JP7554066B2 - Machine learning device, data processing system, and machine learning method - Google Patents

Description

The present disclosure relates to a machine learning device, a data processing system, and a machine learning method that can be applied to a centrifuge system.

Centrifugal separators that use centrifugal force to separate solids and liquids have been used in water treatment facilities for water supply and sewage, industrial wastewater, and human waste. Various types of centrifuges are known, but a type of centrifuge called a decanter is widely used, particularly in facilities that require continuous treatment for long periods of time.

This decanter generally includes a drive motor, a bowl that is rotated by the drive motor and into which the liquid to be treated is fed, a screw conveyor that is coaxially arranged within the bowl, and a differential speed generating device that generates a differential speed between the rotation speed of the bowl and the rotation speed of the screw conveyor. The solid components of the liquid to be treated fed into the bowl are discharged from one end of the bowl, and the liquid components of the liquid to be treated are discharged from the other end of the bowl, thereby achieving solid-liquid separation. Various suitable operation control methods for this decanter have been considered. (See, for example, Patent Document 1 below.)

Patent No. 5442099

According to the method described in the above-mentioned Patent Document 1, the differential speed generated by the differential speed generating device and the centrifugal force of the bowl are controlled based on the conveying torque of the screw conveyor, so that automatic operation of the centrifuge device can be partially realized according to the properties of the liquid to be treated. However, there are various information and parameters related to the control of the centrifuge system including such a centrifuge device other than the conveying torque of the screw conveyor, the differential speed generated by the differential speed generating device, and the centrifugal force of the bowl. In particular, the information (state variables) that affect the control of the control object is diverse, and such information may include information that is difficult to quantify. And, at present, a method for effectively and mechanically taking these state variables into consideration has not been established, and as a result, the reality is that such state variables are reflected in the operation control of the centrifuge system based on the judgment of the engineer (operator) based on the experience and tacit knowledge.

In view of the above, the present disclosure aims to provide a machine learning device, a data processing system, and a machine learning method for achieving suitable operational control of a centrifuge system without relying on the judgment of an operator.

In order to achieve the above object, the machine learning device 20 according to the first aspect of the present disclosure is for a centrifuge system 1 including a bowl 2 that applies centrifugal force to the treated liquid PL1 (not shown) and PL2 (see FIG. 25, etc.) to centrifuge the liquid to be treated into solids M (see FIG. 15, etc.) and a separation liquid SL, as shown in FIG. 3, a screw conveyor 3 that transports the solids M in the bowl 2 toward a discharge outlet 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2, and receives input data including image data of the separation liquid SL discharged from the bowl 2 captured from a predetermined angle of view, and a machine learning device 20 that performs a machine learning process for the input data. The learning data set storage unit 22 stores multiple sets of learning data sets each including an input data set and output data including associated control parameters, the control parameters including at least one of the amount of additive added to the treated liquid PL1, the centrifugal force of the bowl 2, and the differential speed controlled by the differential speed generating device 5; a learning unit 23 that learns a learning model that infers a correlation between the input data and the output data by inputting multiple sets of the learning data set; and a learned model storage unit 24 that stores the learning model learned by the learning unit 23.

This configuration makes it possible to provide a trained model that can infer control parameters for a centrifuge system from image data of the separated liquid.

As shown in FIG. 9, the machine learning device 20A according to the second aspect of the present disclosure is the machine learning device according to the first aspect of the present disclosure, in which the learning unit includes a first learning unit 231 that learns a first learning model that infers features of image data by inputting image data in the learning dataset; and a second learning unit 232 that learns a second learning model that infers the control parameters by inputting features of the image data.

In this configuration, when estimating the control parameters of the centrifuge system from image data of the separated liquid, the number of learning data sets required to obtain the trained models generated in each learning unit can be relatively reduced by dividing the learning unit into two.

As shown in FIG. 11, the machine learning device 20B according to the third aspect of the present disclosure is the machine learning device according to the first aspect of the present disclosure, in which the input data further includes at least one of the slurry concentration of the treated liquid PL1, the moisture content of the dehydrated solids M discharged from the discharge outlet 2a, and the torque value of the screw conveyor 3.

By configuring it in this way, the number of input data increases, making it possible to relatively reduce the number of training datasets required to obtain the trained model generated in the learning unit.

As shown in FIG. 13, the machine learning device 20C according to the fourth aspect of the present disclosure is the machine learning device according to the first aspect of the present disclosure, except that the input data further includes at least one of the slurry concentration of the treated liquid PL1, the moisture content of the dehydrated solids M discharged from the discharge outlet 2a, and the torque value of the screw conveyor 3, and the learning unit includes a first learning unit 231 that learns a first learning model that infers the feature values of the image data by inputting image data in the learning dataset; and a fourth learning unit 234 that learns a fourth learning model that infers the control parameters by inputting at least one of the slurry concentration of the treated liquid PL1, the moisture content of the dehydrated solids M discharged from the discharge outlet 2a, and the torque value of the screw conveyor 3, and the feature values of the image data.

In this configuration, when estimating the control parameters of the centrifuge system from image data of the separated liquid, the learning unit is divided into two, and the number of input data of the learning dataset learned in one of the learning units is increased, making it possible to relatively reduce the number of learning datasets required to obtain the trained model generated in each learning unit.

The machine learning device 120 according to the fifth aspect of the present disclosure is for a centrifuge system 1 including a bowl 2 that applies centrifugal force to the treated liquids PL1 and PL2 to separate them into solids M and a separated liquid SL, a screw conveyor 3 that transports the solids M in the bowl 2 toward a discharge outlet 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2, as shown in FIG. 15, for example. The machine learning device 120 is for a centrifuge system 1 including a bowl 2 that applies centrifugal force to the treated liquids PL1 and PL2 to separate them into solids M and a separated liquid SL, a screw conveyor 3 that transports the solids M in the bowl 2 toward a discharge outlet 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2, and input data including first image data obtained by capturing an image of the separated liquid SL discharged from the bowl 2 from a predetermined angle of view and second image data obtained by capturing an image of the dehydrated solids M discharged from the discharge outlet 2a from a predetermined angle of view, and the input data including first image data obtained by capturing an image of the separated liquid SL discharged from the bowl 2 from a predetermined angle of view and a second image data obtained by capturing an image of the dehydrated solids M discharged from the discharge outlet 2a from a predetermined angle of view. The learning data set storage unit 225 stores multiple sets of learning data sets each including output data including control parameters corresponding to force data, the control parameters including at least one of the amount of additive added to the liquid to be treated PL1, the centrifugal force of the bowl 2, and the differential speed controlled by the differential speed generating device 5; a learning unit 235 that learns a learning model that infers a correlation between the input data and the output data by inputting multiple sets of the learning data sets; and a learned model storage unit 124 that stores the learning model learned by the learning unit.

This configuration makes it possible to provide a trained model that can infer control parameters for the centrifuge system from image data of the separated liquid and image data of the dehydrated solids.

As shown in FIG. 19, the machine learning device 120A according to the sixth aspect of the present disclosure is a machine learning device according to the fifth aspect of the present disclosure, in which the learning units include a first learning unit 231 that learns a first learning model that infers features of first image data by inputting first image data in the learning dataset; a sixth learning unit 236 that learns a sixth learning model that infers features of the second image data by inputting second image data in the learning dataset; and a seventh learning unit 237 that learns a seventh learning model that infers the control parameters by inputting features of the first image data and features of the second image data.

In this configuration, when estimating the control parameters of the centrifuge system from image data of the separated liquid and image data of the dehydrated solids, the number of learning data sets required to obtain the trained model generated in each learning unit can be relatively reduced by dividing the learning unit into three.

A machine learning device 120B according to a seventh aspect of the present disclosure is, for example, as shown in FIG. 21, a machine learning device according to the fifth aspect of the present disclosure, in which the input data further includes at least one of the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3.

By configuring it in this way, the number of input data increases, making it possible to relatively reduce the number of training datasets required to obtain the trained model generated in the learning unit.

As shown in FIG. 23, the machine learning device 120C according to the eighth aspect of the present disclosure is the machine learning device according to the fifth aspect of the present disclosure, in which the input data further includes at least one of the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3, and the learning units include: a first learning unit 231 that learns a first learning model that infers the feature amount of the first image data by inputting the first image data in the learning data set; a sixth learning unit 236 that learns a sixth learning model that infers the feature amount of the second image data by inputting the second image data in the learning data set; and a ninth learning unit 239 that learns a ninth learning model that infers the control parameter by inputting at least one of the slurry concentration of the treated liquid and the torque value of the screw conveyor, the feature amount of the first image data, and the feature amount of the second image data.

In this configuration, when estimating the control parameters of the centrifuge system from the image data of the separated liquid and the image data of the dehydrated solids, the learning unit is divided into three, and the number of input data of the learning dataset learned in one of the learning units is increased, thereby making it possible to relatively reduce the number of learning datasets required to obtain the trained model generated in each learning unit.

As shown in FIG. 25, for example, the machine learning device 320 according to the ninth aspect of the present disclosure is for a centrifuge system 1 including a bowl 2 that applies centrifugal force to the treated liquids PL1 and PL2 to separate them into solids M and a separated liquid SL, a screw conveyor 3 that transports the solids M in the bowl 2 toward a discharge port 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2. The machine learning device 320 receives input data including first image data obtained by capturing an image of the separated liquid SL discharged from the bowl 2 from a predetermined angle of view, and third image data obtained by capturing an image of the treated liquid PL2 from a predetermined angle of view before it is supplied to the bowl 2 and after a predetermined additive has been added thereto. The learning data set storage unit 2210 stores multiple sets of learning data sets including input data and output data including control parameters corresponding to the input data, the control parameters including at least one of the amount of additive added to the treated liquid PL1, the centrifugal force of the bowl 2, and the differential speed controlled by the differential speed generating device 5; a learning unit 2310 that learns a learning model that infers the correlation between the input data and the output data by inputting multiple sets of the learning data sets; and a learned model storage unit 324 that stores the learning model learned by the learning unit 2310.

By configuring it in this way, it is possible to provide a trained model that can infer control parameters of a centrifuge system from image data of the separated liquid and image data of the floc-containing treated liquid.

As shown in FIG. 29, the machine learning device 320A according to the tenth aspect of the present disclosure is the machine learning device according to the ninth aspect of the present disclosure, in which the learning units include a first learning unit 231 that learns a first learning model that infers features of first image data by inputting first image data in the learning dataset; an eleventh learning unit 2311 that learns an eleventh learning model that infers features of the third image data by inputting third image data in the learning dataset; and a twelfth learning unit 2312 that learns a twelfth learning model that infers the control parameters by inputting features of the first image data and features of the third image data.

In this configuration, when estimating the control parameters of the centrifuge system from image data of the separated liquid and image data of the floc-containing treated liquid, the number of learning data sets required to obtain the trained model generated in each learning unit can be relatively reduced by dividing the learning unit into three.

As shown in FIG. 31, for example, a machine learning device 320B according to an eleventh aspect of the present disclosure is a machine learning device according to the ninth aspect of the present disclosure, in which the input data further includes at least one of the slurry concentration of the treated liquid PL1, the moisture content of the dehydrated solids M discharged from the discharge outlet 2a, and the torque value of the screw conveyor 3.

By configuring it in this way, the number of input data increases, making it possible to relatively reduce the number of training datasets required to obtain the trained model generated in the learning unit.

As shown in FIG. 33, for example, a machine learning device 320C according to a twelfth aspect of the present disclosure is a machine learning device according to the ninth aspect of the present disclosure, wherein the input data further includes at least one of the slurry concentration of the treated liquid PL1, the moisture content of the dehydrated solid matter M discharged from the discharge outlet 2a, and the torque value of the screw conveyor 3, and the learning unit includes a first learning unit 231 that inputs first image data in the learning dataset to learn a first learning model that infers features of the first image data; It includes an eleventh learning unit 2311 that learns an eleventh learning model that infers the feature values of the third image data by inputting the third image data in the data set; and a fourteenth learning unit 2314 that learns a fourteenth learning model that infers the control parameters by inputting at least one of the slurry concentration of the treated liquid PL1, the moisture content of the dehydrated solids M discharged from the discharge outlet 2a, and the torque value of the screw conveyor 3, the feature values of the first image data, and the feature values of the third image data.

In this configuration, when estimating the control parameters of the centrifuge system from image data of the separated liquid and image data of the floc-containing treated liquid, the learning unit is divided into three, and the number of input data of the learning dataset learned in one of the learning units is increased, making it possible to relatively reduce the number of learning datasets required to obtain the trained model generated in each learning unit.

As shown in FIG. 35, for example, the machine learning device 520 according to the thirteenth aspect of the present disclosure is for a centrifuge system 1 including a bowl 2 that applies centrifugal force to the treated liquids PL1 and PL2 to separate them into solids M and a separated liquid SL, a screw conveyor 3 that transports the solids M in the bowl 2 toward a discharge port 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2. The machine learning device 520 generates first image data of the separated liquid SL discharged from the bowl 2 taken from a predetermined angle of view, second image data of the dehydrated solids M discharged from the discharge port 2a taken from a predetermined angle of view, and a second image data of the treated liquid PL2 before being supplied to the bowl 2 and after a predetermined additive has been added thereto taken from a predetermined angle of view. The learning data set storage unit 2215 stores multiple sets of learning data sets including input data including third image data captured from a sensor 2216 and output data including control parameters associated with the input data, the control parameters including at least one of the amount of additive added to the treated liquid PL1, the centrifugal force of the bowl 2, and the differential speed controlled by the differential speed generating device 5; a learning unit 2315 that learns a learning model that infers a correlation between the input data and the output data by inputting multiple sets of the learning data sets; and a learned model storage unit 524 that stores the learning model learned by the learning unit 2315.

By configuring it in this way, it is possible to provide a trained model that can infer the control parameters of a centrifuge system from image data of the separated liquid, image data of the dehydrated solids, and image data of the floc-containing liquid to be treated.

As shown in FIG. 38, the machine learning device 520A according to the 14th aspect of the present disclosure includes, in the machine learning device according to the 13th aspect of the present disclosure, a first learning unit 231 that learns a first learning model that infers features of the first image data by inputting the first image data in the learning dataset; a sixth learning unit 236 that learns a sixth learning model that infers features of the second image data by inputting the second image data in the learning dataset; an eleventh learning unit 2311 that learns an eleventh learning model that infers features of the third image data by inputting the third image data in the learning dataset; and a sixteenth learning unit 2316 that learns a sixteenth learning model that infers the control parameter by inputting the features of the first image data, the features of the second image data, and the features of the third image data.

In this configuration, when estimating the control parameters of the centrifuge system from the image data of the separated liquid, the image data of the dewatered solids, and the image data of the floc-containing liquid to be treated, the number of learning data sets required to obtain the trained model generated in each learning unit can be relatively reduced by dividing the learning unit into four.

A machine learning device 520B according to a fifteenth aspect of the present disclosure is, for example, as shown in FIG. 40, a machine learning device according to the thirteenth aspect of the present disclosure, in which the input data further includes at least one of the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3.

By configuring it in this way, the number of input data increases, making it possible to relatively reduce the number of training datasets required to obtain the trained model generated in the learning unit.

As shown in FIG. 42, for example, a machine learning device 520C according to a sixteenth aspect of the present disclosure is a machine learning device according to the thirteenth aspect of the present disclosure, wherein the input data further includes at least one of the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3, and the learning unit includes a first learning unit 231 that learns a first learning model for inferring features of the first image data by inputting first image data in the learning dataset; and a second learning unit 232 that learns a first learning model for inferring features of the second image data by inputting second image data in the learning dataset. The system includes a sixth learning unit 236 that learns a sixth learning model that infers the control parameters; an eleventh learning unit 2311 that learns an eleventh learning model that infers the feature values of the third image data by inputting the third image data in the learning dataset; and an eighteenth learning unit 2318 that learns an eighteenth learning model that infers the control parameters by inputting at least one of the slurry concentration of the treated liquid and the torque value of the screw conveyor, the feature values of the first image data, the feature values of the second image data, and the feature values of the third image data.

In this configuration, when estimating the control parameters of the centrifuge system from the image data of the separated liquid, the image data of the dewatered solids, and the image data of the floc-containing treated liquid, the learning unit is divided into four, and the number of input data of the learning data set learned in one of the learning units is increased, thereby making it possible to relatively reduce the number of learning data sets required to obtain the trained model generated in each learning unit.

A machine learning device according to a seventeenth aspect of the present disclosure is a machine learning device according to any one of the first to sixteenth aspects of the present disclosure, in which the control parameters further include a supply amount of the treated liquid.

By configuring it in this way, the number of control parameters output by the trained model increases, making it possible to achieve more fine-grained parameter adjustment.

The machine learning device according to an eighteenth aspect of the present disclosure is the machine learning device according to any one of the first to seventeenth aspects of the present disclosure, wherein the control parameters further include a dam set diameter of the bowl.

By configuring it in this way, the number of control parameters output by the trained model increases, making it possible to achieve more fine-grained parameter adjustment.

As shown in FIG. 7 and FIG. 10, the data processing system 80, 80A according to the 19th aspect of the present disclosure is used in a centrifuge system 1 including a bowl 2 that applies centrifugal force to the treated liquid PL1, PL2 to separate the solid matter M and the separated liquid SL, a screw conveyor 3 that transports the solid matter M in the bowl 2 toward a discharge port 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2, and includes a first image data acquisition unit 81 for acquiring first image data obtained by capturing an image of the separated liquid SL discharged from the bowl 2 from a predetermined angle of view; and an inference unit 87, 871, 872 that infers control parameters of the centrifuge system 1 by inputting the data acquired by the first image data acquisition unit 81 into a trained model generated by the machine learning device 20, 20A according to the first or second aspect.

With this configuration, control parameters that can achieve suitable operational control of the centrifuge system can be inferred based on the first image data that captures the separated liquid, making it possible to automatically control the operation of the centrifuge system without relying on the judgment of the operator.

As shown in, for example, FIG. 12 and FIG. 14, the data processing systems 80B and 80C according to the twentieth aspect of the present disclosure are used in a centrifugation system 1 including a bowl 2 that applies centrifugal force to the treated liquid PL1, PL2 to separate the treated liquid PL1, PL2 into solid matter M and separated liquid SL, a screw conveyor 3 that transports the solid matter M in the bowl 2 toward a discharge port 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2, and obtain first image data obtained by capturing an image of the separated liquid SL discharged from the bowl 2 at a predetermined angle of view. a first image data acquisition unit 81 for acquiring the slurry concentration of the treated liquid PL1, the moisture content of the dehydrated solids M discharged from the discharge port 2a, and the torque value of the screw conveyor 3; and inference units 873, 871, and 874 for inferring control parameters of the centrifuge system 1 by inputting the data acquired by the first image data acquisition unit 81 and the additional variable acquisition unit 89 into the trained model generated by the machine learning device 20B or 20C according to the third or fourth aspect.

With this configuration, control parameters that can realize suitable operational control of the centrifuge system can be inferred based on at least one of the slurry concentration of the treated liquid, the moisture content of the dewatered solids, and the torque value of the screw conveyor, and the first image data capturing the separated liquid, making it possible to automatically control the operation of the centrifuge system without relying on the judgment of the operator.

As shown in Figs. 17 and 20, for example, the data processing system 180, 180A according to the 21st aspect of the present disclosure is used in a centrifuge system 1 including a bowl 2 that applies centrifugal force to the treated liquid PL1, PL2 to separate the solid matter M and the separated liquid SL, a screw conveyor 3 that transports the solid matter M in the bowl 2 toward a discharge port 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2. The data processing system 180, 180A according to the 21st aspect of the present disclosure is used in a centrifuge system 1 including a bowl 2 that applies centrifugal force to the treated liquid PL1, PL2 to separate the solid matter M and the separated liquid SL, a screw conveyor 3 that conveys the solid matter M in the bowl 2 toward a discharge port 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2. a first image data acquisition unit 81 for acquiring image data of the dehydrated solid matter M discharged from the discharge port 2a; a second image data acquisition unit 181 for acquiring second image data obtained by capturing an image of the dehydrated solid matter M discharged from the discharge port 2a from a predetermined angle of view; and inference units 875, 871, 876, and 877 for inferring control parameters of the centrifuge system 1 by inputting data acquired by the first image data acquisition unit 81 and the second image data acquisition unit 181 into a trained model generated by the machine learning device 120 or 120A according to the fifth or sixth aspect.

With this configuration, control parameters that can achieve suitable operational control of the centrifuge system can be inferred based on the first and second image data that capture the separated liquid and dehydrated solids, making it possible to automatically control the operation of the centrifuge system without relying on the judgment of the operator.

22 and 24, the data processing systems 180B and 180C according to the 22nd aspect of the present disclosure are used in a centrifugation system 1 including a bowl 2 that applies centrifugal force to the treated liquid PL1 and PL2 to separate the solid matter M and the separated liquid SL, a screw conveyor 3 that transports the solid matter M in the bowl 2 toward a discharge port 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2, and include a first image data acquisition unit 81 for acquiring first image data obtained by capturing an image of the separated liquid SL discharged from the bowl 2 at a predetermined angle of view; It includes a second image data acquisition unit 181 for acquiring second image data obtained by capturing an image of the discharged dehydrated solids M from a predetermined angle of view; an additional variable acquisition unit 89 for acquiring at least one of the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3; and inference units 878, 871, 876, and 879 for inferring control parameters of the centrifuge system 1 by inputting the data acquired by the first image data acquisition unit 81, the second image data acquisition unit 181, and the additional variable acquisition unit 89 into the trained model generated by the machine learning device 120B or 120C according to the seventh or eighth aspect.

With this configuration, it is possible to infer control parameters that can realize suitable operational control of the centrifuge system based on at least one of the slurry concentration of the liquid being treated, the torque value of the screw conveyor, and the first and second image data of the separated liquid and dewatered solids, making it possible to automatically control the operation of the centrifuge system without relying on the judgment of the operator.

27 and 30, the data processing system 380, 380A according to the 23rd aspect of the present disclosure is used in a centrifuge system 1 including a bowl 2 that applies centrifugal force to the treated liquid PL1, PL2 to separate the solid matter M and the separated liquid SL, a screw conveyor 3 that transports the solid matter M in the bowl 2 toward a discharge port 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2, and is used to acquire first image data obtained by capturing an image of the separated liquid SL discharged from the bowl 2 from a predetermined angle of view. a first image data acquisition unit 81 for acquiring image data of the treated liquid PL2 from a predetermined angle of view before being supplied to the bowl 2 and after a predetermined additive has been added; and inference units 8710, 871, 8711, 8712 for inferring control parameters of the centrifuge system by inputting data acquired by the first image data acquisition unit 81 and the third image data acquisition unit 381 into a trained model generated by the machine learning device 320, 320A according to the ninth or tenth aspect.

With this configuration, control parameters that can realize suitable operational control of the centrifuge system can be inferred based on the first and third image data of the separated liquid and the floc-containing treated liquid, making it possible to automatically control the operation of the centrifuge system without relying on the judgment of the operator.

As shown in Figures 32 and 34, for example, the data processing systems 380B and 380C according to the 24th aspect of the present disclosure are used in a centrifugation system 1 including a bowl 2 that applies centrifugal force to the treated liquid PL1, PL2 to separate the solid matter M and the separated liquid SL, a screw conveyor 3 that transports the solid matter M in the bowl 2 toward a discharge port 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2, and include a first image data acquisition unit 81 for acquiring first image data obtained by capturing an image of the separated liquid SL discharged from the bowl 2 at a predetermined angle of view; It includes a third image data acquisition unit 381 for acquiring third image data obtained by capturing an image of the processing liquid PL2 from a predetermined angle of view; an additional variable acquisition unit 89 for acquiring at least one of the slurry concentration of the processing liquid PL1, the moisture content of the dehydrated solids M discharged from the discharge port 2a, and the torque value of the screw conveyor 3; and inference units 8713, 871, 8711, and 8714 for inferring control parameters of the centrifuge system 1 by inputting data acquired by the first image data acquisition unit 81, the third image data acquisition unit 381, and the additional variable acquisition unit 89 into a trained model generated by the machine learning device 320B or 320C according to the 11th or 12th aspect.

With this configuration, control parameters that can realize suitable operational control of the centrifuge system can be inferred based on at least one of the slurry concentration of the treated liquid, the moisture content of the dewatered solids, and the torque value of the screw conveyor, and the first and third image data of the separated liquid and the floc-containing treated liquid, making it possible to automatically control the operation of the centrifuge system without relying on the judgment of the operator.

As shown in FIG. 36 and FIG. 39, the data processing system 580, 580A according to the 25th aspect of the present disclosure is used in a centrifugation system 1 including a bowl 2 that applies centrifugal force to the treated liquid PL1, PL2 to separate the solid matter M and the separated liquid SL, a screw conveyor 3 that transports the solid matter M in the bowl 2 toward a discharge port 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2, and includes a first image data acquisition unit 81 for acquiring first image data obtained by capturing an image of the separated liquid SL discharged from the bowl 2 from a predetermined angle of view; a second image data acquisition unit 181 for acquiring second image data captured from the first image data acquisition unit 181; a third image data acquisition unit 381 for acquiring third image data captured from a predetermined angle of view of the treated liquid PL2 before it is supplied to the bowl 2 and after a predetermined additive has been added; and inference units 8715, 871, 876, 8711, 8716 for inferring control parameters of the centrifuge system by inputting data acquired by the first image data acquisition unit 81, the second image data acquisition unit 181, and the third image data acquisition unit 381 into a trained model generated by the machine learning device 520, 520A according to the thirteenth or fourteenth aspect.

With this configuration, control parameters that can realize suitable operational control of the centrifuge system can be inferred based on the first to third image data of the separated liquid, dewatered solids, and floc-containing treated liquid, making it possible to automatically control the operation of the centrifuge system without relying on the judgment of the operator.

As shown in FIG. 41 or 43, for example, the data processing systems 580B and 580C according to the 26th aspect of the present disclosure are used in a centrifugation system 1 including a bowl 2 that applies centrifugal force to the treated liquid PL1, PL2 to separate the solid matter M and the separated liquid SL, a screw conveyor 3 that transports the solid matter M in the bowl 2 toward a discharge port 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2, and include a first image data acquisition unit 81 for acquiring first image data obtained by imaging the separated liquid SL discharged from the bowl 2 from a predetermined angle of view; a second image data acquisition unit 181 for acquiring second image data obtained by imaging the dehydrated solid matter M discharged from the discharge port 2a from a predetermined angle of view; A third image data acquisition unit 381 for acquiring third image data obtained by capturing an image of the treated liquid PL2 from a predetermined angle of view before it is supplied to the bowl 2 and after a predetermined additive has been added; an additional variable acquisition unit 89 for acquiring at least one of the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3; and inference units 8717, 871, 876, 8711, 8718 for inferring control parameters of the centrifuge system 1 by inputting data acquired by the first image data acquisition unit 81, the second image data acquisition unit 181, the third image data acquisition unit 381, and the additional variable acquisition unit 89 into a trained model generated by the machine learning device 520B, 520C according to the 15th or 16th aspect.

With this configuration, it is possible to infer control parameters that can realize suitable operational control of the centrifuge system based on at least one of the slurry concentration of the treated liquid, the torque value of the screw conveyor, and the first to third image data of the separated liquid, the dewatered solids, and the treated liquid containing flocs, making it possible to automatically control the operation of the centrifuge system without relying on the judgment of the operator.

The machine learning method according to the 27th aspect of the present disclosure is implemented by a computer, for example, as shown in FIG. 3 and FIG. 6, and is for a centrifuge system 1 including a bowl 2 that applies centrifugal force to the treated liquids PL1 and PL2 to separate them into solids M and a separated liquid SL, a screw conveyor 3 that transports the solids M in the bowl 2 toward a discharge port 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2, and includes image data obtained by capturing the separated liquid SL discharged from the bowl 2 at a predetermined angle of view. The method includes: step S11 of storing a plurality of sets of learning data sets including input data including control parameters and output data including control parameters associated with the input data, the control parameters including at least one of the amount of additive added to the treated liquid PL1, the centrifugal force of the bowl 2, and the differential speed controlled by the differential speed generating device 5; step S15 of learning a learning model that infers a correlation between the input data and the output data by inputting a plurality of sets of the learning data sets; and step S17 of storing the learned learning model.

This configuration makes it possible to provide a trained model that can infer control parameters for a centrifuge system from image data of the separated liquid.

The machine learning method according to the 28th aspect of the present disclosure is implemented by a computer, for example as shown in FIG. 6 and FIG. 15, and is for a centrifuge system 1 including a bowl 2 that applies centrifugal force to the treated liquids PL1 and PL2 to separate them into solids M and separated liquid SL, a screw conveyor 3 that transports the solids M in the bowl 2 toward a discharge outlet 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2, and the ... centrifuge system 1 including a bowl 2 that applies centrifugal force to the treated liquids PL1 and PL2 to separate them into solids M and separated liquid SL, and the machine learning method according to the 28th aspect of the present disclosure is implemented by a centrifuge system 1 including a bowl 2 that applies centrifugal force to the treated liquids PL1 and PL2 to separate them into solids M and separated liquid SL, and the machine learning method according to the 28th aspect of the present disclosure is implemented by a centrifuge system 1 including a bowl 2 that applies centrifugal force to the treated liquids PL1 and PL2 to separate them into solids M and separated liquid SL, and the machine learning method according to the centrifuge system 1 includes a bowl The method includes: step S11 for storing multiple sets of learning data sets including input data including second image data obtained by capturing an image of a solid object M from a predetermined angle of view, and output data including control parameters associated with the input data, the control parameters including at least one of the amount of additive added to the liquid to be treated PL1, the centrifugal force of the bowl 2, and the differential speed controlled by the differential speed generating device 5; step S15 for learning a learning model that infers a correlation between the input data and the output data by inputting multiple sets of the learning data sets; and step S17 for storing the learned learning model.

This configuration makes it possible to provide a trained model that can infer control parameters for the centrifuge system from image data of the separated liquid and image data of the dehydrated solids.

The machine learning method according to the 29th aspect of the present disclosure is implemented by a computer, for example, as shown in FIG. 6 and FIG. 25, and is for a centrifuge system 1 including a bowl 2 that applies centrifugal force to the treated liquid PL1, PL2 to separate the solid matter M and the separated liquid SL, a screw conveyor 3 that transports the solid matter M in the bowl 2 toward a discharge port 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2, and includes a first image data obtained by capturing an image of the separated liquid SL discharged from the bowl 2 from a predetermined angle of view, and a second image data obtained by capturing an image of the separated liquid SL before it is supplied to the bowl 2 and to which a predetermined additive has been added. The method includes: step S11 for storing a plurality of sets of learning data sets including input data including third image data obtained by capturing an image of the treated liquid PL2 from a predetermined angle of view after the addition of the additive; and output data including control parameters associated with the input data, the control parameters including at least one of the amount of additive added to the treated liquid PL1, the centrifugal force of the bowl 2, and the differential speed controlled by the differential speed generating device 5; step S15 for learning a learning model that infers a correlation between the input data and the output data by inputting a plurality of sets of the learning data sets; and step S17 for storing the learned learning model.

By configuring it in this way, it is possible to provide a trained model that can infer control parameters of a centrifuge system from image data of the separated liquid and image data of the floc-containing treated liquid.

The machine learning method according to the 30th aspect of the present disclosure is implemented by a computer, for example as shown in FIG. 6 and FIG. 35, and is for a centrifuge system 1 including a bowl 2 that applies centrifugal force to the treated liquids PL1 and PL2 to separate them into solids M and separated liquid SL, a screw conveyor 3 that transports the solids M in the bowl 2 toward a discharge outlet 2a, a drive motor 4 that rotates the bowl 2, and a differential speed generating device 5 that rotates the screw conveyor 3 at a differential speed relative to the bowl 2. ... The method includes: step S11 for storing a plurality of sets of learning data sets including input data including third image data obtained by capturing an image of the treated liquid PL2 from a predetermined angle of view before it is supplied and after a predetermined additive has been added, and output data including control parameters associated with the input data, the control parameters including at least one of the amount of additive added to the treated liquid PL1, the centrifugal force of the bowl 2, and the differential speed controlled by the differential speed generating device 5; step S15 for learning a learning model that infers a correlation between the input data and the output data by inputting a plurality of sets of the learning data sets; and step S17 for storing the learned learning model.

By configuring it in this way, it is possible to provide a trained model that can infer the control parameters of a centrifuge system from image data of the separated liquid, image data of the dehydrated solids, and image data of the floc-containing liquid to be treated.

According to the present disclosure, it becomes possible to obtain control parameters that can mechanically achieve suitable operational control of the centrifuge system without relying on the judgment of the operator.

FIG. 2 is a schematic diagram showing a general structure of a decanter body according to the first embodiment of the present disclosure. FIG. 1 is a schematic diagram showing a centrifugal separation system including a piping structure of a decanter according to a first embodiment of the present disclosure. 1 is a schematic block diagram of a machine learning device according to a first embodiment of the present disclosure. 3 shows an example of image data obtained by the separation liquid monitoring system according to the first embodiment of the present disclosure. FIG. 2 is a diagram showing an example of a neural network model for supervised learning implemented in the machine learning device according to the first embodiment of the present disclosure. 1 is a flowchart illustrating an example of a machine learning method according to a first embodiment of the present disclosure. 1 is a schematic block diagram illustrating a data processing system according to a first embodiment of the present disclosure. 1 is a flowchart illustrating an example of parameter adjustment by the data processing system according to the first embodiment of the present disclosure. FIG. 11 is a schematic block diagram of a machine learning device according to a second embodiment of the present disclosure. FIG. 11 is a schematic block diagram illustrating a data processing system according to a second embodiment of the present disclosure. FIG. 13 is a schematic block diagram of a machine learning device according to a third embodiment of the present disclosure. FIG. 13 is a schematic block diagram illustrating a data processing system according to a third embodiment of the present disclosure. FIG. 13 is a schematic block diagram of a machine learning device according to a fourth embodiment of the present disclosure. FIG. 13 is a schematic block diagram illustrating a data processing system according to a fourth embodiment of the present disclosure. FIG. 13 is a schematic block diagram of a machine learning device according to a fifth embodiment of the present disclosure. 13 shows an example of image data obtained by a dehydrated solids monitoring system according to a fifth embodiment of the present disclosure. FIG. 13 is a schematic block diagram illustrating a data processing system according to a fifth embodiment of the present disclosure. 13 is a flowchart illustrating an example of parameter adjustment by a data processing system according to a fifth embodiment of the present disclosure. FIG. 13 is a schematic block diagram of a machine learning device according to a sixth embodiment of the present disclosure. FIG. 13 is a schematic block diagram showing a data processing system according to a sixth embodiment of the present disclosure. FIG. 13 is a schematic block diagram of a machine learning device according to a seventh embodiment of the present disclosure. FIG. 13 is a schematic block diagram illustrating a data processing system according to a seventh embodiment of the present disclosure. FIG. 23 is a schematic block diagram of a machine learning device according to an eighth embodiment of the present disclosure. FIG. 13 is a schematic block diagram showing a data processing system according to an eighth embodiment of the present disclosure. FIG. 13 is a schematic block diagram of a machine learning device according to a ninth embodiment of the present disclosure. 13 shows an example of image data obtained by a treatment liquid monitoring system according to a ninth embodiment of the present disclosure. FIG. 13 is a schematic block diagram showing a data processing system according to a ninth embodiment of the present disclosure. 23 is a flowchart showing an example of parameter adjustment by a data processing system according to a ninth embodiment of the present disclosure. FIG. 23 is a schematic block diagram of a machine learning device according to a tenth embodiment of the present disclosure. FIG. 23 is a schematic block diagram showing a data processing system according to a tenth embodiment of the present disclosure. FIG. 23 is a schematic block diagram of a machine learning device according to an eleventh embodiment of the present disclosure. FIG. 23 is a schematic block diagram showing a data processing system according to an eleventh embodiment of the present disclosure. FIG. 23 is a schematic block diagram of a machine learning device according to a twelfth embodiment of the present disclosure. FIG. 23 is a schematic block diagram showing a data processing system according to a twelfth embodiment of the present disclosure. FIG. 23 is a schematic block diagram of a machine learning device according to a thirteenth embodiment of the present disclosure. A schematic block diagram showing a data processing system according to a thirteenth embodiment of the present disclosure. 23 is a flowchart showing an example of parameter adjustment by a data processing system according to a thirteenth embodiment of the present disclosure. FIG. 23 is a schematic block diagram of a machine learning device according to a fourteenth embodiment of the present disclosure. A schematic block diagram showing a data processing system according to a fourteenth embodiment of the present disclosure. FIG. 23 is a schematic block diagram of a machine learning device according to a fifteenth embodiment of the present disclosure. FIG. 23 is a schematic block diagram showing a data processing system according to a fifteenth embodiment of the present disclosure. FIG. 23 is a schematic block diagram of a machine learning device according to a sixteenth embodiment of the present disclosure. A schematic block diagram showing a data processing system according to a sixteenth embodiment of the present disclosure.

Each embodiment for carrying out the present disclosure will be described below with reference to the drawings. Note that the scope necessary for the explanation to achieve the objectives of the present disclosure will be shown in a schematic manner below, and the scope necessary for the explanation of the relevant parts of the present disclosure will be mainly explained, and the parts for which explanation is omitted will be explained as being based on publicly known technology.

First Embodiment
Before describing the machine learning device, data processing system, and machine learning method according to the first embodiment of the present disclosure, a centrifuge system to which the machine learning device, data processing system, and machine learning method are applied will be briefly described. The centrifuge system according to the present embodiment uses a centrifuge device that includes a horizontal decanter 1. Note that the specific aspects of the centrifuge system according to the present disclosure are not limited to those shown below, and can also be applied to centrifuge systems that include a centrifuge device such as a vertical or straight-body decanter.

Figure 1 is a schematic diagram showing the general structure of a decanter body according to a first embodiment of the present disclosure. Also, Figure 2 is a schematic diagram showing a centrifuge system including a piping structure of a decanter according to a first embodiment of the present disclosure. The horizontal decanter 1 shown in this embodiment mainly includes a bowl 2, a screw conveyor 3, a drive motor 4, a differential speed generating device 5, and a casing 6, as shown in Figures 1 and 2.

The bowl 2 is made of a cylindrical member with one end processed into a weight shape, and is supported so as to be rotatable around a horizontal axis. One or more solid discharge outlets 2a are provided at the weight-shaped end of the bowl 2, and a dam 2c with one or more separated liquid discharge outlets 2b is attached to the other end. Solid components contained in the liquid to be treated that is poured into the bowl 2 are mainly discharged from the solid discharge outlet 2a, and liquid components contained in the liquid to be treated that is also poured into the bowl 2 are mainly discharged from the separated liquid discharge outlet 2b. A pulley 4a that transmits power from a drive motor 4, which will be described later, is attached to one end of the bowl 2 (the end processed into a weight shape in FIG. 1). The solid components discharged from the solid discharge outlet 2a are discharged in a state in which most of the liquid components in the liquid to be treated have been dehydrated and removed, and contain a relatively small amount of liquid (separated liquid) components, so hereinafter, these solid components are referred to as liquid-containing solids (or dehydrated solids) M.

The screw conveyor 3 is arranged coaxially within the bowl 2 and is composed of a member with a spiral screw blade 3a formed around it. The screw blade 3a is a member for transporting and/or squeezing the solid matter in the bowl 2. The body 3b of this screw conveyor 3 is provided with a space 3c for receiving the liquid to be treated therein, and a discharge port 3d for pouring the liquid to be treated stored in this space into the bowl 2. In addition, a differential speed generating device 5, which will be described later, is connected to one end of this screw conveyor 3.

The drive motor 4 is a motor for applying a rotational force to the bowl 2, and is connected to a pulley 4a attached to one end of the bowl 2 via a belt 4b. A relatively large motor is preferably used for the drive motor 4 in order to rotate the bowl 2 in the range of, for example, 2000 to 5000 rpm, and the rotation speed is preferably variable as appropriate by inverter control. The differential speed generating device 5 is a device for generating a differential speed between the rotation speed of the bowl 2 and the rotation speed of the screw conveyor 3, and is a device that can rotate the screw conveyor 3 slightly slower (for example, about 1 to 3 rpm) than the bowl 2. The differential speed generating device 5 includes a gear box 5a connected to the screw conveyor 3 and a differential motor (also called a "back drive motor") 5b that applies a braking force to the screw conveyor 3. The specific operating principle of the differential speed generating device 5 is conventionally known, so its description will be omitted.

The casing 6 is provided to cover the bowl 2 and the screw conveyor 3, and guides the liquid-containing solids (dehydrated solids) M, which are solids containing a small amount of liquid and discharged from the solid discharge outlet 2a of the bowl 2, to the solids chute 6a provided below, and similarly guides the separated liquid SL (see FIG. 3) discharged from the separated liquid discharge outlet 2b of the bowl 2 to the separated liquid chute 6b provided below. In addition, the bowl 2 and the screw conveyor 3 covered by the casing 6 are supported together by the frame 6c.

The solid matter chute 6a of the casing 6 is connected to a solid matter discharge conduit 7, and the solid matter discharged from the solid matter chute 6a is transported, for example, via this solid matter discharge conduit 7 to a main transport path 7a to a drying device, incineration equipment, or the like (not shown). In addition, the separated liquid chute 6b of the casing 6 is connected to a separated liquid discharge conduit 8, and the separated liquid discharged from the separated liquid chute 6b is transported, for example, via this separated liquid discharge conduit 8 to a water purification equipment, or the like (not shown).

The decanter 1 further includes a liquid supply pipe 9 for feeding the liquid to be treated into the bowl 2. One end of this liquid supply pipe 9 is connected to the space 3c in the body 3b of the screw conveyor 3, and the other end is connected to a source 10 of the liquid to be treated and a source 11 of additives, and serves as a conduit for feeding the liquid to be treated, etc., that flows in from the other end, into the bowl 2.

The liquid supply pipe 9 and the treated liquid supply source 10 are fluidly connected by a treated liquid pipe 10b equipped with a pump 10a for supplying the treated liquid. By controlling this pump 10a, the amount of treated liquid supplied into the bowl can be controlled. Here, the treated liquid supplied from the treated liquid supply source 10 is often a slurry (suspension) containing sludge as a solid matter.

The liquid supply pipe 9 and the additive supply source 11 are fluidly connected by an additive pipe 11c equipped with a pump 11a for supplying an additive, such as a chemical, to be added to the liquid being treated, and multiple valves 11b for controlling the supply position and timing of the chemical. Here, the chemical supplied from the additive supply source 11 is a flocculant, particularly a polymer flocculant or an inorganic flocculant, which is administered to the slurry to change the sludge contained in the slurry into a floc-like state. In the following, the liquid being treated before the chemical is added will be referred to as the "liquid being treated PL1" and the liquid being treated after the chemical has been added will be referred to as the "floc-containing liquid being treated PL2" to distinguish between the two.

As shown in FIG. 2, the additive pipe 11c according to this embodiment is branched into two pipes at its midpoint, one of which is connected to the middle of the treated liquid pipe 10b, and the other is connected to the liquid supply pipe 9. When the chemical is supplied via the pipe connected to the middle of the treated liquid pipe 10b, the treated liquid PL1 reacts with the chemical in the treated liquid pipe 10b (so-called "line chemical supply"); when the chemical is supplied via the pipe connected to the liquid supply pipe 9, the treated liquid PL1 reacts with the chemical mainly in the space 3c and the bowl 2 (so-called "in-machine chemical supply"). By adopting such a piping structure, the position and timing at which the chemical is added to the treated liquid PL1 can be adjusted, and the effect of generating sludge flocs (also called "floc aggregates") by the chemical, which is mainly influenced by the state of the treated liquid PL1, can be optimally adjusted. Note that the connection position of the pipes shown in FIG. 2 is only one example, and the piping structure can be appropriately changed in consideration of the state of the treated liquid, etc.

In addition, it is preferable that the various pipes of the decanter 1 according to this embodiment are provided with a monitoring system for monitoring objects passing through the pipes. In detail, this monitoring system includes a separated liquid monitoring system 40 provided in the separated liquid discharge conduit 8, a dehydrated solids monitoring system 50 provided in the solids discharge conduit 7, and a treated liquid monitoring system 60 provided in the treated liquid pipe 10b. In addition to these monitoring systems, a screw conveyor torque monitoring system 70 for monitoring the torque value of the screw conveyor 3 may also be included. Several examples of the specific configuration of each monitoring system will be described in accordance with each embodiment described later. In the first embodiment, no monitoring systems other than the separated liquid monitoring system 40 are used, so monitoring systems other than the separated liquid monitoring system 40 will be described in the second embodiment and subsequent embodiments.

In the solid-liquid separation by the decanter 1 having the above-mentioned configuration, first, the floc-containing liquid PL2 to be treated is fed through the liquid feed pipe 9 into the bowl 2 rotating at a predetermined rotation speed, and the solids in the floc-containing liquid PL2 (which has become flocculated due to the effect of the chemicals) settle on the inner wall surface of the bowl 2 due to the action of the centrifugal force generated by the drive motor 4. The settled solids are conveyed while being squeezed toward the one end of the bowl 2, which is machined into a weight shape, by the screw blades 3a of the screw conveyor 3, which rotates with the bowl 2 at a rotation speed slightly lower than the rotation speed of the bowl 2 due to the action of the differential speed generating device 5, and are discharged from the solid discharge port 2a to the solid chute 6a together with a predetermined amount of fluid components. In addition, the residue in the bowl 2 after the solids have been settled and removed as described above remains in the bowl 2 as the separated liquid SL for a certain period of time, and then overflows from the separated liquid discharge port 2b and is discharged to the separated liquid chute 6b.

<Machine learning device>
The above-mentioned solid-liquid separation process is performed in a suitable manner by adjusting various control parameters of the decanter 1. The control parameters here mainly consist of the centrifugal force of the bowl 2, the amount of chemicals supplied from the additive supply source 11, and the differential speed between the bowl 2 and the screw conveyor 3. Hereinafter, a machine learning device 20 according to a first embodiment of the present disclosure that learns an inference model (trained model) capable of estimating optimal control parameters in the decanter 1 will be described.

FIG. 3 is a schematic block diagram of a machine learning device 20 according to a first embodiment of the present disclosure. As shown in FIG. 3, the machine learning device 20 according to this embodiment includes a learning dataset acquisition unit 21, a learning dataset storage unit 22, a learning unit 23, and a trained model storage unit 24.

The learning data set acquisition unit 21 is an interface unit that acquires a plurality of data constituting a learning (training) data set, for example, via a wired or wireless communication line. The specific contents of the plurality of data acquired here can be appropriately changed according to the trained model to be generated. In this embodiment, the plurality of data are acquired as image data of the separation liquid SL and control parameters associated with this image data. The control parameters include the supply amount of the chemical added to the treated liquid PL1, the centrifugal force of the bowl 2, and the differential speed controlled by the differential speed generating device 5. In this embodiment, the above-mentioned three data are acquired as control parameters, but the control parameters constituting the learning data set may include at least one of the above-mentioned three data. In addition, the control parameters are not limited to the above-mentioned three, and may also include, for example, the supply amount of the treated liquid PL1 from the treated liquid supply source 10 and the dam set diameter of the dam 2c of the bowl 2. Therefore, the control parameters constituting the output data may be four or more. In general, if the number of control parameters is increased, more detailed parameter adjustments can be achieved when applied to the operation control of the decanter 1 described below, but as the number of control parameters increases, the number of training data sets required to obtain a trained model capable of inference with sufficient accuracy tends to increase. Therefore, it is preferable to determine the number of control parameters as output data taking into consideration the number of training data sets that can be prepared, etc.

An example of a method for acquiring a plurality of data acquired by the learning data set acquisition unit 21 will be described. As shown in FIG. 3, the learning data set acquisition unit 21 is connected to a computer PC1 and acquires desired data from the computer PC1. The computer PC1 may be, for example, at least a part of the control unit 30 that controls various operations of the decanter 1 described above, or may be a computer communicably connected to the control unit 30. This allows the computer PC1 to acquire various control parameters of the decanter 1. In addition, as shown in FIG. 2, the computer PC1 is directly connected to a separated liquid monitoring system 40 provided in the separated liquid discharge conduit 8 or indirectly connected via the control unit 30, and can acquire image data of the separated liquid SL from the separated liquid monitoring system 40. The control unit 30 of the decanter 1 referred to here is a device for controlling the entire decanter 1, and may be composed of a well-known computer equipped with a processor, memory, etc., connected to various sensors and driving means provided in the decanter 1.

Here, as shown in FIG. 3, the separated liquid monitoring system 40 may include a sight glass 41 connected to an arbitrary position of the separated liquid discharge conduit 8, and a separated liquid imaging camera 43 that is installed at a predetermined angle of view in a window 42 provided in the sight glass 41 and can capture image data of the separated liquid SL flowing in the sight glass 41. Of these, a well-known camera capable of capturing two-dimensional images may be used as the separated liquid imaging camera 43. In addition, in this separated liquid monitoring system 40, a light source (not shown) that illuminates the separated liquid SL in the sight glass 41 and a polarizing filter (not shown) that can be attached to the separated liquid imaging camera 43 may be appropriately used so that the state of the separated liquid SL is easily reflected in the image data captured by the separated liquid imaging camera 43. Note that the configuration of the separated liquid monitoring system 40 is not limited to this, and various configurations may be used as long as they are capable of acquiring image data of the separated liquid SL. As a specific example, a configuration such as the separated liquid monitoring system 40A shown in FIG. 9 described later may be used.

Figure 4 shows an example of image data obtained by the separated liquid imaging camera 43 of the separated liquid monitoring system 40 according to the first embodiment of the present disclosure. In the technical field of water treatment equipment, the amount of solid matter contained in the separated liquid SL discharged after the solid-liquid separation process by the centrifugal separation system, i.e., a low solid matter content concentration, can be an indicator for inferring that the solid-liquid separation process is being performed well. In addition, in the image data obtained by the separated liquid imaging camera 43 (also referred to as the "first image data"), if the separated liquid SL contains a large amount of solid matter, it tends to be in a state in which a large amount of small particulate solid matter is mixed in. Therefore, as shown in Figure 4, it can be inferred that image data in which the solid matter content concentration in the separated liquid SL is relatively high (for example, as shown in Figure 4(b)) tends to have a larger amount of color change between pixels than image data in which the solid matter content concentration is relatively low (for example, as shown in Figure 4(a)).

The learning dataset storage unit 22 is a database for storing a plurality of data constituting the learning dataset acquired by the learning dataset acquisition unit 21, associating related input data and output data (also called "teacher data") into one learning dataset. In the present embodiment, the learning dataset stored may be one in which the input data is image data of the separated liquid SL discharged from the bowl 2 captured at a predetermined angle of view, and the output data is a control parameter associated with the image data as the input data. The specific configuration of the database constituting the learning dataset storage unit 22 may be adjusted as appropriate. For example, in FIG. 3, for convenience of explanation, the learning dataset storage unit 22 and the trained model storage unit 24 described later are shown as separate storage means, but they may also be configured by a single storage medium (database).

As described above, the learning dataset stored in the learning dataset storage unit 22 is composed of one image data and a control parameter corresponding to this one image data. On the other hand, in order to generate one trained model in the learning unit 23, it is necessary to perform learning using many learning datasets (for example, several thousand to tens of thousands of sets). Therefore, in order to prepare a large number of learning datasets in a relatively short time, it is preferable to perform data augmentation in the learning dataset storage unit 22. As a specific method of this data augmentation, for example, a method can be adopted in which a number of square partial image data smaller than the original image data is randomly extracted from one rectangular original image data (for example, 100 pieces), and each of the extracted a number of partial image data is associated with a control parameter corresponding to the original image data, thereby acquiring a number of learning datasets.

The learning unit 23 learns a learning model that infers the correlation between the input data and the output data in the learning dataset by inputting a plurality of sets of the learning datasets stored in the learning dataset storage unit 22. In this embodiment, as will be described in detail later, supervised learning using a neural network is adopted as a specific method of machine learning. However, the specific method of machine learning is not limited to this, and other learning methods can be adopted as long as they can learn the correlation between input and output from the learning dataset. For example, ensemble learning (random forest, boosting, stacking, etc.) can also be used.

The trained model storage unit 24 is a database for storing trained models generated by the learning unit 23. The trained models stored in the trained model storage unit 24 are applied to the real system upon request via communication lines including the Internet or storage media. Specific aspects of applying the trained models to the real system (data processing system 80) will be described in detail later.

When preparing a learning dataset for the decanter 1 according to this embodiment, it is necessary to identify optimal control parameters that will serve as teacher data. As a method for identifying these optimal control parameters, for example, a method in which the operator EN manually identifies them by taking into consideration the image data acquired by the separated liquid monitoring system 40 and the actual control parameters at that time can be adopted. In this case, it is preferable to assign a skilled engineer or multiple engineers to the operator EN who identifies the optimal control parameters that will serve as teacher data. The optimal control parameters identified in this way are organized and stored in the learning dataset storage unit 22 in the machine learning device 20 as teacher data for the learning dataset in a state associated with the above-mentioned image data.

Here, the optimal control parameters to be identified may be at least one of the centrifugal force of the bowl 2, the amount of drug supply from the additive supply source 11, and the differential speed between the bowl 2 and the screw conveyor 3, as described above. Since these control parameters are considered to be the control parameters that have the greatest effect on the solid-liquid separation process by the decanter 1, it is preferable that the control parameters to be identified here are all of the above-mentioned three control parameters as exemplified in this embodiment. Of course, only one or only two of these three control parameters may be adopted as output data, and it is also possible to further add parameters other than these three control parameters (e.g., drug supply position, etc.) to the output data. Of the above three control parameters, the centrifugal force of the bowl 2 can be adjusted mainly by controlling the rotation speed of the bowl 2 by the drive motor 4, but the control of changing this rotation speed has a characteristic of being less responsive than the other two controls (control of drug supply amount and differential speed). Therefore, if only two of the three control parameters are used as output data, it is advisable to use the amount of drug supplied from the additive supply source 11 and the differential speed between the bowl 2 and the screw conveyor 3.

Next, a learning method in the learning unit 23 using the multiple learning data sets obtained as described above will be briefly described. FIG. 5 is a diagram showing an example of a neural network model for supervised learning implemented in the machine learning device according to the first embodiment of the present disclosure. The neural network in the neural network model shown in FIG. 5 is composed of l neurons (x1 to x1) in the input layer, m neurons (y11 to y1m) in the first hidden layer, n neurons (y21 to y2n) in the second hidden layer, and o neurons (z1 to zo) in the output layer. The first and second hidden layers are also called hidden layers, and the neural network may have multiple hidden layers in addition to the first and second hidden layers, or may have only the first hidden layer as a hidden layer.

In addition, between the input layer and the first hidden layer, between the first hidden layer and the second hidden layer, and between the second hidden layer and the output layer, there are nodes that connect the neurons between the layers, and each node is assigned a weight wi (i is a natural number).

The neural network in the neural network model according to this embodiment uses a learning dataset to learn the correlation between the input data and output data of the learning dataset. Specifically, image data of the separation liquid SL as state variables constituting the input data is associated with the neurons in the input layer, and the values of the neurons in the output layer are calculated using a general neural network output value calculation method, that is, the value of the output side neuron is calculated as the sum of a sequence of multiplication values of the value of the input side neuron connected to the neuron and the weight wi associated with the node connecting the output side neuron and the input side neuron, for all neurons other than the neurons in the input layer. Note that when the state variables are input to the neurons in the input layer, the format in which the information acquired as the state variables is input can be appropriately set in consideration of the accuracy of the trained model to be generated, etc.

Then, the calculated values of o neurons z1 to zo in the output layer (i.e., the control parameters in this embodiment) are compared with the teacher data t1 to to, which are also composed of control parameters and form part of the learning dataset, to find the error, and the weights wi associated with each node are adjusted (backpropagation) so that the found error becomes small.

Then, when a certain condition is met, such as the above-mentioned series of steps being repeated a certain number of times or the error being smaller than an allowable value, the learning is terminated and the neural network model (all weights wi associated with each of the nodes) is stored as a learned model in the learned model storage unit 24.

<Machine learning method>
In relation to the above, the present disclosure provides a machine learning method. FIG. 6 is a flowchart showing an example of a machine learning method according to a first embodiment of the present disclosure. In the machine learning method described below, the description will be based on the above-mentioned machine learning device 20, but the premise of the configuration is not limited to the above-mentioned machine learning device 20. In addition, the machine learning method is realized by using a computer, but various computers can be applied, such as a computer constituting the control unit 30, a server device arranged on a network, or a computer PC1 shown in FIG. 3. In addition, as for the specific configuration of this computer, for example, one including at least a calculation device consisting of a CPU, a GPU, etc., a storage device consisting of a volatile or non-volatile memory such as a RAM or a ROM, a communication device for communicating with a network or other devices, and a bus connecting these devices can be adopted. Furthermore, the machine learning method may be provided in the form of a program having one or more instructions for causing a computer to execute a predetermined operation, or in the form of a non-transitory computer-readable medium storing the program.

In the supervised learning as a machine learning method according to the present embodiment, a desired number of training datasets are first prepared as a preliminary step before starting machine learning, and the prepared multiple training datasets are stored in the training dataset storage unit 22 (step S11). The number of training datasets to be prepared here should be set taking into consideration the inference accuracy required for the trained model that is ultimately obtained. In addition, an example of a method for preparing training datasets has already been given above, so a description thereof will be omitted here.

After step S11 is completed, a pre-learning neural network model is prepared (S12) to start machine learning in the learning unit 23. The pre-learning neural network model prepared here may have the structure shown in FIG. 5, for example, and the weights of each node may be set to an initial value. Then, for example, one learning dataset is randomly selected from the multiple learning datasets stored in the learning dataset storage unit 22 (step S13), and the input data in the one learning dataset is input to the input layer (see FIG. 5) of the prepared pre-learning neural network model (step S14). Note that various methods can be used to input the input data in the learning dataset to the input layer of the pre-learning neural network model. As a specific example, a method can be used in which the luminance value and/or color value (e.g., RGB value) of each pixel of the image data is input to the neurons of each input layer. In addition, a predetermined pre-processing such as a dimensional reduction process or noise removal process for adjusting the number of data may be performed before inputting the image data as input data to the input layer.

Here, the control parameters of the output layer (see FIG. 5) generated as a result of step S14 are generated by the neural network model before learning, and therefore in most cases are values different from the desired result. Therefore, next, machine learning is performed using the output data as teacher data in one learning dataset acquired in step S13, i.e., the control parameters, and the control parameters of the output layer generated in step S13 (step S15). The machine learning performed here may be, for example, a process (backpropagation) in which the control parameters constituting the teacher data and the control parameters constituting the output layer are compared, an error between the two is detected, and a weight associated with each node in the neural network model before learning is adjusted so that an output layer in which the error is reduced is obtained. In addition, the number and format of the control parameters output to the output layer of the neural network model before learning are the same as those of the teacher data in the learning dataset as the learning target. Therefore, for example, if the control parameters as teacher data in the learning dataset are composed of data corresponding to two control parameters, the control parameters output to the output layer by the neural network model are also data corresponding to the two control parameters.

Once machine learning is performed in step S15, it is determined whether or not it is necessary to continue machine learning, for example, based on the remaining number of unlearned training datasets stored in the training dataset storage unit 22 (step S16). If machine learning is to be continued (No in step S16), the process returns to step S13, and if machine learning is to be terminated (Yes in step S16), the process proceeds to step S17. If the machine learning is to be continued, steps S13 to S15 are performed multiple times on the neural network model being trained using unlearned training datasets. The accuracy of the trained model that is finally generated generally tends to be higher in proportion to the number of times.

When the machine learning is to be terminated (Yes in step S16), the weights associated with each node are adjusted through a series of steps to generate a neural network, which is stored as a trained model in the trained model storage unit 24 (step S17), and the series of learning processes is terminated. The trained model stored here can be applied and used in various data processing systems, and specific examples of such data processing systems will be described later.

In the above-mentioned learning process and machine learning method of the machine learning device, in order to generate one trained model, multiple machine learning processes are repeatedly performed on one (pre-training) neural network model to improve its inference accuracy and obtain a trained model that is suitable for application to a data processing system. However, the present invention is not limited to this acquisition method. For example, a plurality of trained models that have undergone a predetermined number of machine learning processes are stored as candidates in the trained model storage unit 24, and input data from a common data set for validity judgment is input to the group of multiple trained models to generate an output layer (neuron values), and the accuracy of the control parameters identified in this output layer is compared to select one of the best trained models to be applied to the data processing system. The validity judgment data set may be composed of a data set similar to the training data set used for learning, and may be one that has not been used for learning.

Optionally, instead of the pre-trained neural network model prepared in step S12 above in which the node weights are set to initial values, a trained model previously generated for, for example, any image recognition can be used. In this case, the desired trained model is generated by so-called fine tuning or transfer learning. Therefore, a highly accurate trained model can be generated with a smaller number of training datasets than in the normal machine learning process described above.

As described above, by applying the machine learning device 20 and machine learning method according to this embodiment, a trained model capable of deriving optimal control parameters from image data acquired by the separation liquid monitoring system 40 can be obtained.

<Data Processing System>
Next, an application example of the trained model generated by the above-mentioned machine learning device and machine learning method will be described with reference to Fig. 7. Fig. 7 is a schematic block diagram showing a data processing system 80 according to a first embodiment of the present disclosure. As an example of the data processing system 80 according to the present embodiment, a system applied to the control unit 30 of the above-mentioned horizontal decanter 1 is shown.

As shown in FIG. 7, the data processing system 80 mainly includes a first image data acquisition unit 81, a parameter adjustment unit 82, a calculation unit 83, a database (DB) 84, a user interface 85, and an internal bus 86 for interconnecting these. Note that FIG. 7 shows only the components related to the data processing system 80, and omits the description of other components provided in the control unit 30 that are not directly related to the data processing system 80 according to this embodiment.

The first image data acquisition unit 81 is for acquiring image data, more specifically, image data of the separation liquid SL. Specifically, it is connected to the above-mentioned separation liquid monitoring system 40 and acquires image data captured by the separation liquid imaging camera 43 of the separation liquid monitoring system 40.

The parameter adjustment unit 82 adjusts various control units, which are adjusted by the control unit 30 to realize optimal operation control, based on the inference results of the inference unit 87 described later. This parameter adjustment unit 82 is connected to a drive motor control unit 31 that is connected to the drive motor 4 and controls its rotation speed, a pump control unit 32 that is provided in the additive piping 11c and controls the pump 11a that varies the amount of drug supply, and a differential motor control unit 33 that is connected to a differential motor 5b and controls its rotation speed to control the rotation speed of the screw conveyor 3 (relative to the bowl 2). Note that the parameter adjustment unit 82 in this embodiment is connected to the three control units described above, but the connection destinations can be appropriately adjusted according to the number and type of control parameters output by the inference unit 87.

The arithmetic unit 83 constitutes a processor for realizing various processes in the data processing system 80, and includes at least an inference unit 87. The arithmetic unit 83 is connected to a learned model storage unit 88 that stores a learned model used in the inference unit 87. The inference unit 87 infers a control parameter to be adjusted by the parameter adjustment unit 82 from image data as a state variable acquired by the first image data acquisition unit 81 by taking into consideration one learned model stored in the learned model storage unit 88. It is preferable that a plurality of learned models are stored in the learned model storage unit 88 according to the purpose and various conditions (for example, environmental conditions such as season, weather, temperature and humidity, or the type of liquid to be treated, etc.). Furthermore, the task of selecting an appropriate learned model from a plurality of learned models may be automatically selected using various sensors, etc., or may be manually selected by an operator EN, etc.

The database 84 is made of a well-known recording medium and is used to temporarily or continuously store various data handled by the data processing system 80, such as image data acquired by the first image data acquisition unit 81 and the results of calculations by the calculation unit 83. The user interface 85 is, for example, a GUI (graphical user interface), and is used to display the status of the decanter 1 and receive input operations from the operator EN, etc.

Figure 8 is a flowchart showing an example of parameter adjustment by the data processing system 80 according to the first embodiment of the present disclosure. When the decanter 1 to which the data processing system 80 is applied starts to operate, as shown in Figure 8, the data processing system 80 first determines whether it is time to adjust the parameters (step S21). Here, the timing for adjusting the control parameters can be specified automatically or manually. In addition, the adjustment may be performed periodically at a predetermined time interval while the decanter 1 is operating, or the adjustment may be performed only at a specific timing, such as when the operation of the decanter 1 is started, when the type of the treated liquid to be treated by the decanter 1 is changed, when a predetermined time has elapsed since operation, or when the treated liquid monitoring system 60 or the dehydrated solids monitoring system 50 is capable of detecting a change in the concentration of the treated liquid or the moisture content of the dehydrated solids and the change exceeds a predetermined threshold.

When it is detected that it is time to adjust the parameters (Yes in step S21), the separated liquid image capturing camera 43 in the separated liquid monitoring system 40 is operated to capture an image of the separated liquid SL flowing in the separated liquid discharge conduit 8 (step S22). The captured image data is acquired by the first image data acquisition unit 81 and sent to the data processing system 80 (step S23).

Then, the image data acquired by the first image data acquisition unit 81 is sent to the inference unit 87 via the internal bus 86, and the inference unit 87 infers control parameters by inputting this image data to the input layer of one trained model in the trained model storage unit 88 that has been specified in advance by the inference unit 87 (step S24). Then, the parameter adjustment unit 82 adjusts each control unit using the values of the control parameters inferred here (step S25). After that, the process returns to step S21 and enters a standby state.

As described above, according to the data processing system 80 of this embodiment, the preferred control parameters of the decanter 1 can be estimated based on image data of the separated liquid, which is information that is relatively easy to obtain, and therefore this data processing system 80 can be applied to the decanter 1 relatively easily. Furthermore, adjustment of the control parameters can be easily performed without relying on the judgment of the operator EN, etc.

Optionally, in the above embodiment, the trained model taken into consideration in the inference unit 87 is generated through so-called batch learning as shown in FIG. 6, but the data processing system 80 of the present invention is not limited to this. Specifically, for example, the trained model generated through the above-mentioned batch learning may be further subjected to online learning to further improve accuracy. As a specific method, first, the image data as state variables acquired in the first image data acquisition unit 81 and the control parameters inferred by the inference unit 87, particularly the control parameters when the solid-liquid separation processing result is improved as a result of reflecting the adjustment in the parameter adjustment unit 82, are temporarily stored in the database 84 as a set of online learning data sets. Then, machine learning (specifically fine tuning) similar to that shown in FIG. 6 can be performed using the online learning data set stored in the database 84 at any time.

Although the above-mentioned data processing system 80 is applied in the control unit 30 of the horizontal decanter 1, it can be applied to, for example, a computer connected to the control unit 30 of the decanter 1 in a communicable manner, or a server device connected to the control unit 30 of the decanter 1 via a network or the like. However, the centrifuge systems used for sludge treatment, which are mainly assumed in this embodiment, usually have significantly different processing contents (type of liquid to be treated, processing amount per unit time, etc.) and surrounding environments (climate, etc.) for each system. Therefore, generating a general-purpose trained model that can be applied to all of these centrifuge systems requires a very large number of training data sets for the generation, and the contents of the data must be obtained under various conditions without bias, which tends to increase the generation cost. Therefore, it is cost-effective and preferable to apply and operate the data processing system 80 and the above-mentioned trained model to each centrifuge system.

Second Embodiment
In the machine learning device, machine learning method, and data processing system according to the first embodiment, a case where a control parameter is inferred from image data of the separation liquid SL using one trained model has been described. However, although a correlation is recognized between the image data of the separation liquid SL and the control parameter, the degree of correlation between the two is not extremely high. In general, in a machine learning process of a neural network model, the number of training data sets required to obtain a trained model capable of inference with sufficient accuracy tends to be small in proportion to the degree of correlation between the input data and output data of the training data set. Therefore, in the first embodiment, in order to learn the correlation between the image data of the separation liquid SL and the control parameter, it is necessary to prepare a relatively large number of training data sets. Therefore, as one aspect for generating a trained model capable of inference with sufficient accuracy with a smaller number of training data sets, a case where two trained models are used to infer a control parameter from image data of the separation liquid SL will be described below as a second embodiment of the present disclosure. Among the components of the machine learning device 20A and data processing system 80A according to the second embodiment described below, those common to the components of the machine learning device 20 and data processing system 80 according to the first embodiment are denoted by the same reference numerals and their description is omitted. Similarly to the first embodiment, the machine learning device, machine learning method, and data processing system according to the second embodiment described below are described by taking as an example a case where they are applied to the centrifuge system shown in Figures 1 and 2. Furthermore, all of the modified examples described in the embodiments described above or below are applicable to this embodiment to the extent that no contradiction occurs.

FIG. 9 is a schematic block diagram of a machine learning device 20A according to a second embodiment of the present disclosure. As shown in FIG. 9, the machine learning device 20A according to this embodiment is similar to the machine learning device 20 according to the first embodiment, except that the components of the machine learning device 20A according to this embodiment include a first learning dataset storage unit 221 and a second learning dataset storage unit 222 as learning dataset storage units, and a first learning unit 231 and a second learning unit 232 as learning units. In relation to this, the multiple data acquired in the learning dataset acquisition unit 21 are also different from those in the first embodiment.

The multiple data acquired by the learning data set acquisition unit 21 according to this embodiment are acquired from the computer PC1, and in addition to the image data of the separated liquid SL and the control parameters, the multiple data further include feature quantities of the image data of the separated liquid SL (hereinafter also referred to as "feature quantities of the separated liquid SL"). As the feature quantities of the separated liquid SL, various pieces of information identified based on the captured image of the separated liquid SL can be adopted. Specifically, for example, the solid matter concentration of the separated liquid SL, color value, average particle size of particles contained in the separated liquid SL, particle distribution, degree of foaming, etc. can be included in one or more of them. The feature quantities can be identified by actually sampling the separated liquid SL and measuring and analyzing it using various measuring instruments, etc., or by the operator EN visually judging it.

In order to measure the above-mentioned characteristic quantities, it is preferable to adopt a configuration capable of sampling and extracting the separated liquid SL in the separated liquid monitoring system 40A according to the present embodiment. Specifically, as shown in FIG. 9, the separated liquid monitoring system 40A can adopt a system including a branch pipe 44 provided at an appropriate position in the separated liquid discharge conduit 8, a valve 45 provided in the branch pipe 44, a transparent container 46 disposed at the end of the branch pipe 44, and a separated liquid imaging camera 43 installed on the side of the transparent container 46. The arrangement of the separated liquid imaging camera 43 is not limited to this, and it may be installed in a position where the container 46 can be imaged from above, for example. In the separated liquid monitoring system 40A having such a configuration, the separated liquid SL flowing in the separated liquid discharge conduit 8 is sampled and extracted in the transparent container 46 by opening the valve 45 at any timing. Then, the separated liquid SL in this container 46 is imaged by the separated liquid imaging camera 43 to generate image data of the separated liquid SL, and the corresponding desired control parameter is specified by the operator EN. Furthermore, by measuring and analyzing the separated liquid SL sampled in the container 46, the characteristic quantities of the separated liquid SL can be identified. Of course, the separated liquid monitoring system 40 described in the first embodiment can also be used in this embodiment as an alternative.

The multiple data acquired in the learning dataset acquisition unit 21 are stored separately as two learning datasets in the first learning dataset storage unit 221 and the second learning dataset storage unit 222, taking into consideration their respective correspondences. The first learning dataset stored in the first learning dataset storage unit 221 may include image data of the separated liquid SL, obtained by capturing an image of the separated liquid discharged from the bowl 2 from a predetermined angle of view, as the first input data, and may include a feature amount of the separated liquid SL associated with the first input data as the first output data. The second learning dataset stored in the second learning dataset storage unit 222 may include a feature amount of the separated liquid SL as the second input data, and may include a control parameter associated with the second input data as the second output data. When splitting and generating two learning data sets from a plurality of data in this manner, for example, the image data of the separation liquid SL, the features of the separation liquid SL, and the control parameters associated with the same separation liquid SL may be split into a set of the image data of the separation liquid SL and the features of the separation liquid SL, and a set of the features of the separation liquid SL and the control parameters, and each set may be treated as a first and a second learning data set. Note that the first and second learning data sets obtained after splitting in this case do not need to be stored in a format that maintains their association, since each set is referenced in a different learning unit, which will be described later.

The first and second learning data sets stored in the first learning data set storage unit 221 and the second learning data set storage unit 222, respectively, are referred to only by different learning units. The first learning unit 231 learns a learning model that infers the correlation between the first input data and the first output data by inputting a plurality of sets of the first learning data set. In other words, the first learning unit 231 learns a first learning model that infers the feature amount of the separation liquid SL in this image data, i.e., the feature amount of the image data, by inputting the image data of the separation liquid SL in the first learning data set. The second learning unit 232 learns a learning model that infers the correlation between the second input data and the second output data by inputting a plurality of sets of the second learning data set. In other words, the second learning unit 232 learns a learning model that infers the correlation between the second input data and the second output data by inputting the feature amount of the image data in the second learning data set.

The specific machine learning methods in the first and second learning units 231, 232 use different learning datasets, but the process is similar to the supervised learning process shown in FIG. 6. The first and second trained models obtained through a series of machine learning processes are then stored in the trained model storage unit 24.

Figure 10 is a schematic block diagram showing a data processing system 80A according to a second embodiment of the present disclosure. As shown in Figure 10, the data processing system 80A according to this embodiment has the same components as the data processing system 80 according to the first embodiment described above, except that the data processing system 80A according to this embodiment has a first inference unit 871 and a second inference unit 872 as inference units in the arithmetic unit 83.

The first inference unit 871 performs inference using the first trained model generated by the above-mentioned machine learning device 20A and stored in the trained model storage unit 88. Therefore, when image data of the separation liquid SL acquired by the first image data acquisition unit 81 is input, the first inference unit 871 outputs the feature amount of this image data to the output layer. Moreover, the second inference unit 872 performs inference using the second trained model generated by the above-mentioned machine learning device 20A and stored in the trained model storage unit 88. Therefore, when the feature amount of the image data inferred by the first inference unit 871 is input, the second inference unit 872 outputs a control parameter to the output layer.

When adjusting the control parameters in a decanter 1 to which a data processing system 80A including the above-mentioned first and second inference units 871, 872 is applied, the same process as that shown in FIG. 8 described above can be performed. However, when inferring the control parameters in step S24 of the process shown in FIG. 8, first, image data of the separated liquid SL is input to the first inference unit 871, and the feature quantities of the output image data are input to the second inference unit 872.

As described above, according to the machine learning device, machine learning method, and data processing system of this embodiment, the feature quantities of image data that have a higher degree of correlation with the image data of the separation liquid SL than the control parameters are inferred, and then the control parameters are inferred using the feature quantities of image data that have a higher degree of correlation with the control parameters than the image data of the separation liquid SL. This is expected to relatively reduce the number of training datasets required to generate each trained model that enables inference with sufficient accuracy.

Third Embodiment
In the machine learning device and the machine learning method according to the first and second embodiments, only the image data of the separation liquid SL is used as the input data of the learning data set. However, when only the image data of the separation liquid SL is used as the input data, the number of learning data sets required to generate a trained model capable of inference with sufficient accuracy tends to increase. Therefore, as one aspect for generating a trained model capable of inference with sufficient accuracy with a smaller number of learning data sets, a case in which the number of input data of the learning data set shown in the above-mentioned first embodiment is increased will be described below as a third embodiment of the present disclosure. Note that, among the components of the machine learning device 20B and the data processing system 80B according to the third embodiment described below, those common to the components of the machine learning device 20 and the data processing system 80 according to the first embodiment are given the same reference numerals and their description will be omitted. In addition, the machine learning device, the machine learning method, and the data processing system according to the third embodiment described below are described by taking as an example a case in which they are applied to the centrifuge system shown in FIG. 1 and FIG. 2, similarly to those according to the first and second embodiments. Furthermore, all of the modifications described in the embodiments described above or below are applicable to this embodiment as well to the extent that no contradiction occurs.

FIG. 11 is a schematic block diagram of a machine learning device 20B according to a third embodiment of the present disclosure. As shown in FIG. 11, the machine learning device 20B according to this embodiment is similar to the machine learning device 20 according to the first embodiment described above, except that the machine learning device 20B according to this embodiment includes a third learning data set storage unit 223 as a learning data set storage unit and a third learning unit 233 as a learning unit. In addition, the multiple data acquired in the learning data set acquisition unit 21 are different from those according to the first embodiment. As shown in FIG. 11, the learning data set acquisition unit 21 according to this embodiment is connected to a computer PC1 and acquires desired data from this computer PC1. Like the computer PC1 according to the first embodiment, this computer PC1 is connected to a control unit 30 and a separated liquid monitoring system 40, but in addition to these, it is also connected to a dewatered solids monitoring system 50, a treated liquid monitoring system 60, and a screw conveyor torque monitoring system 70.

As shown in Figs. 2 and 11, the dehydrated solids monitoring system 50 is preferably installed in the solids discharge conduit 7, for example, and is capable of monitoring the moisture content of the dehydrated solids (also called "dehydrated cake") M passing through the solids discharge conduit 7. The dehydrated solids monitoring system 50 measures or estimates the moisture content of the dehydrated solids M passing through the solids discharge conduit 7. Although various methods can be used to measure or estimate the moisture content of the dehydrated solids M, in this embodiment, a method is used to directly measure the moisture content of the dehydrated solids M passing through the solids discharge conduit 7 using a well-known moisture content meter that uses microwaves or the like. In Fig. 2, the solids discharge conduit 7 is shown as an example of the location where the dehydrated solids monitoring system 50 is installed, but instead, for example, the main conveying path 7a may be used.

As shown in Figs. 2 and 11, the treated liquid monitoring system 60 is arranged in the treated liquid pipe 10b, for example, and monitors the slurry concentration of the treated liquid PL1 supplied from the treated liquid supply source 10. A specific method for monitoring the slurry concentration may be to directly detect the concentration value of the treated liquid using a known laser, optical, or ultrasonic concentration sensor. In this embodiment, the treated liquid monitoring system 60 is provided at the position of the treated liquid pipe 10b, particularly at the junction with the additive pipe 11c. Alternatively, for example, the treated liquid monitoring system 60 may be provided at a position before the treated liquid pipe 10b joins with the additive pipe 11c, in the liquid supply pipe 9, or in the space 3c. In addition, it is also possible to measure the slurry concentration not in the treated liquid PL1 before the chemical is added, but in the floc-containing treated liquid PL2 after the chemical is added to the treated liquid PL1.

As shown in Figures 2 and 11, the screw conveyor torque monitoring system 70 is attached to, for example, the rotating shaft of the screw conveyor 3, and monitors the torque value of the screw conveyor 3 by detecting the reaction force acting on the screw conveyor 3.

Various data acquired by these four monitoring systems 40, 50, 60 and 70, specifically, image data of the separated liquid SL, the moisture content of the dehydrated solids M, the slurry concentration of the treated liquid PL1 (or the floc-containing treated liquid PL2) and the torque value of the screw conveyor 3 are sent directly or via the control unit 30 to the computer PC1, which then sends them to the learning dataset acquisition unit 21 together with the corresponding desired control parameters. The multiple data constituting the learning dataset acquired by the learning dataset acquisition unit 21 are stored in the third learning dataset storage unit 223 in the form of a third learning dataset in which the image data of the separated liquid SL, the moisture content of the dehydrated solids M, the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3 are third input data, and the control parameters associated with these data are third output data. Then, in the third learning unit 233, machine learning is performed using the third learning data set in a manner similar to the machine learning method of the first embodiment described above, and the obtained third learned model is stored in the learned model storage unit 24. In this embodiment, the moisture content of the dehydrated solid matter M, the slurry concentration of the treated liquid PL1, and the torque value of the screw conveyor 3 are all data of one value among the third input data, whereas the image data of the separated liquid SL are the brightness and/or color values of all pixels constituting the image data, and therefore the total number of data is significantly different. Therefore, if these input data are directly associated with the input layer, the obtained inference result will be extremely influenced by the image data of the separated liquid SL. Therefore, in this embodiment, it is particularly preferable to perform preprocessing of the data to adjust the number of data before associating the third input data with the input layer, so that the degree of influence of input data other than the image data of the separated liquid SL is not too small. In addition, the preprocessing can be performed in the same manner during inference by storing the preprocessing in the learned model storage unit 24 together with the neural network model obtained as part of the learned model.

Figure 12 is a schematic block diagram showing a data processing system 80B according to a third embodiment of the present disclosure. As shown in Figure 12, the data processing system 80B according to this embodiment has the same components as the data processing system 80 according to the first embodiment described above, except that it has a third inference unit 873 as an inference unit and an additional variable acquisition unit 89.

As shown in FIG. 12, the additional variable acquisition unit 89 is connected to the dewatered solids monitoring system 50, the treated liquid monitoring system 60, and the screw conveyor torque monitoring system 70, and acquires the moisture content of the dewatered solids M, the slurry concentration of the treated liquid PL1, and the torque value of the screw conveyor 3 from these monitoring systems. It is preferable that the moisture content of the dewatered solids M, the slurry concentration of the treated liquid PL1, and the torque value of the screw conveyor 3 acquired by this additional variable acquisition unit 89 are acquired at the same time as the image data acquired by the first image data acquisition unit 81 from the separated liquid monitoring system 40 is captured.

When adjusting the control parameters in the decanter 1 to which the data processing system 80B equipped with the additional variable acquisition unit 89 described above is applied, the same process as that shown in FIG. 8 described above can be performed. However, in the process shown in FIG. 8, the additional variable acquisition unit 89 acquires the moisture content of the dehydrated solids M, the slurry concentration of the treated liquid PL1, and the torque value of the screw conveyor 3 at the same time as step S23 or at a predetermined timing before or after it. These three values are then associated with the input layer of the trained model in the third inference unit 873, together with the image data of the separated liquid SL.

In the above-mentioned third embodiment, the computer PC1 is connected to three systems: the control unit 30 and the separated liquid monitoring system 40, as well as the dehydrated solids monitoring system 50, the treated liquid monitoring system 60, and the screw conveyor torque monitoring system 70, and the three values acquired from these systems are used as input data for the learning data set together with image data, but the present invention is not limited to this. In more detail, it is sufficient that at least one of the values acquired by the dehydrated solids monitoring system 50, the treated liquid monitoring system 60, and the screw conveyor torque monitoring system 70 is used as input data for the learning data set. The same applies to the number of pieces of information acquired by the additional variable acquisition unit 89.

As described above, according to the machine learning device, machine learning method, and data processing system of this embodiment, in addition to image data of the separated liquid SL, the moisture content of the dehydrated solids M, the slurry concentration of the treated liquid PL1, and the torque value of the screw conveyor 3 are used as input data, making it easier to identify correlations between input data and output data in the learning stage of the neural network model. This is expected to reduce the number of learning data sets required to generate each trained model capable of inference with sufficient accuracy.

<Fourth embodiment>
As described above, the second and third embodiments are both examples showing a mode in which the number of learning data sets required to generate a trained model capable of inference with sufficient accuracy can be reduced. The two embodiments can be combined. Therefore, a case in which the technical ideas shown in the second and third embodiments described above are combined will be described below as a fourth embodiment of the present disclosure. Note that, among the components of the machine learning device 20C and the data processing system 80C according to the fourth embodiment described below, those components common to the machine learning devices 20, 20A, and 20B and the data processing systems 80, 80A, and 80B according to the first to third embodiments are given the same reference numerals and their description will be omitted. In addition, the machine learning device, the machine learning method, and the data processing system according to the third embodiment described below are described by taking as an example a case in which they are applied to the centrifuge system shown in FIG. 1 and FIG. 2, as in the first to third embodiments. Furthermore, all the modified examples described in the above or later-described embodiments can be applied to this embodiment to the extent that no contradiction occurs.

FIG. 13 is a schematic block diagram of a machine learning device 20C according to a fourth embodiment of the present disclosure. As shown in FIG. 13, the machine learning device 20C according to this embodiment is similar to the machine learning device 20 according to the first embodiment, except that the components of the machine learning device 20C according to this embodiment include a first learning dataset storage unit 221 and a fourth learning dataset storage unit 224 as learning dataset storage units, and a first learning unit 231 and a fourth learning unit 234 as learning units. In relation to this, the multiple data acquired in the learning dataset acquisition unit 21 also differ from those of the first embodiment.

The learning data set acquisition unit 21 according to this embodiment is connected to a computer PC1 as shown in FIG. 13, and acquires desired data from the computer PC1. This computer PC1 is connected to the control unit 30 and the separated liquid monitoring system 40A, as in the computer PC1 according to the second embodiment, but in addition to these, it is also connected to at least one of the dewatered solid matter monitoring system 50, the treated liquid monitoring system 60, and the screw conveyor torque monitoring system 70. The specific configurations of the control unit 30, the separated liquid monitoring system 40A, the dewatered solid matter monitoring system 50, the treated liquid monitoring system 60, and the screw conveyor torque monitoring system 70 connected to the computer PC1 are the same as those exemplified in the other embodiments already described above. In addition to the image data of the separated liquid SL and the control parameters, the multiple data acquired by the learning data set acquisition unit 21 in this embodiment further include at least one of the moisture content of the dewatered solid matter M, the slurry concentration of the treated liquid PL1 (or the floc-containing treated liquid PL2), and the torque value of the screw conveyor 3, and the characteristic amount of the separated liquid SL. In this embodiment, as in the third embodiment, the computer PC1 is connected to all of the dehydrated solids monitoring system 50, the treated liquid monitoring system 60, and the screw conveyor torque monitoring system 70, and the multiple data acquired by the learning data set acquisition unit 21 includes, for example, the moisture content of the dehydrated solids M, the slurry concentration of the treated liquid PL1, and the torque value of the screw conveyor 3.

The multiple data acquired in the learning data set acquisition unit 21 are stored separately as two learning data sets in the first learning data set storage unit 221 and the fourth learning data set storage unit 224 while taking into consideration their respective correspondences. The first learning data set stored in the first learning data set storage unit 221 may include image data of the separated liquid SL obtained by capturing the separated liquid discharged from the bowl 2 from a predetermined angle of view as the first input data, and may include the feature amount of the separated liquid SL associated with the first input data as the first output data. The fourth learning data set stored in the fourth learning data set storage unit 224 may include the feature amount of the separated liquid SL, the moisture content of the dehydrated solids M, the slurry concentration of the treated liquid PL1, and the torque value of the screw conveyor 3 as the fourth input data, and may include the control parameters associated with the fourth input data as the fourth output data. In this way, when splitting and generating two learning data sets from multiple data, for example, the image data of the separated liquid SL, the moisture content of the dehydrated solids M, the slurry concentration of the liquid to be treated PL1, the torque value of the screw conveyor 3, the features and control parameters of the separated liquid SL associated with the same separated liquid SL can be split into a set of the image data of the separated liquid SL and the features of the separated liquid SL, and a set of the moisture content of the dehydrated solids M, the slurry concentration of the liquid to be treated PL1, the torque value of the screw conveyor 3, and the features and control parameters of the separated liquid SL, and each can be used as a first and fourth learning data set.

The first and fourth learning data sets stored in the first learning data set storage unit 221 and the fourth learning data set storage unit 224, respectively, are referred to only by different learning units. The first learning unit 231 learns a learning model that infers the correlation between the first input data and the first output data by inputting a plurality of sets of the first learning data set. In other words, the first learning model learns a first learning model that infers the feature amount of the image data by inputting the image data of the separated liquid SL in the first learning data set. And the fourth learning unit 234 learns a learning model that infers the correlation between the fourth input data and the fourth output data by inputting a plurality of sets of the fourth learning data set. In other words, the fourth learning model learns a fourth learning model that infers the control parameters by inputting the feature amount of the image data in the fourth learning data set, the moisture content of the dehydrated solids M, the slurry concentration of the liquid PL1 to be treated, and the torque value of the screw conveyor 3.

The specific machine learning methods in the first and fourth learning units 231, 234 use different learning datasets, but the process is similar to the supervised learning process shown in FIG. 6. The first and fourth trained models obtained through a series of machine learning processes are then stored in the trained model storage unit 24.

Figure 14 is a schematic block diagram showing a data processing system 80C according to a fourth embodiment of the present disclosure. As shown in Figure 14, the data processing system 80C according to this embodiment has the same components as the data processing system 80 according to the first embodiment described above, except that it has a first inference unit 871 and a fourth inference unit 874 as inference units in the arithmetic unit 83, and also has an additional variable acquisition unit 89.

As shown in FIG. 14, the additional variable acquisition unit 89 is connected to the dewatered solids monitoring system 50, the treated liquid monitoring system 60, and the screw conveyor torque monitoring system 70, and acquires the moisture content of the dewatered solids M, the slurry concentration of the treated liquid PL1, and the torque value of the screw conveyor 3 from these monitoring systems. The first inference unit 871 performs inference using the first trained model generated by the above-mentioned machine learning device 20C and stored in the trained model storage unit 88. Therefore, when the image data of the separated liquid SL acquired by the first image data acquisition unit 81 is input, the first inference unit 871 outputs the feature amount of this image data to the output layer. Furthermore, the fourth inference unit 874 performs inference using the fourth trained model generated by the above-mentioned machine learning device 20C and stored in the trained model storage unit 88. Therefore, when the feature quantities of the image data inferred by the first inference unit 871, the moisture content of the dehydrated solids M, the slurry concentration of the liquid to be treated PL1, and the torque value of the screw conveyor 3 are input, the fourth inference unit 874 outputs control parameters to the output layer.

When adjusting the control parameters in the decanter 1 to which the data processing system 80C including the above-mentioned additional variable acquisition unit 89 and the first and fourth inference units 871 and 874 is applied, the same process as that shown in FIG. 8 may be performed. However, the process shown in FIG. 8 further includes a step in which the additional variable acquisition unit 89 acquires the moisture content of the dehydrated solids M, the slurry concentration of the treated liquid PL1, and the torque value of the screw conveyor 3 at the same time as step S23 or at a predetermined timing before or after it. In addition, when inferring the control parameters in step S24, first, the image data of the separated liquid SL is input to the first inference unit 871, and the feature amount of the output image data is input to the fourth inference unit 874 together with the three values acquired by the additional variable acquisition unit 89.

As described above, according to the machine learning device, machine learning method, and data processing system of this embodiment, it is expected that the number of training data sets required to generate each trained model capable of inference with sufficient accuracy can be further reduced compared to the second and third embodiments described above.

Fifth embodiment
The image data included in the input data in the first to fourth embodiments described above is only the image data of the separated liquid SL. However, there are other image data that can be used as input data besides the image data of the separated liquid SL. Therefore, as a fifth embodiment, the image data included in the input data includes image data of the dehydrated solid matter M in addition to the image data of the separated liquid SL. Note that, among the components of the machine learning device 120 and the data processing system 180 according to the fifth embodiment described below, those common to the components of the machine learning devices 20, 20A to 20C and the data processing systems 80, 80A to 80C according to the first to fourth embodiments are given the same reference numerals and their description is omitted. In addition, the machine learning device, the machine learning method, and the data processing system according to the fifth embodiment described below are described by taking as an example a case where they are applied to the centrifuge system shown in FIG. 1 and FIG. 2, as in the first to fourth embodiments. Furthermore, all the modified examples described in the above or later embodiments are applicable to this embodiment to the extent that no contradiction occurs.

FIG. 15 is a schematic block diagram of a machine learning device 120 according to a fifth embodiment of the present disclosure. As shown in FIG. 15, the machine learning device 120 according to this embodiment includes a learning dataset acquisition unit 121, a fifth learning dataset storage unit 225, a fifth learning unit 235, and a learned model storage unit 124.

The learning dataset acquisition unit 121 is similar to the learning dataset acquisition unit 21 of the first embodiment, and is an interface unit that acquires multiple data constituting a learning dataset via a wired or wireless communication line. The learning dataset acquisition unit 121 acquires multiple data, including first image data capturing an image of the separated liquid SL, second image data capturing an image of the dehydrated solids M, and control parameters associated with these image data. Examples of control parameters in this embodiment include the amount of chemical agent added to the treated liquid PL1, the centrifugal force of the bowl 2, and the differential speed controlled by the differential speed generating device 5.

An example of a method for acquiring multiple data acquired by the learning data set acquisition unit 121 will be described. As shown in FIG. 15, the learning data set acquisition unit 121 is connected to a computer PC1 and acquires desired data from the computer PC1. In addition to the control unit 30 and the separated liquid monitoring system 40, the computer PC1 is also connected directly or indirectly via the control unit 30 to a dehydrated solids monitoring system 50 installed in the solids discharge conduit 7. This allows each control parameter to be acquired from the control unit 30, first image data from the separated liquid monitoring system 40, and second image data from the dehydrated solids monitoring system 50.

Here, as shown in FIG. 15, the dehydrated solids monitoring system 50 according to this embodiment can include a window 52 provided at an arbitrary position in the solids discharge conduit 7, and a dehydrated solids imaging camera 53 that is installed in the window 52 at a predetermined angle of view and can capture image data of the dehydrated solids M passing through the solids discharge conduit 7. Of these, a well-known camera capable of capturing two-dimensional images can be used as the dehydrated solids imaging camera 53. In addition, in this dehydrated solids monitoring system 50, a light source (not shown) that illuminates the dehydrated solids M passing through the solids discharge conduit 7 and a polarizing filter (not shown) that can be attached to the dehydrated solids imaging camera 53 can be appropriately used so that the image data captured by the dehydrated solids imaging camera 53 can easily reflect the state of the dehydrated solids M, particularly the surface state. In addition, in FIG. 2, the solid discharge conduit 7 extends substantially vertically, and the dehydrated solids M pass through the solid discharge conduit 7 so as to fall naturally, but the installation angle of the solid discharge conduit 7 can be changed as appropriate. Furthermore, in this embodiment, the dehydrated solids monitoring system 5 is installed in the solids discharge conduit 7, but the dehydrated solids monitoring system 5 can be installed at any position where the dehydrated solids M pass, such as the main conveying path 7a or the solids chute 6a. Note that the configuration of the dehydrated solids monitoring system 50 is not limited to this, and various configurations can be adopted as long as they are capable of acquiring image data of the dehydrated solids M. As a specific example, a configuration such as the dehydrated solids monitoring system 50A shown in FIG. 19 described later may be adopted.

Figure 16 shows an example of image data obtained by the dehydrated solid imaging camera 53 of the dehydrated solid monitoring system 50 according to the fifth embodiment of the present disclosure. In the technical field of water treatment equipment, the amount of liquid components contained in the dehydrated solid M discharged after the solid-liquid separation process by the centrifuge system, i.e., a small moisture content, can be used as an indicator for inferring that the solid-liquid separation process is being performed well. In addition, in the image data obtained by the dehydrated solid monitoring system 50, if the amount of liquid components contained in the dehydrated solid M is large, the color tends to become darker and the brightness tends to increase. Therefore, as shown in Figure 16, it can be inferred that image data in which the moisture content of the dehydrated solid M is relatively high (for example, as shown in Figure 16 (b)) tends to have a larger amount of color change between pixels than image data in which the moisture content is relatively low (for example, as shown in Figure 16 (a)).

The fifth learning data set storage unit 225 is a database for storing a plurality of data constituting the learning data set acquired by the learning data set acquisition unit 121, by associating related input data and output data into one learning data set. In this embodiment, the fifth learning data set stored in the fifth learning data set storage unit 225 may be a fifth input data set consisting of first image data obtained by imaging the separated liquid SL discharged from the bowl 2 from a predetermined angle of view and second image data obtained by imaging the dehydrated solids M discharged from the solids discharge port 2a from a predetermined angle of view, and a fifth output data set consisting of control parameters associated with the image data as the fifth input data. The control parameters may include at least one of the centrifugal force of the bowl 2, the amount of chemicals supplied from the additive supply source 11, and the differential speed between the bowl 2 and the screw conveyor 3.

The fifth learning unit 235 learns a learning model that infers a correlation between the fifth input data and the fifth output data by inputting multiple sets of the fifth learning datasets stored in the fifth learning dataset storage unit 225. In this embodiment, as in the first embodiment described above, supervised learning using a neural network is adopted as a specific method of machine learning. Although the specific machine learning method in this fifth learning unit 235 uses a different learning dataset for learning, the process is the same as the machine learning method according to the first embodiment shown in FIG. 6. In addition, the learned model storage unit 124 is a database for storing the learned model generated in the fifth learning unit 235.

FIG. 17 is a schematic block diagram showing a data processing system 180 according to a fifth embodiment of the present disclosure. As shown in FIG. 17, the data processing system 180 according to this embodiment mainly includes a first image data acquisition unit 81, a second image data acquisition unit 181, a parameter adjustment unit 82, a calculation unit 183, a database (DB) 84, a user interface 85, and an internal bus 86 for connecting these to each other. Of the above-mentioned components, the first image data acquisition unit 81, the parameter adjustment unit 82, the database 84, the user interface 85, and the internal bus 86 are the same as those of the data processing system 80 according to the first embodiment described above, so the same reference numerals are used and their description is omitted.

The second image data acquisition unit 181 is for acquiring second image data. Specifically, it is connected to the dehydrated solids monitoring system 50 and acquires the second image data, which is image data of the dehydrated solids M captured by the dehydrated solids imaging camera 53.

The arithmetic unit 183 constitutes a processor for implementing various processes in the data processing system 180, and includes at least a fifth inference unit 875. The arithmetic unit 183 is also connected to a learned model storage unit 188 that stores a learned model used in the fifth inference unit 875, i.e., a fifth learned model learned in the fifth learning unit 235. The fifth inference unit 875 infers control parameters to be adjusted by the parameter adjustment unit 82 from the first and second image data as state variables acquired by the first and second image data acquisition units 81, 181, by taking into account a learned model stored in the learned model storage unit 188.

Figure 18 is a flowchart showing an example of parameter adjustment by the data processing system according to the fifth embodiment of the present disclosure. When the decanter 1 to which the data processing system 180 is applied starts to operate, as shown in Figure 18, the data processing system 180 first determines whether it is time to adjust the parameters (step S21). When it detects that it is time to adjust the parameters (Yes in step S21), it operates the separated liquid imaging camera 43 in the separated liquid monitoring system 40 and the dehydrated solid imaging camera 53 in the dehydrated solid monitoring system 50 to capture images of the separated liquid SL flowing in the separated liquid discharge conduit 8 and the dehydrated solid M passing through the solid discharge conduit 7 (step S221). The first and second image data captured here are acquired in the data processing system 180 by the first and second image data acquisition units 81 and 181 (step S231).

Next, the first and second image data acquired by the first and second image data acquisition units 81, 181 are sent to the fifth inference unit 875 via the internal bus 86, and the image data are input to the input layer of one learned model in the learned model storage unit 188 that has been specified in advance in the fifth inference unit 875, thereby inferring control parameters (step S24). Then, the parameter adjustment unit 82 adjusts each control unit using the values of the control parameters inferred here (step S25). After that, the process returns to step S21 and enters a standby state.

As described above, according to the machine learning device 120 and machine learning method of this embodiment, by adopting two pieces of image data as input data, it becomes easier to learn the correlation between the input data and the output data, and each trained model capable of inference with sufficient accuracy can be generated with a relatively small number of training data sets. Furthermore, according to the data processing system 180 of this embodiment, the preferred control parameters of the decanter 1 can be estimated from only two pieces of image data, the image data of the separated liquid SL and the image data of the dehydrated solids M, so that this data processing system 180 can be applied to the decanter 1 relatively easily. Furthermore, adjustment of the control parameters can be easily performed without relying on the judgment of the operator EN.

Sixth embodiment
In the above fifth embodiment, the machine learning device, the machine learning method, and the data processing system have been described in terms of inferring a control parameter from the first and second image data using one trained model. In the following, as a modification of the fifth embodiment, a case in which a plurality of trained models are used to infer a control parameter from the first and second image data will be described as a sixth embodiment of the present disclosure. Note that among the components of the machine learning device 120A and the data processing system 180A according to the sixth embodiment described below, those components common to the components of the machine learning device 120 and the data processing system 180 according to the fifth embodiment are given the same reference numerals and their description will be omitted. In addition, the machine learning device, the machine learning method, and the data processing system according to the sixth embodiment described below are described by taking as an example a case in which they are applied to the centrifuge system shown in FIG. 1 and FIG. 2, as in the first to fifth embodiments. Furthermore, all the modifications described in the above or later-described embodiments are applicable to this embodiment to the extent that no contradiction occurs.

FIG. 19 is a schematic block diagram of a machine learning device 120A according to a sixth embodiment of the present disclosure. As shown in FIG. 19, the machine learning device 120A according to this embodiment includes, as its components, a first learning dataset storage unit 221, a sixth learning dataset storage unit 226, and a seventh learning dataset storage unit 227 as learning dataset storage units, and includes a first learning unit 231, a sixth learning unit 236, and a seventh learning unit 237 as learning units. Except for this, the machine learning device 120 according to the fifth embodiment is similar to the machine learning device 120 according to the fifth embodiment. In addition, in relation to this, the multiple data acquired in the learning dataset acquisition unit 121 are also different from those in the fifth embodiment. Note that the first learning dataset storage unit 221 and the first learning unit 231 are similar to those described in the machine learning device 20A according to the second embodiment described above, and therefore the same reference numerals are used and their description is omitted.

The multiple data acquired by the learning dataset acquisition unit 121 according to this embodiment include, in addition to the first and second image data and the control parameters, the feature quantities of the separated liquid SL and the feature quantities of the second image data of the image of the dehydrated solids M (hereinafter also referred to as the "feature quantities of the dehydrated solids M"). As the feature quantities of the dehydrated solids M, various pieces of information identified based on the image of the dehydrated solids M can be adopted. Specifically, for example, one or more of the moisture content, color value, density, etc. of the dehydrated solids M can be included. The feature quantities can be identified, for example, by actually sampling the dehydrated solids M and measuring and analyzing them using various measuring instruments, or by the operator EN visually judging the feature quantities.

In order to measure the above-mentioned characteristic quantities, it is preferable that the dehydrated solid matter monitoring system 50A according to this embodiment adopts a configuration in which the dehydrated solid matter M is sampled and then the second image data is acquired. Specifically, as shown in FIG. 19, the dehydrated solid matter monitoring system 50A may include a tray 54 having a predetermined depth capable of being filled with the dehydrated solid matter M, a photographing box 55 in which the tray 54 can be placed and which has a light source therein, and a dehydrated solid matter imaging camera 53 capable of imaging the surface of the dehydrated solid matter M filled in the tray 54 from a predetermined angle of view. The dehydrated solid matter M filled in the tray 54 is extracted at a predetermined timing using a sampling unit (not shown) provided at an appropriate position on the conveying path of the dehydrated solid matter M, such as the solid matter discharge conduit 7 or the main conveying path 7a. Then, the dehydrated solid matter M extracted in the tray 54 is imaged by the dehydrated solid matter imaging camera 53 to generate the second image data, and the desired control parameters corresponding to this image data are specified by the operator EN. Furthermore, by measuring and analyzing the dehydrated solids M sampled in the tray 54, it is possible to identify the characteristics of the dehydrated solids M.

The multiple data acquired in the learning data set acquisition unit 121 are stored separately as three learning data sets in the first learning data set storage unit 221, the sixth learning data set storage unit 226, and the seventh learning data set storage unit 227, taking into consideration their respective correspondences. The first learning data set stored in the first learning data set storage unit 221 may include, as first input data, first image data obtained by capturing the separated liquid SL discharged from the bowl 2 from a predetermined angle of view, and may include, as first output data, a feature amount of the separated liquid SL associated with the first input data. The sixth learning data set stored in the sixth learning data set storage unit 226 may include, as sixth input data, second image data obtained by capturing the dehydrated solid matter M discharged from the bowl 2 from a predetermined angle of view, and may include, as sixth output data, a feature amount of the dehydrated solid matter M associated with the sixth input data. Furthermore, the seventh learning data set stored in the seventh learning data set storage unit 227 may include the characteristic quantities of the separated liquid SL and the characteristic quantities of the dehydrated solids M as seventh input data, and may include control parameters associated with the seventh input data as seventh output data.

The first, sixth and seventh learning datasets stored in the first learning dataset storage unit 221, the sixth learning dataset storage unit 226 and the seventh learning dataset storage unit 227, respectively, are referred to only by other learning units. The first learning unit 231 learns a learning model that infers the correlation between the first input data and the first output data by inputting a plurality of sets of the first learning dataset. In other words, the first learning model that infers the feature amount of the first image data by inputting the first image data in the first learning dataset. The sixth learning unit 236 learns a learning model that infers the correlation between the sixth input data and the sixth output data by inputting a plurality of sets of the sixth learning dataset. In other words, the sixth learning unit 236 learns a learning model that infers the correlation between the sixth input data and the sixth output data by inputting the second image data in the sixth learning dataset. Furthermore, the seventh learning unit 237 learns a learning model that infers a correlation between the seventh input data and the seventh output data by inputting multiple sets of the seventh learning data set. In other words, the seventh learning model that infers a control parameter is learned by inputting the features of the first and second image data in the seventh learning data set.

The specific machine learning methods in the first, sixth and seventh learning units 231, 236 and 237 use different learning datasets, but the steps are similar to the supervised learning steps shown in FIG. 6. The first, sixth and seventh trained models obtained through a series of machine learning steps are then stored in the trained model storage unit 124, respectively.

Fig. 20 is a schematic block diagram showing a data processing system 180A according to a sixth embodiment of the present disclosure. As shown in Fig. 20, the data processing system 180A according to this embodiment has the same components as the data processing system 180 according to the fifth embodiment described above, except that the data processing system 180A according to this embodiment has a first inference unit 871, a sixth inference unit 876, and a seventh inference unit 877 as inference units in the arithmetic unit 183.

The first inference unit 871 performs inference using the first learned model generated by the above-mentioned machine learning device 120A and stored in the learned model storage unit 188. Therefore, when the first image data acquired by the first image data acquisition unit 81 is input, the first inference unit 871 outputs the feature amount of the first image data to the output layer. Furthermore, the sixth inference unit 876 performs inference using the sixth learned model generated by the above-mentioned machine learning device 120A and stored in the learned model storage unit 188. Therefore, when the second image data acquired by the second image data acquisition unit 181 is input, the sixth inference unit 876 outputs the feature amount of the second image data to the output layer. Furthermore, the seventh inference unit 877 performs inference using the seventh learned model generated by the above-mentioned machine learning device 120A and stored in the learned model storage unit 188. Therefore, when the seventh inference unit 877 receives the features of the first and second image data inferred by the first inference unit 871 and the sixth inference unit 876, it outputs control parameters to the output layer.

When adjusting the control parameters in a decanter 1 to which a data processing system 180A including the above-mentioned first, sixth and seventh inference units 871, 876 and 877 is applied, a process similar to that shown in FIG. 18 described above may be performed. However, when inferring the control parameters in step S24 of the process shown in FIG. 18, the first and second image data are first input to the first inference unit 871 and the sixth inference unit 876, and the feature quantities of the output first and second image data are input to the seventh inference unit 877.

As described above, according to the machine learning device, machine learning method, and data processing system of the present embodiment, by obtaining inference results for control parameters using multiple trained models, it is expected that the number of training data sets required to generate each trained model capable of inference with sufficient accuracy can be relatively reduced.

Seventh embodiment
In the machine learning device, the machine learning method, and the data processing system according to the fifth embodiment, the case where the control parameters are inferred only from the first and second image data has been described. In the following, as a modified example of the fifth embodiment, a case where other variables are adopted as input data in addition to the first and second image data will be described as the seventh embodiment of the present disclosure. Note that among the components of the machine learning device 120B and the data processing system 180B according to the seventh embodiment described below, those components common to the components of the machine learning devices 120, 120A and the data processing systems 180, 180A according to the fifth and sixth embodiments are given the same reference numerals and their description will be omitted. In addition, the machine learning device, the machine learning method, and the data processing system according to the seventh embodiment described below are described by taking as an example a case where they are applied to the centrifuge system shown in FIG. 1 and FIG. 2, as in the first to sixth embodiments. Furthermore, all the modified examples described in the embodiments described above or below can be applied to this embodiment to the extent that no contradiction occurs.

FIG. 21 is a schematic block diagram of a machine learning device 120B according to a seventh embodiment of the present disclosure. As shown in FIG. 21, the machine learning device 120B according to this embodiment is similar to the machine learning device 120 according to the fifth embodiment described above, except that the machine learning device 120B according to this embodiment includes an eighth learning data set storage unit 228 as a learning data set storage unit and an eighth learning unit 238 as a learning unit. In addition, the multiple data acquired in the learning data set acquisition unit 121 according to this embodiment are different from those according to the fifth embodiment. As shown in FIG. 21, the learning data set acquisition unit 121 according to this embodiment is connected to a computer PC1 and acquires desired data from this computer PC1. Like the computer PC1 according to the fifth embodiment, this computer PC1 is connected to the control unit 30, the separated liquid monitoring system 40, and the dewatered solids monitoring system 50, but in addition to these, it is also connected to at least one of the treated liquid monitoring system 60 and the screw conveyor torque monitoring system 70. In this embodiment, the computer PC1 is connected to both the treated liquid monitoring system 60 and the screw conveyor torque monitoring system 70, and the multiple data acquired by the learning data set acquisition unit 121 includes both the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3. The treated liquid monitoring system 60 and the screw conveyor torque monitoring system 70 may be similar to those exemplified in the third embodiment described above, and detailed description thereof will be omitted here.

Various data acquired by these four monitoring systems 40, 50, 60 and 70, specifically, the first and second image data, the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3, are sent directly or via the control unit 30 to the computer PC1, and are sent from the computer PC1 to the learning dataset acquisition unit 121 together with the corresponding desired control parameters. The multiple data constituting the learning dataset acquired by the learning dataset acquisition unit 121 are stored in the eighth learning dataset storage unit 228 in the form of an eighth learning dataset in which the first and second image data, the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3 are the eighth input data, and the control parameters associated with these data are the eighth output data. Then, in the eighth learning unit 238, machine learning is performed using the eighth learning dataset in a manner similar to the machine learning method of the fifth embodiment described above, and the obtained eighth trained model is stored in the trained model storage unit 124.

Figure 22 is a schematic block diagram showing a data processing system 180B according to a seventh embodiment of the present disclosure. As shown in Figure 22, the data processing system 180B according to this embodiment has the same components as the data processing system 180 according to the fifth embodiment described above, except that it has an eighth inference unit 878 as an inference unit and an additional variable acquisition unit 89. Of these, the additional variable acquisition unit 89 is connected to the treated liquid monitoring system 60 and the screw conveyor torque monitoring system 70 as shown in Figure 22, and acquires the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3 from these monitoring systems.

When adjusting the control parameters in a decanter 1 to which this data processing system 180B is applied, a process similar to that shown in FIG. 18 described above can be performed. However, in the process shown in FIG. 18, simultaneously with step S231 or at a predetermined timing before or after it, the additional variable acquisition unit 89 acquires the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3. These values are then associated with the input layer of the trained model in the eighth inference unit 878, together with the first and second image data.

As described above, according to the machine learning device, machine learning method, and data processing system of this embodiment, in addition to the first and second image data, the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3 are used as input data, making it easier to identify the correlation between the input data and the output data in the learning stage of the neural network model. This is expected to reduce the number of learning datasets required to generate each trained model capable of inference with sufficient accuracy.

Eighth embodiment
Next, a case where the technical ideas described in the sixth and seventh embodiments are combined will be described below as an eighth embodiment of the present disclosure. Among the components of the machine learning device 120C and the data processing system 180C according to the eighth embodiment described below, those components common to the machine learning devices 120, 120A, and 120B and the data processing systems 180, 180A, and 180B according to the fifth to seventh embodiments are given the same reference numerals and their description will be omitted. In addition, the machine learning device, the machine learning method, and the data processing system according to the eighth embodiment described below are described by taking as an example a case where they are applied to the centrifuge system shown in FIG. 1 and FIG. 2, as in the first to seventh embodiments. Furthermore, all the modified examples described in the above or later embodiments are applicable to this embodiment to the extent that no contradiction occurs.

FIG. 23 is a schematic block diagram of a machine learning device 120C according to an eighth embodiment of the present disclosure. As shown in FIG. 23, the machine learning device 120C according to this embodiment is similar to the machine learning device 120 according to the fifth embodiment except that, as components, it has a first learning dataset storage unit 221, a sixth learning dataset storage unit 226, and a ninth learning dataset storage unit 229 as learning dataset storage units, and has a first learning unit 231, a sixth learning unit 236, and a ninth learning unit 239 as learning units. In relation to this, the multiple data acquired in the learning dataset acquisition unit 121 are also different from those in the fifth embodiment.

The learning data set acquisition unit 121 according to this embodiment is connected to a computer PC1 as shown in FIG. 23, and acquires desired data from the computer PC1. This computer PC1 is connected to the control unit 30, the separated liquid monitoring system 40A, and the dehydrated solid matter monitoring system 50A, as in the computer PC1 according to the fifth embodiment, but in addition to these, it is also connected to at least one of the treated liquid monitoring system 60 and the screw conveyor torque monitoring system 70. The specific configurations of the control unit 30, the separated liquid monitoring system 40A, the dehydrated solid matter monitoring system 50A, the treated liquid monitoring system 60, and the screw conveyor torque monitoring system 70 connected to the computer PC1 are the same as those exemplified in the other embodiments already described above. In addition to the first and second image data and the control parameters, the multiple data acquired by the learning data set acquisition unit 121 in this embodiment further include at least one of the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3, the feature value of the separated liquid SL, and the feature value of the dehydrated solid matter M. In this embodiment, as in the seventh embodiment, the computer PC1 is connected to both the treated liquid monitoring system 60 and the screw conveyor torque monitoring system 70, and the multiple data acquired by the learning data set acquisition unit 121 includes both the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3.

The multiple data acquired in the learning data set acquisition unit 121 are stored separately as three learning data sets in the first learning data set storage unit 221, the sixth learning data set storage unit 226, and the ninth learning data set storage unit 229, taking into consideration their respective correspondences. The first learning data set stored in the first learning data set storage unit 221 may include image data (first image data) of the separated liquid SL obtained by capturing the separated liquid discharged from the bowl 2 from a predetermined angle of view as the first input data, and may include the feature amount of the separated liquid SL associated with the first input data as the first output data. The sixth learning data set stored in the sixth learning data set storage unit 226 may include image data (second image data) of the dehydrated solid matter M discharged from the solid matter discharge port 2a captured from a predetermined angle of view as the sixth input data, and may include the feature amount of the dehydrated solid matter M associated with the sixth input data as the sixth output data. Furthermore, the ninth learning data set stored in the ninth learning data set storage unit 229 may include, as ninth input data, the characteristic quantities of the separated liquid SL, the characteristic quantities of the dehydrated solids M, the slurry concentration of the treated liquid PL1, and the torque value of the screw conveyor 3, and may include, as ninth output data, control parameters associated with the ninth input data.

The first, sixth and ninth learning datasets stored in the first learning dataset storage unit 221, the sixth learning dataset storage unit 226 and the ninth learning dataset storage unit 229, respectively, are referred to only by other learning units. The first learning unit 231 learns a learning model that infers the correlation between the first input data and the first output data by inputting a plurality of sets of the first learning dataset. In other words, the first learning model that infers the feature amount of the first image data by inputting the first image data in the first learning dataset. The sixth learning unit 236 learns a learning model that infers the correlation between the sixth input data and the sixth output data by inputting a plurality of sets of the sixth learning dataset. In other words, the sixth learning unit 236 learns a learning model that infers the correlation between the sixth input data and the sixth output data by inputting the second image data in the sixth learning dataset. The ninth learning unit 239 learns a learning model that infers the correlation between the ninth input data and the ninth output data by inputting multiple sets of the ninth learning data set. In other words, the ninth learning model that infers the control parameters is learned by inputting the feature amounts of the first and second image data in the ninth learning data set, the slurry concentration of the liquid to be treated PL1, and the torque value of the screw conveyor 3.

The specific machine learning methods in the first, sixth and ninth learning units 231, 236 and 239 use different learning datasets, but the steps are similar to the supervised learning steps shown in FIG. 6. The first, sixth and ninth trained models obtained through a series of machine learning steps are then stored in the trained model storage unit 124.

FIG. 24 is a schematic block diagram showing a data processing system 180C according to an eighth embodiment of the present disclosure. As shown in FIG. 24, the data processing system 180C according to this embodiment includes a first inference unit 871, a sixth inference unit 876, and a ninth inference unit 879 as inference units in an arithmetic unit 183, and includes the same components as the data processing system 180 according to the fifth embodiment described above, except that the data processing system 180C includes an additional variable acquisition unit 89. Of these, the additional variable acquisition unit 89 may be the same as that described in the seventh embodiment.

The first inference unit 871 performs inference using the first learned model generated by the above-mentioned machine learning device 120C and stored in the learned model storage unit 188. Therefore, when the first image data acquired by the first image data acquisition unit 81 is input, the first inference unit 871 outputs the feature amount of the first image data to the output layer. Furthermore, the sixth inference unit 876 performs inference using the sixth learned model generated by the above-mentioned machine learning device 120C and stored in the learned model storage unit 188. Therefore, when the second image data acquired by the second image data acquisition unit 181 is input, the sixth inference unit 876 outputs the feature amount of the second image data to the output layer. Furthermore, the ninth inference unit 879 performs inference using the ninth learned model generated by the above-mentioned machine learning device 120C and stored in the learned model storage unit 188. Therefore, when the ninth inference unit 879 receives the features of the first image data inferred by the first inference unit 871, the features of the second image data inferred by the sixth inference unit 876, the slurry concentration of the treated liquid PL1, and the torque value of the screw conveyor 3, it outputs control parameters to the output layer.

When adjusting the control parameters in the decanter 1 to which the data processing system 180C including the above-mentioned additional variable acquisition unit 89 and the first, sixth and ninth inference units 871, 876 and 879 is applied, the same process as that shown in FIG. 18 may be performed. However, the process shown in FIG. 18 further includes a step in which the additional variable acquisition unit 89 acquires the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3 at the same time as step S231 or at a predetermined timing before or after it. In addition, when inferring the control parameters in step S24, first, the first and second image data are input to the first inference unit 871 and the sixth inference unit 876, respectively, and the feature quantities of the output first and second image data are input to the ninth inference unit 879 together with the two values acquired by the additional variable acquisition unit 89.

As described above, according to the machine learning device, machine learning method, and data processing system of this embodiment, it is expected that the number of training data sets required to generate each trained model capable of inference with sufficient accuracy can be further reduced compared to the sixth and seventh embodiments described above.

Ninth embodiment
Next, as a ninth embodiment, a description will be given of an input data including image data of the floc-containing liquid to be treated PL2 in addition to image data of the separation liquid SL. Among the components of the machine learning device 320 and the data processing system 380 according to the ninth embodiment described below, those components common to the components of the machine learning device and the data processing system according to the first to eighth embodiments are given the same reference numerals and their description will be omitted. Furthermore, the machine learning device, the machine learning method, and the data processing system according to the ninth embodiment described below are described by taking as an example a case where they are applied to the centrifuge system shown in FIG. 1 and FIG. 2, as in the first to eighth embodiments. Furthermore, all the modified examples described in the above-mentioned or later-mentioned embodiments are also applicable to this embodiment to the extent that no contradiction occurs.

FIG. 25 is a schematic block diagram of a machine learning device 320 according to a ninth embodiment of the present disclosure. As shown in FIG. 25, the machine learning device 320 according to this embodiment includes a learning dataset acquisition unit 321, a tenth learning dataset storage unit 2210, a tenth learning unit 2310, and a trained model storage unit 324.

The learning dataset acquisition unit 321 is similar to the learning dataset acquisition units 21 and 121 described above, and is an interface unit that acquires multiple data constituting a learning dataset via a wired or wireless communication line. This learning dataset acquisition unit 321 acquires, as multiple data, first image data capturing an image of the separation liquid SL, third image data capturing an image of the floc-containing treated liquid PL2, and control parameters associated with these image data. Examples of control parameters in this embodiment include the supply amount of chemicals added to the treated liquid PL1, the centrifugal force of the bowl 2, and the differential speed controlled by the differential speed generating device 5.

An example of a method for acquiring multiple data acquired by the learning data set acquisition unit 321 will be described. As shown in FIG. 25, the learning data set acquisition unit 321 is connected to a computer PC1 and acquires desired data from the computer PC1. In addition to the control unit 30 and the separated liquid monitoring system 40, the computer PC1 is also connected directly or indirectly via the control unit 30 to the treated liquid monitoring system 60 installed in the treated liquid piping 10b. This makes it possible to acquire each control parameter from the control unit 30, the first image data from the separated liquid monitoring system 40, and the third image data from the treated liquid monitoring system 60.

Here, as the treated liquid monitoring system 60 according to the present embodiment, as shown in FIG. 25, a configuration similar to that of the separated liquid monitoring system 40 can be adopted. That is, a system including a sight glass 61 connected to an arbitrary location downstream of the connection position of at least the additive pipe 11c of the treated liquid piping 10b, and a floc-containing treated liquid imaging camera 63 installed at a predetermined angle of view in a window 62 provided in the sight glass 61 and capable of capturing image data of the floc-containing treated liquid PL2 flowing in the sight glass 61 can be adopted. Of these, a well-known camera capable of capturing two-dimensional images can be adopted as the floc-containing treated liquid imaging camera 63. In addition, in this treated liquid monitoring system 60, a light source (not shown) that illuminates the floc-containing treated liquid PL2 in the sight glass 61 and a polarizing filter (not shown) that can be attached to the floc-containing treated liquid imaging camera 63 can be appropriately adopted so that the state of the floc-containing treated liquid PL2 is easily reflected in the image data captured by the floc-containing treated liquid imaging camera 63. The configuration of the treated liquid monitoring system 60 is not limited to this, and various configurations can be adopted as long as they are capable of acquiring image data of the floc-containing treated liquid PL2. As a specific example, when the chemical is supplied through the additive pipe 11c connected to the liquid supply pipe 9, a configuration may be adopted in which the floc-containing treated liquid imaging camera 63 is disposed at a position capable of imaging the space 3c inside the screw conveyor 3, or a configuration such as the treated liquid monitoring system 60A shown in FIG. 33, which will be described later, may be adopted.

Figure 26 shows an example of image data obtained by the floc-containing treated liquid imaging camera 63 of the treated liquid monitoring system 60 according to the fifth embodiment of the present disclosure. It is generally known that when performing solid-liquid separation processing using a centrifuge system, if the solid components in the treated liquid fed into the decanter 1 are sufficiently aggregated by the chemical solution, the solid components and the liquid components can be well separated. In addition, in the third image data obtained by the treated liquid monitoring system 60, if the floc aggregation of the floc-containing treated liquid PL2 is well performed, one floc aggregate tends to become coarse and the transparency of the area where the floc aggregate is not present tends to increase. Therefore, as shown in Figure 26, it can be inferred that the image data in which the floc aggregation reaction in the floc-containing treated liquid PL2 is well proceeding (for example, the one shown in Figure 26(b)) tends to have a greater amount of color change between pixels than the image data in which the reaction is not well proceeding due to, for example, a shortage of the chemical supply (for example, the one shown in Figure 26(a)).

The tenth learning data set storage unit 2210 is a database for storing a plurality of data constituting the learning data set acquired by the learning data set acquisition unit 321, by associating related input data and output data into one learning data set. In this embodiment, the tenth learning data set stored in the tenth learning data set storage unit 2210 may be tenth input data including first image data obtained by capturing an image of the separated liquid SL discharged from the bowl 2 from a predetermined angle of view, and third image data obtained by capturing an image of the floc-containing treated liquid PL2 from a predetermined angle of view before being supplied to the bowl 2 and after the addition of the chemical, and tenth output data including a control parameter associated with the image data as the tenth input data. The control parameter may include at least one of the centrifugal force of the bowl 2, the amount of chemical supplied from the additive supply source 11, and the differential speed between the bowl 2 and the screw conveyor 3.

The tenth learning unit 2310 learns a learning model that infers a correlation between the tenth input data and the tenth output data by inputting a plurality of sets of the fifth learning dataset stored in the tenth learning dataset storage unit 2210. In this embodiment, as in the first and fifth embodiments described above, supervised learning using a neural network is adopted as a specific method of machine learning. The specific machine learning method in this tenth learning unit 2310 is similar to the machine learning method according to the first embodiment shown in FIG. 6, although the learning dataset used for learning is different. In addition, the trained model storage unit 324 is similar to the trained model storage units 24 and 124 described above, and is a database for storing trained models generated by the tenth learning unit 2310.

FIG. 27 is a schematic block diagram showing a data processing system 380 according to a ninth embodiment of the present disclosure. As shown in FIG. 27, the data processing system 380 according to this embodiment mainly includes a first image data acquisition unit 81, a third image data acquisition unit 381, a parameter adjustment unit 82, a calculation unit 383, a database (DB) 84, a user interface 85, and an internal bus 86 for connecting these to each other. Of the above-mentioned components, the first image data acquisition unit 81, the parameter adjustment unit 82, the database 84, the user interface 85, and the internal bus 86 may be similar to those of the data processing systems 80 and 180 according to the first and fifth embodiments described above.

The third image data acquisition unit 381 is for acquiring third image data. Specifically, it is connected to the treated liquid monitoring system 60 and acquires the third image data, which is image data of the floc-containing treated liquid PL2 captured by the floc-containing treated liquid imaging camera 63.

The arithmetic unit 383 constitutes a processor for implementing various processes in the data processing system 380, and includes at least a tenth inference unit 8710. The arithmetic unit 383 is also connected to a learned model storage unit 388 that stores the learned model used in the tenth inference unit 8710, i.e., the tenth learned model learned in the tenth learning unit 2310. The tenth inference unit 8710 infers control parameters to be adjusted by the parameter adjustment unit 82 from the first and third image data as state variables acquired by the first and third image data acquisition units 81, 381, by taking into account a learned model stored in the learned model storage unit 388.

Figure 28 is a flowchart showing an example of parameter adjustment by the data processing system 380 according to the ninth embodiment of the present disclosure. When the decanter 1 to which the data processing system 380 is applied starts to be driven, as shown in Figure 28, the data processing system 380 first determines whether it is time to adjust the parameters (step S21). When it detects that it is time to adjust the parameters (Yes in step S21), it operates the separated liquid imaging camera 43 in the separated liquid monitoring system 40 and the floc-containing treated liquid imaging camera 63 in the treated liquid monitoring system 60 to capture images of the separated liquid SL flowing in the separated liquid discharge conduit 8 and the floc-containing treated liquid PL2 flowing in the treated liquid piping 10b (step S222). The first and third image data captured here are acquired in the data processing system 380 by the first and third image data acquisition units 81 and 381 (step S232).

Next, the first and third image data acquired by the first and third image data acquisition units 81, 381 are sent to the tenth inference unit 8710 via the internal bus 86, and the image data are input to the input layer of one learned model in the learned model storage unit 388 that has been specified in advance in the tenth inference unit 8710, thereby inferring control parameters (step S24). Then, the parameter adjustment unit 82 adjusts each control unit using the values of the control parameters inferred here (step S25). After that, the process returns to step S21 and enters a standby state.

As described above, according to the machine learning device 320 and machine learning method of this embodiment, by adopting two pieces of image data as input data, it becomes easier to learn the correlation between the input data and the output data, and each trained model capable of inference with sufficient accuracy can be generated with a relatively small number of training data sets. Furthermore, according to the data processing system 380 of this embodiment, the preferred control parameters of the decanter 1 can be estimated from only two pieces of image data, the image data of the separated liquid SL and the image data of the floc-containing treated liquid PL2, so that this data processing system 380 can be applied to the decanter 1 relatively easily. Furthermore, adjustment of the control parameters can be easily performed without relying on the judgment of the operator EN.

Tenth embodiment
In the above ninth embodiment, the machine learning device, the machine learning method, and the data processing system have been described in terms of inferring control parameters from the first and third image data using one trained model. In the following, as a modification of the above ninth embodiment, a case in which a plurality of trained models are used to infer control parameters from the first and third image data will be described as a tenth embodiment of the present disclosure. Note that among the components of the machine learning device 320A and the data processing system 380A according to the tenth embodiment described below, those components common to the components of the machine learning device 320 and the data processing system 380 according to the ninth embodiment are given the same reference numerals and their description will be omitted. In addition, the machine learning device, the machine learning method, and the data processing system according to the tenth embodiment described below are described, as in the first to ninth embodiments, by taking as an example a case in which they are applied to the centrifuge system shown in FIG. 1 and FIG. 2. Furthermore, all the modifications described in the above or later-described embodiments are applicable to this embodiment to the extent that no contradiction occurs.

FIG. 29 is a schematic block diagram of a machine learning device 320A according to a tenth embodiment of the present disclosure. As shown in FIG. 29, the machine learning device 320A according to this embodiment includes, as its components, a first learning dataset storage unit 221, an eleventh learning dataset storage unit 2211, and a twelfth learning dataset storage unit 2212 as learning dataset storage units, and includes a first learning unit 231, an eleventh learning unit 2311, and a twelfth learning unit 2312 as learning units. Except for this, the machine learning device 320A according to the ninth embodiment is similar to the machine learning device 320 according to the ninth embodiment. In addition, in relation to this, the multiple data acquired in the learning dataset acquisition unit 321 are also different from those in the ninth embodiment. Note that the first learning dataset storage unit 221 and the first learning unit 231 are similar to those described in the machine learning device 20A according to the second embodiment described above, and therefore the same reference numerals are used and their description is omitted.

The multiple data acquired by the learning dataset acquisition unit 321 according to this embodiment include, in addition to the first and third image data and the control parameters, the feature quantities of the separated liquid SL and the feature quantities of the third image data of the image of the floc-containing treated liquid PL2 (hereinafter also referred to as the "feature quantities of the floc-containing treated liquid PL2"). As the feature quantities of the floc-containing treated liquid PL2, various pieces of information identified based on the image of the floc-containing treated liquid PL2 can be adopted. Specifically, for example, one or more of the size, color value, density, etc. of the floc aggregates can be included. The feature quantities can be identified, for example, by actually sampling the floc-containing treated liquid PL2 and measuring and analyzing it using various measuring instruments, or by the operator EN visually judging it.

In order to measure the above-mentioned characteristic quantities, it is preferable that the treated liquid monitoring system 60A according to this embodiment adopts a configuration in which the floc-containing treated liquid PL2 is sampled and then the third image data is acquired. Specifically, the treated liquid monitoring system 60A may be the same as the above-mentioned separation liquid monitoring system 40A, that is, the system may include a branch pipe 64 provided in the treated liquid piping 10b, a valve 65 provided in the branch pipe 64, a transparent container 66 disposed at the end of the branch pipe 64, and a camera 63 for capturing images of the floc-containing treated liquid installed on the side of the transparent container 66.

The multiple data acquired in the learning data set acquisition unit 321 according to this embodiment are stored separately as three learning data sets in the first learning data set storage unit 221, the eleventh learning data set storage unit 2211, and the twelfth learning data set storage unit 2212, taking into consideration their respective correspondences. The first learning data set stored in the first learning data set storage unit 221 may include, as first input data, first image data obtained by capturing the separation liquid SL discharged from the bowl 2 from a predetermined angle of view, and may include, as first output data, a feature amount of the separation liquid SL associated with the first input data. The eleventh learning data set stored in the eleventh learning data set storage unit 2211 may include, as eleventh input data, third image data obtained by capturing the floc-containing treatment liquid PL2 from a predetermined angle of view before it is supplied to the bowl and after the agent is added, and may include, as eleventh output data, a feature amount of the floc-containing treatment liquid PL2 associated with the eleventh input data. Furthermore, the 12th learning data set stored in the 12th learning data set storage unit 2212 may include the feature amounts of the separated liquid SL and the feature amounts of the floc-containing treated liquid PL2 as 12th input data, and may include control parameters associated with the 12th input data as 12th output data.

The first, eleventh and twelfth learning datasets stored in the first learning dataset storage unit 221, the eleventh learning dataset storage unit 2211 and the twelfth learning dataset storage unit 2212, respectively, are referred to only by other learning units. The first learning unit 231 learns a learning model that infers the correlation between the first input data and the first output data by inputting a plurality of sets of the first learning dataset. In other words, the first learning model that infers the feature amount of the first image data by inputting the first image data in the first learning dataset. The eleventh learning unit 2311 learns a learning model that infers the correlation between the eleventh input data and the eleventh output data by inputting a plurality of sets of the eleventh learning dataset. In other words, the eleventh learning unit learns an eleventh learning model that infers the feature amount of the third image data by inputting the third image data in the eleventh learning dataset. Furthermore, the 12th learning unit 2312 learns a learning model that infers a correlation between the 12th input data and the 12th output data by inputting a plurality of sets of the 12th learning data set. In other words, the 12th learning model that infers a control parameter is learned by inputting the features of the first and third image data in the 12th learning data set.

The specific machine learning methods in the first, eleventh, and twelfth learning units 231, 2311, and 2312 use different learning datasets, but the steps are similar to the supervised learning steps shown in FIG. 6. The first, eleventh, and twelfth trained models obtained through a series of machine learning steps are stored in the trained model storage unit 324, respectively.

Fig. 30 is a schematic block diagram showing a data processing system 380A according to a tenth embodiment of the present disclosure. As shown in Fig. 30, the data processing system 380A according to this embodiment has the same components as the data processing system 380 according to the ninth embodiment described above, except that the data processing system 380A according to this embodiment has a first inference unit 871, an eleventh inference unit 8711, and a twelfth inference unit 87!2 as inference units in the arithmetic unit 383.

The first inference unit 871 performs inference using the first trained model generated by the above-mentioned machine learning device 320A and stored in the trained model storage unit 388. Therefore, when the first image data acquired by the first image data acquisition unit 81 is input, the first inference unit 871 outputs the feature amount of the first image data to the output layer. Furthermore, the eleventh inference unit 8711 performs inference using the eleventh trained model generated by the above-mentioned machine learning device 320A and stored in the trained model storage unit 388. Therefore, when the third image data acquired by the third image data acquisition unit 381 is input, the eleventh inference unit 8711 outputs the feature amount of the third image data to the output layer. Furthermore, the twelfth inference unit 8712 performs inference using the twelfth trained model generated by the above-mentioned machine learning device 320A and stored in the trained model storage unit 388. Therefore, when the features of the first and third image data inferred by the first and eleventh inference units 8711 are input, the twelfth inference unit 8712 outputs control parameters to the output layer.

When adjusting the control parameters in a decanter 1 to which a data processing system 380A including the above-mentioned first, eleventh and twelfth inference units 871, 8711 and 8712 is applied, a process similar to that shown in FIG. 28 described above may be performed. However, when inferring the control parameters in step S24 of the process shown in FIG. 28, the first and third image data are first input to the first inference unit 871 and the eleventh inference unit 8711, and the feature quantities of the output first and third image data are input to the twelfth inference unit 8712.

As described above, according to the machine learning device, machine learning method, and data processing system of the present embodiment, by obtaining inference results for control parameters using multiple trained models, it is expected that the number of training data sets required to generate each trained model capable of inference with sufficient accuracy can be relatively reduced.

Eleventh embodiment
In the machine learning device, the machine learning method, and the data processing system according to the ninth embodiment, the case where the control parameters are inferred only from the first and third image data has been described. In the following, as a modified example of the ninth embodiment, a case where other variables are adopted as input data in addition to the first and third image data will be described as an eleventh embodiment of the present disclosure. Note that among the components of the machine learning device 320B and the data processing system 380B according to the eleventh embodiment described below, those components common to the components of the machine learning devices 320, 320A and the data processing systems 380, 380A according to the ninth and tenth embodiments are given the same reference numerals and their description will be omitted. In addition, the machine learning device, the machine learning method, and the data processing system according to the eleventh embodiment described below are described by taking as an example a case where they are applied to the centrifuge system shown in FIG. 1 and FIG. 2, as in the first to tenth embodiments. Furthermore, all the modified examples described in the embodiments described above or below can be applied to this embodiment to the extent that no contradiction occurs.

FIG. 31 is a schematic block diagram of a machine learning device 320B according to an eleventh embodiment of the present disclosure. As shown in FIG. 31, the machine learning device 320B according to this embodiment is similar to the machine learning device 320 according to the ninth embodiment described above, except that the machine learning device 320B according to this embodiment includes a thirteenth training dataset storage unit 2213 as a training dataset storage unit and a thirteenth training unit 2313 as a training unit. In addition, the multiple data acquired in the training dataset acquisition unit 321 are also different from those according to the ninth embodiment.

As shown in FIG. 31, the learning data set acquisition unit 321 according to this embodiment is connected to a computer PC1 and acquires desired data from the computer PC1. Like the computer PC1 according to the ninth embodiment, this computer PC1 is connected to the control unit 30, the separated liquid monitoring system 40, and the treated liquid monitoring system 60, but in addition to these, it is also connected to at least one of the treated liquid concentration sensor 67, the dehydrated solids moisture content meter 56, and the screw conveyor torque monitoring system 70. Of these, the treated liquid concentration sensor 67 may be arranged, for example, at a predetermined position of the treated liquid piping 10b, particularly before it merges with the additive piping 11c, and may directly measure the slurry concentration of the treated liquid PL1 supplied from the treated liquid supply source 10. As the treated liquid concentration sensor 67, for example, a well-known laser type, optical type, or ultrasonic type concentration sensor can be used. The dehydrated solid moisture content meter 56 may be arranged in the solid discharge conduit 7 and directly measure the moisture content of the dehydrated solids M passing through the solid discharge conduit 7. For example, a well-known microwave moisture content meter may be used as the dehydrated solid moisture content meter 56. The treated liquid concentration sensor 67 and the dehydrated solid moisture content meter 56 are not shown in FIG. 2. In this embodiment, the computer PC1 is connected to all of the treated liquid concentration sensor 67, the dehydrated solid moisture content meter 56, and the screw conveyor torque monitoring system 70, and the learning data set acquisition unit 321 acquires a plurality of data including the slurry concentration of the treated liquid PL1 measured by the treated liquid concentration sensor 67, the moisture content of the dehydrated solids M discharged from the solid discharge port 2a measured by the dehydrated solid moisture content meter 56, and the torque value of the screw conveyor 3 measured by the screw conveyor torque monitoring system 70. The dehydrated solids moisture content meter 56 and screw conveyor torque monitoring system 70 in this embodiment may be similar to those exemplified in the third embodiment described above, and detailed description thereof will be omitted here.

Various data acquired by the above-mentioned monitoring system, specifically, the first and third image data, the slurry concentration of the liquid to be treated PL1, the moisture content of the dehydrated solids M, and the torque value of the screw conveyor 3 are sent directly or via the control unit 30 to the computer PC1, which then sends them to the learning dataset acquisition unit 321 together with the corresponding desired control parameters. The multiple data constituting the learning dataset acquired by the learning dataset acquisition unit 321 are stored in the 13th learning dataset storage unit 2213 in the form of a 13th learning dataset in which the first and third image data, the slurry concentration of the liquid to be treated PL1, the moisture content of the dehydrated solids M, and the torque value of the screw conveyor 3 are used as 13th input data, and the control parameters associated with these data are used as 13th output data. Then, in the 13th learning unit 2313, machine learning is performed using the 13th learning dataset in a manner similar to the machine learning method of the 13th embodiment described above, and the obtained 13th trained model is stored in the trained model storage unit 324.

Figure 32 is a schematic block diagram showing a data processing system 380B according to an eleventh embodiment of the present disclosure. As shown in Figure 32, the data processing system 380B according to this embodiment has the same components as the data processing system 380 according to the ninth embodiment described above, except that it has a thirteenth inference unit 8713 as an inference unit and an additional variable acquisition unit 89. Of these, the additional variable acquisition unit 89 is connected to the treated liquid concentration sensor 67, the dehydrated solids moisture content meter 56, and the screw conveyor torque monitoring system 70 as shown in Figure 32, and acquires the slurry concentration of the treated liquid PL1, the moisture content of the dehydrated solids M, and the torque value of the screw conveyor 3 from these monitoring systems and sensors.

When adjusting the control parameters in a decanter 1 to which this data processing system 380B is applied, a process similar to that shown in FIG. 28 described above can be performed. However, in the process shown in FIG. 28, simultaneously with step S232 or at a predetermined timing before or after it, the additional variable acquisition unit 89 acquires the slurry concentration of the treated liquid PL1, the moisture content of the dehydrated solids M, and the torque value of the screw conveyor 3. These values are then associated with the input layer of the trained model in the 13th inference unit 8713, together with the first and third image data.

As described above, according to the machine learning device, machine learning method, and data processing system of this embodiment, in addition to the first and third image data, the slurry concentration of the treated liquid PL1, the moisture content of the dehydrated solids M, and the torque value of the screw conveyor 3 are used as input data, making it easier to identify correlations between input data and output data in the learning stage of the neural network model. This is expected to reduce the number of learning datasets required to generate each trained model capable of inference with sufficient accuracy.

Twelfth embodiment
Next, a case where the technical ideas described in the tenth and eleventh embodiments are combined will be described below as a twelfth embodiment of the present disclosure. Among the components of the machine learning device 320C and the data processing system 380C according to the twelfth embodiment described below, those components common to the machine learning devices 320, 320A, 320B and the data processing systems 380, 380A, 380B according to the ninth to eleventh embodiments are given the same reference numerals and their description will be omitted. In addition, the machine learning device, the machine learning method, and the data processing system according to the twelfth embodiment described below are described by taking as an example a case where they are applied to the centrifuge system shown in FIG. 1 and FIG. 2, as in the first to eleventh embodiments. Furthermore, all the modified examples described in the above or later embodiments can be applied to this embodiment to the extent that no contradiction occurs.

FIG. 33 is a schematic block diagram of a machine learning device 320C according to a twelfth embodiment of the present disclosure. As shown in FIG. 33, the machine learning device 320C according to this embodiment is similar to the machine learning device 120 according to the ninth embodiment except that, as components, it has a first learning dataset storage unit 221, an eleventh learning dataset storage unit 2211, and a fourteenth learning dataset storage unit 2214 as learning dataset storage units, and has a first learning unit 231, an eleventh learning unit 2311, and a fourteenth learning unit 2314 as learning units. In relation to this, the multiple data acquired in the learning dataset acquisition unit 121 are also different from those in the ninth embodiment.

As shown in FIG. 33, the learning data set acquisition unit 321 according to this embodiment is connected to a computer PC1 and acquires desired data from the computer PC1. Like the computer PC1 according to the ninth embodiment, this computer PC1 is connected to the control unit 30, the separated liquid monitoring system 40A, and the treated liquid monitoring system 60A, but in addition to these, it is also connected to at least one of the treated liquid concentration sensor 67, the dehydrated solids moisture content meter 56, and the screw conveyor torque monitoring system 70. The specific configurations of the control unit 30, the separated liquid monitoring system 40A, the treated liquid monitoring system 60A, the treated liquid concentration sensor 67, the dehydrated solids moisture content meter 56, and the screw conveyor torque monitoring system 70 connected to the computer PC1 are the same as those exemplified in the other embodiments already described above. In addition to the first and third image data and the control parameters, the multiple data acquired by the learning data set acquisition unit 321 in this embodiment further includes at least one of the slurry concentration of the treated liquid PL1, the moisture content of the dehydrated solids M, and the torque value of the screw conveyor 3, the feature value of the separated liquid SL, and the feature value of the floc-containing treated liquid PL2. In this embodiment, as in the ninth embodiment, the computer PC1 is connected to all of the treated liquid concentration sensor 67, the dehydrated solids moisture content meter 56, and the screw conveyor torque monitoring system 70, and the multiple data acquired by the learning data set acquisition unit 121 includes three items: the slurry concentration of the treated liquid PL1, the dehydrated solids moisture content meter 56, and the torque value of the screw conveyor 3.

The multiple data acquired in the learning dataset acquisition unit 321 are stored separately as three learning datasets in the first learning dataset storage unit 221, the eleventh learning dataset storage unit 2211, and the fourteenth learning dataset storage unit 2214, taking into consideration their respective correspondences. The first learning dataset stored in the first learning dataset storage unit 221 may include image data (first image data) of the separated liquid SL obtained by capturing an image of the separated liquid discharged from the bowl 2 from a predetermined angle of view as first input data, and may include a feature amount of the separated liquid SL associated with the first input data as first output data. The eleventh learning data set stored in the eleventh learning data set storage unit 2211 may include, as eleventh input data, image data (third image data) of the floc-containing treated liquid PL2 before being supplied to the bowl 2 and after the addition of the chemical agent, captured from a predetermined angle of view, and may include, as eleventh output data, feature amounts of the floc-containing treated liquid PL2 associated with the eleventh input data. Furthermore, the fourteenth learning data set stored in the fourteenth learning data set storage unit 2214 may include, as fourteenth input data, feature amounts of the separated liquid SL, feature amounts of the floc-containing treated liquid PL2, the slurry concentration of the treated liquid PL1, the moisture content of the dehydrated solids M, and the torque value of the screw conveyor 3, and may include, as fourteenth output data, control parameters associated with the fourteenth input data.

The first, eleventh and fourteenth learning datasets stored in the first learning dataset storage unit 221, the eleventh learning dataset storage unit 2211 and the fourteenth learning dataset storage unit 2214, respectively, are referred to only by other learning units. The first learning unit 231 learns a learning model that infers the correlation between the first input data and the first output data by inputting a plurality of sets of the first learning dataset. In other words, the first learning model that infers the feature amount of the first image data by inputting the first image data in the first learning dataset. The eleventh learning unit 2311 learns a learning model that infers the correlation between the eleventh input data and the eleventh output data by inputting a plurality of sets of the eleventh learning dataset. In other words, the eleventh learning unit learns an eleventh learning model that infers the feature amount of the third image data by inputting the third image data in the eleventh learning dataset. The 14th learning unit 2314 learns a learning model that infers the correlation between the 14th input data and the 14th output data by inputting multiple sets of the 14th learning data set. In other words, the 14th learning model that infers the control parameters is learned by inputting the feature amounts of the first and third image data in the 14th learning data set, the slurry concentration of the treated liquid PL1, the moisture content of the dehydrated solids M, and the torque value of the screw conveyor 3.

The specific machine learning methods in the first, eleventh and fourteenth learning units 231, 2311 and 2314 use different learning datasets, but the steps are similar to the supervised learning steps shown in FIG. 6. The first, eleventh and fourteenth trained models obtained through a series of machine learning steps are each stored in the trained model storage unit 324.

FIG. 34 is a schematic block diagram showing a data processing system 380C according to a twelfth embodiment of the present disclosure. As shown in FIG. 34, the data processing system 380C according to this embodiment includes a first inference unit 871, an eleventh inference unit 8711, and a fourteenth inference unit 8714 as inference units in an arithmetic unit 383, and includes the same components as the data processing system 380 according to the ninth embodiment described above, except that the data processing system 380C according to this embodiment includes an additional variable acquisition unit 89. Of these, the additional variable acquisition unit 89 may be the same as that described in the eleventh embodiment.

The first inference unit 871 performs inference using the first learned model generated by the above-mentioned machine learning device 320C and stored in the learned model storage unit 388. Therefore, when the first image data acquired by the first image data acquisition unit 81 is input, the first inference unit 871 outputs the feature amount of the first image data to the output layer. Furthermore, the eleventh inference unit 8711 performs inference using the eleventh learned model generated by the above-mentioned machine learning device 320C and stored in the learned model storage unit 388. Therefore, when the third image data acquired by the third image data acquisition unit 381 is input, the eleventh inference unit 8711 outputs the feature amount of the third image data to the output layer. Furthermore, the fourteenth inference unit 8714 performs inference using the fourteenth learned model generated by the above-mentioned machine learning device 320C and stored in the learned model storage unit 188. Therefore, when the 14th inference unit 8714 receives the feature values of the first image data inferred by the first inference unit 871, the feature values of the third image data inferred by the 11th inference unit 8711, the slurry concentration of the treated liquid PL1, the moisture content of the dehydrated solids M, and the torque value of the screw conveyor 3, it outputs control parameters to the output layer.

When adjusting the control parameters in the decanter 1 to which the data processing system 380C including the above-mentioned additional variable acquisition unit 89 and the first, eleventh and fourteenth inference units 871, 8711 and 8714 is applied, the same process as that shown in FIG. 28 may be performed. However, the process shown in FIG. 28 further includes a step in which the additional variable acquisition unit 89 acquires the slurry concentration of the treated liquid PL1, the water content of the dehydrated solid matter M and the torque value of the screw conveyor 3 at the same time as step S232 or at a predetermined timing before or after it. In addition, when inferring the control parameters in step S24, first, the first and third image data are input to the first inference unit 871 and the eleventh inference unit 8711, respectively, and the feature quantities of the output first and third image data are input to the fourteenth inference unit 8714 together with the three values acquired by the additional variable acquisition unit 89.

As described above, according to the machine learning device, machine learning method, and data processing system of this embodiment, it is expected that the number of training data sets required to generate each trained model capable of inference with sufficient accuracy can be further reduced compared to the tenth and eleventh embodiments described above.

Thirteenth embodiment
Next, as a thirteenth embodiment, a description will be given of a case in which the image data included in the input data includes image data of the dehydrated solid matter M and image data of the floc-containing liquid to be treated PL2 in addition to image data of the separated liquid SL. Among the components of the machine learning device 520 and the data processing system 580 according to the thirteenth embodiment described below, those components common to the components of the machine learning device and the data processing system according to the first to twelfth embodiments are given the same reference numerals and their description will be omitted. In addition, the machine learning device, the machine learning method, and the data processing system according to the thirteenth embodiment described below are described by taking as an example a case in which they are applied to the centrifuge system shown in FIG. 1 and FIG. 2, as in the first to twelfth embodiments. Furthermore, all the modified examples described in the above-mentioned or later-mentioned embodiments are applicable to this embodiment to the extent that no contradiction occurs.

FIG. 35 is a schematic block diagram of a machine learning device 520 according to a thirteenth embodiment of the present disclosure. As shown in FIG. 35, the machine learning device 520 according to this embodiment includes a learning dataset acquisition unit 521, a fifteenth learning dataset storage unit 2215, a fifteenth learning unit 2315, and a trained model storage unit 524.

The learning data set acquisition unit 521 is an interface unit similar to the learning data set acquisition units 21, 121, and 321 of the first, fifth, and ninth embodiments, and acquires a plurality of data constituting a learning data set via a wired or wireless communication line. The learning data set acquisition unit 521 acquires a plurality of data, including first image data of the separated liquid SL, second image data of the dehydrated solid matter M, and third image data of the floc-containing treated liquid PL2, as well as control parameters associated with these image data. In addition, examples of the control parameters according to this embodiment include the amount of chemicals added to the treated liquid PL1, the centrifugal force of the bowl 2, and the differential speed controlled by the differential speed generating device 5. In addition, the learning data set acquisition unit 521 is connected to a computer PC1 as shown in FIG. 35, and acquires desired data from the computer PC1. In addition to the control unit 30 and the separated liquid monitoring system 40, the computer PC1 is also connected directly or indirectly via the control unit 30 to the dehydrated solids monitoring system 50 and the treated liquid monitoring system 60. This allows the control unit 30 to obtain each control parameter, the separated liquid monitoring system 40 to obtain first image data, the dehydrated solids monitoring system 50 to obtain second image data, and the treated liquid monitoring system 60 to obtain third image data. The specific configurations of the separated liquid monitoring system 40, the dehydrated solids monitoring system 50, and the treated liquid monitoring system 60 according to this embodiment have already been exemplified in the above-mentioned embodiments, and will not be described here.

The 15th learning data set storage unit 2215 is a database for storing a plurality of data constituting the learning data set acquired by the learning data set acquisition unit 521, by associating related input data and output data with each other to form one learning data set. In this embodiment, the fifth learning data set stored in the 15th learning data set storage unit 2215 may be a 15th input data set consisting of a first image data obtained by imaging the separated liquid SL discharged from the bowl 2 from a predetermined angle of view, a second image data obtained by imaging the dehydrated solids M discharged from the solids discharge port 2a from a predetermined angle of view, and a third image data obtained by imaging the floc-containing treated liquid PL2 from a predetermined angle of view before being supplied to the bowl 2 and after the chemical solution has been added, and a control parameter associated with the image data as the 15th input data may be a 15th output data set. In this embodiment, the control parameters are exemplified as including three parameters: the centrifugal force of the bowl 2, the amount of chemicals supplied from the additive supply source 11, and the differential speed between the bowl 2 and the screw conveyor 3, but it is sufficient to include at least one of these. In addition, four or more control parameters may be adopted, including further control parameters such as the amount of liquid to be treated PL1 supplied and the dam set diameter of the dam 2c of the bowl 2.

The fifteenth learning unit 2315 learns a learning model that infers a correlation between the fifteenth input data and the fifteenth output data by inputting a plurality of sets of the fifth learning dataset stored in the fifteenth learning dataset storage unit 2215. In this embodiment, as in the first embodiment and the like described above, supervised learning using a neural network is adopted as a specific method of machine learning. The specific machine learning method in this fifteenth learning unit 2315 is similar to the machine learning method according to the first embodiment shown in FIG. 6, although the learning dataset used for learning is different. The trained model storage unit 524 is a database for storing the trained model generated in the fifteenth learning unit 2315.

Figure 36 is a schematic block diagram showing a data processing system 580 according to the fifteenth embodiment of the present disclosure. As shown in Figure 36, the data processing system 580 according to this embodiment mainly includes a first image data acquisition unit 81, a second image data acquisition unit 181, a third image data acquisition unit 381, a parameter adjustment unit 82, a calculation unit 583, a database (DB) 84, a user interface 85, and an internal bus 86 for connecting these to each other. Among the above-mentioned components, the first image data acquisition unit 81, the parameter adjustment unit 82, the second image data acquisition unit 181, the third image data acquisition unit 381, the database 84, the user interface 85, and the internal bus 86 are the same as those of the data processing systems according to the first, fifth, and ninth embodiments described above, so they are denoted by the same reference numerals and their description is omitted.

The arithmetic unit 583 constitutes a processor for implementing various processes in the data processing system 580, and includes at least a fifteenth inference unit 8715. The arithmetic unit 583 is also connected to a learned model storage unit 588 that stores a learned model used in the fifteenth inference unit 8715, i.e., a fifteenth learned model learned in the fifteenth learning unit 2315. The fifteenth inference unit 8715 infers control parameters to be adjusted by the parameter adjustment unit 82 from the first to third image data as state variables acquired by the first to third image data acquisition units 81, 181, and 381, by taking into consideration a learned model stored in the learned model storage unit 588.

Figure 37 is a flowchart showing an example of parameter adjustment by the data processing system according to the fifteenth embodiment of the present disclosure. When the decanter 1 to which the data processing system 580 is applied starts to be driven, as shown in Figure 37, the data processing system 580 first determines whether it is time to adjust the parameters (step S21). When it detects that it is time to adjust the parameters (Yes in step S21), it operates the separated liquid imaging camera 43 in the separated liquid monitoring system 40, the dehydrated solid imaging camera 53 in the dehydrated solid monitoring system 50, and the floc-containing treated liquid imaging camera 63 in the treated liquid monitoring system 60 to capture images of the separated liquid SL flowing in the separated liquid discharge conduit 8, the dehydrated solids M passing through the solid discharge conduit 7, and the floc-containing treated liquid PL2 flowing in the treated liquid piping 10b (step S223). The first to third image data captured here are acquired in the data processing system 580 by the first to third image data acquisition units 81, 181, and 381 (step S233).

Next, the first to third image data acquired by the first to third image data acquisition units 81, 181, 381 are sent to the 15th inference unit 8715 via the internal bus 86, and the image data are input to the input layer of one learned model in the learned model storage unit 588 previously specified in the 15th inference unit 8715, thereby inferring control parameters (step S24). Then, the parameter adjustment unit 82 adjusts each control unit using the values of the control parameters inferred here (step S25). After that, the process returns to step S21 and enters a standby state.

As described above, according to the machine learning device 520 and machine learning method of this embodiment, by adopting three image data as input data, it becomes easier to learn the correlation between the input data and the output data, and each trained model capable of inference with sufficient accuracy can be generated with a relatively small number of training data sets. Furthermore, according to the data processing system 580 of this embodiment, the preferred control parameters of the decanter 1 can be estimated from only three image data, namely the image data of the separated liquid SL, the image data of the dehydrated solids M, and the image data of the floc-containing treated liquid PL2, so that this data processing system 580 can be applied to the decanter 1 relatively easily. Furthermore, the control parameters can be easily adjusted without relying on the judgment of the operator EN.

<Fourteenth embodiment>
In the above-described thirteenth embodiment, the machine learning device, the machine learning method, and the data processing system have been described in terms of inferring control parameters from the first to third image data using one trained model. In the following, as a modification of the above-described thirteenth embodiment, a case in which a plurality of trained models are used to infer control parameters from the first to third image data will be described as a fourteenth embodiment of the present disclosure. Note that among the components of the machine learning device 520A and the data processing system 580A according to the fourteenth embodiment described below, those components common to the components of the machine learning device 520 and the data processing system 580 according to the thirteenth embodiment are given the same reference numerals and their description will be omitted. In addition, the machine learning device, the machine learning method, and the data processing system according to the fourteenth embodiment described below are described by taking as an example a case in which they are applied to the centrifuge system shown in FIG. 1 and FIG. 2, as in the first to thirteenth embodiments. Furthermore, all the modifications described in the above-described or later-described embodiments are applicable to this embodiment to the extent that no contradiction occurs.

FIG. 38 is a schematic block diagram of a machine learning device 520A according to a fourteenth embodiment of the present disclosure. As shown in FIG. 38, the machine learning device 520A according to this embodiment includes, as its components, a first learning dataset storage unit 221, a sixth learning dataset storage unit 226, an eleventh learning dataset storage unit 2211, and a sixteenth learning dataset storage unit 2216 as learning dataset storage units, and includes a first learning unit 231, a sixth learning unit 236, an eleventh learning unit 2311, and a sixteenth learning unit 2316 as learning units. Except for this, the machine learning device 520 according to the thirteenth embodiment is similar to the machine learning device 520 according to the thirteenth embodiment. In addition, in relation to this, the multiple data acquired in the learning dataset acquisition unit 521 are also different from those in the thirteenth embodiment. Note that the first learning data set storage unit 221, the sixth learning data set storage unit 226, and the eleventh learning data set storage unit 2211, as well as the first learning unit 231, the sixth learning unit 236, and the eleventh learning unit 2311 have already been exemplified in the above-mentioned embodiments, and their description will be omitted here.

The multiple data acquired by the learning dataset acquisition unit 521 in this embodiment include, in addition to the first and second image data and the control parameters, the characteristic quantities of the separated liquid SL, the characteristic quantities of the dehydrated solids M, and the characteristic quantities of the floc-containing treated liquid PL2. These characteristic quantities have already been exemplified in the above-mentioned embodiments, and their description will be omitted here.

The multiple data acquired in the learning dataset acquisition unit 521 are stored separately as four learning datasets in the first learning dataset storage unit 221, the sixth learning dataset storage unit 226, the eleventh learning dataset storage unit 2211, and the sixteenth learning dataset storage unit 2216, taking into consideration their respective correspondences. Of these, the sixteenth learning dataset stored in the sixteenth learning dataset storage unit 2216 may include, as sixteenth input data, the characteristic amounts of the separated liquid SL, the characteristic amounts of the dehydrated solids M, and the characteristic amounts of the floc-containing treated liquid PL2, and may include, as sixteenth output data, control parameters associated with the sixteenth input data.

The first, sixth, eleventh and sixteenth learning datasets stored in the first learning dataset storage unit 221, the sixth learning dataset storage unit 226, the eleventh learning dataset storage unit 2211 and the sixteenth learning dataset storage unit 2216, respectively, are each referenced only by another learning unit. Of these, the sixteenth learning unit 2316 learns a learning model that infers the correlation between the sixteenth input data and the sixteenth output data by inputting a plurality of sets of the sixteenth learning dataset. In other words, it learns a sixteenth learning model that infers a control parameter by inputting the features of the first to third image data in the sixteenth learning dataset.

The specific machine learning methods in the first, sixth, eleventh and sixteenth learning units 231, 236, 2311 and 2316 use different learning datasets, but the steps are similar to the supervised learning steps shown in FIG. 6. The first, sixth, eleventh and sixteenth trained models obtained through a series of machine learning steps are each stored in the trained model storage unit 524.

Fig. 39 is a schematic block diagram showing a data processing system 580A according to a fourteenth embodiment of the present disclosure. As shown in Fig. 39, the data processing system 580A according to this embodiment has the same components as the data processing system 580 according to the thirteenth embodiment described above, except that the data processing system 580A according to this embodiment has a first inference unit 871, a sixth inference unit 876, an eleventh inference unit 8711, and a sixteenth inference unit 8716 as inference units in the arithmetic unit 583.

Of these inference units, the 16th inference unit 8716 performs inference using the 16th learned model generated by the above-mentioned machine learning device 520A and stored in the learned model storage unit 588. Therefore, when the feature quantities of the first to third image data inferred by the first inference unit 871, the sixth inference unit 876, and the 11th inference unit 8711 are input, the 16th inference unit 8716 outputs control parameters to the output layer. Note that the first inference unit 871, the sixth inference unit 876, and the 11th inference unit 8711 have already been exemplified in each of the above-mentioned embodiments, so their description will be omitted here.

When adjusting the control parameters in the decanter 1 to which the data processing system 580A including the above-mentioned first, sixth, eleventh and sixteenth inference units 871, 876, 8711 and 8716 is applied, the same process as that shown in FIG. 37 above may be performed. However, when inferring the control parameters in step S24 of the process shown in FIG. 37, the first to third image data are first input to the first inference unit 871, the sixth inference unit 876 and the eleventh inference unit 8711, and the feature quantities of the output first to third image data are input to the sixteenth inference unit 8716.

As described above, according to the machine learning device, machine learning method, and data processing system of the present embodiment, by obtaining inference results for control parameters using multiple trained models, it is expected that the number of training datasets required to generate each trained model capable of inference with sufficient accuracy can be relatively reduced.

<Fifteenth embodiment>
In the machine learning device, the machine learning method, and the data processing system according to the thirteenth embodiment, the case where the control parameters are inferred only from the first to third image data has been described. In the following, as a modified example of the thirteenth embodiment, a case where other variables are adopted as input data in addition to the first to third image data will be described as a fifteenth embodiment of the present disclosure. Note that among the components of the machine learning device 520B and the data processing system 580B according to the fifteenth embodiment described below, those components common to the components of the machine learning devices 520, 520A and the data processing systems 580, 580A according to the thirteenth and fourteenth embodiments are given the same reference numerals and their description will be omitted. In addition, the machine learning device, the machine learning method, and the data processing system according to the fifteenth embodiment described below are described by taking as an example a case where they are applied to the centrifuge system shown in FIG. 1 and FIG. 2, as in the first to fourteenth embodiments. Furthermore, all the modified examples described in the embodiments described above or below can be applied to this embodiment to the extent that no contradiction occurs.

Figure 40 is a schematic block diagram of a machine learning device 520B according to the fifteenth embodiment of the present disclosure. As shown in Figure 40, the machine learning device 520B according to this embodiment is similar to the machine learning device 520 according to the thirteenth embodiment described above, except that the machine learning device 520B according to this embodiment includes a seventeenth learning data set storage unit 2217 as a learning data set storage unit and a seventeenth learning unit 2317 as a learning unit. In addition, the multiple data acquired in the learning data set acquisition unit 521 are also different from those according to the thirteenth embodiment. As shown in Figure 40, the learning data set acquisition unit 521 according to this embodiment is connected to a computer PC1 and acquires desired data from this computer PC1. Like the computer PC1 according to the thirteenth embodiment, this computer PC1 is connected to the control unit 30, the separated liquid monitoring system 40, the dewatered solids monitoring system 50, and the treated liquid monitoring system 60, but in addition to these, it is also connected to at least one of the treated liquid concentration sensor 67 and the screw conveyor torque monitoring system 70. In this embodiment, the computer PC1 is connected to both the treated liquid concentration sensor 67 and the screw conveyor torque monitoring system 70, and the multiple data acquired by the learning data set acquisition unit 521 includes both the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3. The treated liquid concentration sensor 67 and the screw conveyor torque monitoring system 70 may be the same as those exemplified in the above-mentioned embodiment, and detailed description thereof will be omitted here.

Various data acquired by these monitoring systems, specifically, the first to third image data, the slurry concentration of the treated liquid PL1, and the torque value of the screw conveyor 3 are sent directly or via the control unit 30 to the computer PC1, and are sent from the computer PC1 to the learning dataset acquisition unit 521 together with the corresponding desired control parameters. The multiple data constituting the learning dataset acquired by the learning dataset acquisition unit 521 are stored in the 17th learning dataset storage unit 2217 in the form of a 17th learning dataset in which the first to third image data, the slurry concentration of the treated liquid PL1, and the torque value of the screw conveyor 3 are the 17th input data, and the control parameters associated with these data are the 17th output data. Then, in the 17th learning unit 2317, machine learning is performed using the 17th learning dataset in a manner similar to the machine learning method of the 13th embodiment described above, and the obtained 17th learned model is stored in the learned model storage unit 524.

Figure 41 is a schematic block diagram showing a data processing system 580B according to a fifteenth embodiment of the present disclosure. As shown in Figure 41, the data processing system 580B according to this embodiment has the same components as the data processing system 580 according to the thirteenth embodiment described above, except that it has a seventeenth inference unit 8717 as an inference unit and an additional variable acquisition unit 89. Of these, the additional variable acquisition unit 89 is connected to the treated liquid concentration sensor 67 and the screw conveyor torque monitoring system 70 as shown in Figure 41, and acquires the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3.

When adjusting the control parameters in a decanter 1 to which this data processing system 580B is applied, a process similar to that shown in FIG. 37 described above can be performed. However, in the process shown in FIG. 37, the additional variable acquisition unit 89 acquires the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3 at the same time as step S233 or at a predetermined timing before or after it. These values are then associated with the input layer of the trained model in the 17th inference unit 8717 together with the first to third image data.

As described above, according to the machine learning device, machine learning method, and data processing system of this embodiment, by adopting the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3 as input data in addition to the first to third image data, it becomes easier to identify the correlation between the input data and the output data in the learning stage of the neural network model. This is expected to reduce the number of learning datasets required to generate each trained model capable of inference with sufficient accuracy.

<Sixteenth embodiment>
Next, a combination of the technical ideas described in the above fourteenth and fifteenth embodiments will be described below as a sixteenth embodiment of the present disclosure. Among the components of the machine learning device 520C and the data processing system 580C according to the sixteenth embodiment described below, those components common to the machine learning devices 520, 520A, 520B and the data processing systems 580, 580A, 580B according to the thirteenth to fifteenth embodiments will be given the same reference numerals and their description will be omitted. In addition, the machine learning device, the machine learning method, and the data processing system according to the sixteenth embodiment described below will be described by taking as an example a case where they are applied to the centrifuge system shown in FIG. 1 and FIG. 2, as in the first to fifteenth embodiments. Furthermore, all the modified examples described in the above or later embodiments can be applied to this embodiment to the extent that no contradiction occurs.

FIG. 42 is a schematic block diagram of a machine learning device 520C according to a sixteenth embodiment of the present disclosure. As shown in FIG. 42, the machine learning device 520C according to this embodiment includes, as its components, a first learning dataset storage unit 221, a sixth learning dataset storage unit 226, an eleventh learning dataset storage unit 2211, and an eighteenth learning dataset storage unit 2218 as learning dataset storage units, and includes a first learning unit 231, a sixth learning unit 236, an eleventh learning unit 2311, and an eighteenth learning unit 2318 as learning units, except that the machine learning device 520 according to the thirteenth embodiment is similar to the machine learning device 520 according to the thirteenth embodiment. In addition, in relation to this, the multiple data acquired in the learning dataset acquisition unit 521 are also different from those in the fifth embodiment. Note that the first learning data set storage unit 221, the sixth learning data set storage unit 226, and the eleventh learning data set storage unit 2211, as well as the first learning unit 231, the sixth learning unit 236, and the eleventh learning unit 2311 have already been exemplified in the above-mentioned embodiments, and their description will be omitted here.

The learning data set acquisition unit 521 according to this embodiment is connected to a computer PC1 as shown in FIG. 42, and acquires desired data from the computer PC1. This computer PC1 is connected to the control unit 30, the separated liquid monitoring system 40A, the dehydrated solid matter monitoring system 50A, and the treated liquid monitoring system 60A, as in the computer PC1 according to the thirteenth embodiment, but in addition to these, it is also connected to at least one of the treated liquid concentration sensor 67 and the screw conveyor torque monitoring system 70. The specific configurations of the control unit 30, the separated liquid monitoring system 40A, the dehydrated solid matter monitoring system 50A, the treated liquid monitoring system 60A, the treated liquid concentration sensor 67, and the screw conveyor torque monitoring system 70 connected to the computer PC1 are the same as those exemplified in the other embodiments already described above. In addition to the first to third image data and the control parameters, the multiple data acquired by the learning data set acquisition unit 521 in this embodiment further includes at least one of the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3, the characteristic amount of the separated liquid SL, the characteristic amount of the dehydrated solids M, and the characteristic amount of the floc-containing treated liquid PL2. In this embodiment, as in the fourteenth embodiment, the computer PC1 is connected to both the treated liquid concentration sensor 67 and the screw conveyor torque monitoring system 70, and the multiple data acquired by the learning data set acquisition unit 521 includes both the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3.

The multiple data acquired in the learning data set acquisition unit 521 are stored separately as three learning data sets in the first learning data set storage unit 221, the sixth learning data set storage unit 226, the eleventh learning data set storage unit 2211, and the eighteenth learning data set storage unit 2218, taking into consideration their respective correspondences. Of these, the eighteenth learning data set stored in the eighteenth learning data set storage unit 2218 may include, as the eighteenth input data, the characteristic amount of the separated liquid SL, the characteristic amount of the dewatered solids M, the characteristic amount of the floc-containing treated liquid PL2, the slurry concentration of the treated liquid PL1, and the torque value of the screw conveyor 3, and may include, as the eighteenth output data, the control parameters associated with the eighteenth input data. Note that,

The first, sixth, eleventh and eighteenth learning data sets stored in the first learning data set storage unit 221, the sixth learning data set storage unit 226, the eleventh learning data set storage unit 2211 and the eighteenth learning data set storage unit 2218, respectively, are each referred to only by another learning unit. Among these, the eighteenth learning unit 2318 learns a learning model that infers the correlation between the eighteenth input data and the eighteenth output data by inputting a plurality of sets of the eighteenth learning data set. In other words, the eighteenth learning model that infers the control parameters is learned by inputting the feature amounts of the first to third image data in the eighteenth learning data set, the slurry concentration of the liquid to be treated PL1, and the torque value of the screw conveyor 3.

The specific machine learning methods in the first, sixth, eleventh and eighteenth learning units 231, 236, 2311 and 2318 use different learning datasets, but the steps are similar to the supervised learning steps shown in FIG. 6. The first, sixth, eleventh and eighteenth trained models obtained through a series of machine learning steps are each stored in the trained model storage unit 524.

Figure 43 is a schematic block diagram showing a data processing system 180C according to a sixteenth embodiment of the present disclosure. As shown in Figure 43, the data processing system 580C according to this embodiment includes a first inference unit 871, a sixth inference unit 876, an eleventh inference unit 8711, and an eighteenth inference unit 8718 as inference units in an arithmetic unit 583, and includes the same components as the data processing system 580 according to the thirteenth embodiment described above, except that the data processing system 580C according to this embodiment includes an additional variable acquisition unit 89. Of these, the additional variable acquisition unit 89 may be the same as that described in the fifteenth embodiment.

Of these inference units, the 18th inference unit 8718 performs inference using the 18th learned model generated by the machine learning device 520C described above and stored in the learned model storage unit 588. Therefore, when the feature amounts of the first to third image data inferred by the first inference unit 871, the sixth inference unit 876, and the 11th inference unit 8711, the slurry concentration of the treated liquid PL1, and the torque value of the screw conveyor 3 are input, the 18th inference unit 8718 outputs control parameters to the output layer. Note that the first inference unit 871, the sixth inference unit 876, and the 11th inference unit 8711 have already been exemplified in each of the above-mentioned embodiments, so their description will be omitted here.

When adjusting the control parameters in the decanter 1 to which the data processing system 580C including the above-mentioned additional variable acquisition unit 89 and the first, sixth, eleventh and eighteenth inference units 871, 876, 8711 and 8718 is applied, the same process as that shown in FIG. 37 may be performed. However, the process shown in FIG. 37 further includes a step in which the additional variable acquisition unit 89 acquires the slurry concentration of the treated liquid PL1 and the torque value of the screw conveyor 3 at the same time as step S233 or at a predetermined timing before or after it. In addition, when inferring the control parameters in step S24, first, the first to third image data are input to the first inference unit 871, the sixth inference unit 876 and the eleventh inference unit 8711, respectively, and the feature quantities of the output first to third image data are input to the 18th inference unit 8718 together with the two values acquired by the additional variable acquisition unit 89.

As described above, according to the machine learning device, machine learning method, and data processing system of this embodiment, it is expected that the number of training data sets required to generate each trained model capable of inference with sufficient accuracy can be further reduced compared to the 14th and 15th embodiments described above.

The above-described embodiment is merely an example, and the present disclosure is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit of the present disclosure. All of these are included in the technical ideas of the present disclosure. In addition, in the present disclosure, each component may exist in one or more than one form, as long as no contradiction arises.

1 Decanter (Centrifugal Separator System)
2 Bowl 2a Solid discharge outlet 2b Separated liquid discharge outlet 2c Dam 3 Screw conveyor 3c Space 4 Drive motor 5 Differential speed generating device 6 Casing 6a Solid chute 6b Separated liquid chute 7 Solid discharge conduit 7a Main conveying path 8 Separated liquid discharge conduit 9 Liquid supply pipe 10 Treated liquid supply source 10b Treated liquid piping 11 Additive supply source 11c Additive piping 20, 20A to C, 120, 120A to C, 320, 320A to C, 520, 520A to C Machine learning device 21, 121, 321, 521 Learning data set acquisition unit 22, 221 to 2218 Learning data set storage unit 23, 231 to 2318 Learning unit 24, 124, 324, 524 Learned model storage unit 30 Control unit 40, 40A Separated liquid monitoring system 50, 50A Dehydrated solids monitoring system 56 Dehydrated solids moisture content meter 60, 60A Treated liquid monitoring system 67 Treated liquid concentration sensor 70 Screw conveyor torque monitoring system 80, 80A-C, 180, 180A-C, 380, 380A-C, 580, 580A-C Data processing system 81 First image data acquisition unit 181 Second image data acquisition unit 381 Third image data acquisition unit 82 Parameter adjustment unit 83, 183, 383, 583 Arithmetic units 87, 871-8718 Inference units 88, 188, 388, 588 Trained model storage unit 89 Additional variable acquisition unit SL Separated liquid M Liquid-containing solids (dehydrated solids)
PL1: Liquid to be treated PL2: Floc-containing liquid to be treated

Claims (30)

A machine learning device for a centrifugation system including: a bowl that applies centrifugal force to a liquid to be treated to separate the liquid into solids and a separated liquid; a screw conveyor that transports the solids in the bowl toward a discharge outlet; a drive motor that rotates the bowl; and a differential speed generating device that rotates the screw conveyor at a differential speed relative to the bowl,
a learning dataset storage unit that stores a plurality of learning datasets each including input data including image data obtained by capturing an image of the separated liquid discharged from the bowl at a predetermined angle of view, and output data including control parameters associated with the input data, the control parameters including at least one of an amount of additive added to the liquid to be treated, a centrifugal force of the bowl, and a differential speed controlled by the differential speed generating device;
a learning unit that learns a learning model that infers a correlation between the input data and the output data by inputting a plurality of sets of the learning data sets;
A learned model storage unit that stores the learning model learned by the learning unit;
Machine learning device. The learning unit includes:
a first learning unit that learns a first learning model that infers features of the image data by inputting the image data in the learning dataset;
a second learning unit that learns a second learning model that infers the control parameter by inputting a feature amount of the image data;
The machine learning device according to claim 1 . The input data further includes at least one of a slurry concentration of the liquid to be treated, a moisture content of the dehydrated solid matter discharged from the discharge port, and a torque value of the screw conveyor.
The machine learning device according to claim 1 . The input data further includes at least one of a slurry concentration of the liquid to be treated, a moisture content of the dehydrated solid matter discharged from the discharge port, and a torque value of the screw conveyor;
The learning unit includes:
a first learning unit that learns a first learning model that infers features of the image data by inputting the image data in the learning dataset;
a fourth learning unit that learns a fourth learning model that infers the control parameters by inputting at least one of a slurry concentration of the liquid to be treated, a moisture content of the dewatered solid matter discharged from the discharge port, and a torque value of the screw conveyor, and a feature amount of the image data;
The machine learning device according to claim 1 . A machine learning device for a centrifugation system including: a bowl that applies centrifugal force to a liquid to be treated to separate the liquid into solids and a separated liquid; a screw conveyor that transports the solids in the bowl toward a discharge outlet; a drive motor that rotates the bowl; and a differential speed generating device that rotates the screw conveyor at a differential speed relative to the bowl,
a learning dataset storage unit that stores a plurality of learning datasets each including input data including first image data obtained by imaging the separated liquid discharged from the bowl at a predetermined angle of view and second image data obtained by imaging the dehydrated solid matter discharged from the discharge outlet at a predetermined angle of view, and output data including control parameters associated with the input data, the control parameters including at least one of an amount of additive added to the liquid to be treated, a centrifugal force of the bowl, and a differential speed controlled by the differential speed generating device;
a learning unit that learns a learning model that infers a correlation between the input data and the output data by inputting a plurality of sets of the learning data sets;
A learned model storage unit that stores the learning model learned by the learning unit;
Machine learning device. The learning unit includes:
a first learning unit that learns a first learning model that infers a feature amount of the first image data by inputting the first image data in the learning dataset;
a sixth learning unit that learns a sixth learning model that infers features of the second image data by inputting the second image data in the learning dataset;
a seventh learning unit that learns a seventh learning model that infers the control parameter by inputting the feature amount of the first image data and the feature amount of the second image data;
The machine learning device according to claim 5 . The input data further includes at least one of a slurry concentration of the liquid to be treated and a torque value of the screw conveyor.
The machine learning device according to claim 5 . The input data further includes at least one of a slurry concentration of the liquid to be treated and a torque value of the screw conveyor,
The learning unit includes:
a first learning unit that learns a first learning model that infers a feature amount of the first image data by inputting the first image data in the learning dataset;
a sixth learning unit that learns a sixth learning model that infers features of the second image data by inputting the second image data in the learning dataset;
a ninth learning unit that learns a ninth learning model that infers the control parameter by inputting at least one of a slurry concentration of the liquid to be treated and a torque value of the screw conveyor, a feature amount of the first image data, and a feature amount of the second image data;
The machine learning device according to claim 5 . A machine learning device for a centrifugation system including: a bowl that applies centrifugal force to a liquid to be treated to separate the liquid into solids and a separated liquid; a screw conveyor that transports the solids in the bowl toward a discharge outlet; a drive motor that rotates the bowl; and a differential speed generating device that rotates the screw conveyor at a differential speed relative to the bowl,
a learning dataset storage unit that stores a plurality of learning datasets each including input data including first image data obtained by capturing an image of the separated liquid discharged from the bowl at a predetermined angle of view and third image data obtained by capturing an image of the liquid to be treated from a predetermined angle of view before being supplied to the bowl and after a predetermined additive has been added thereto, and output data including control parameters associated with the input data, the control parameters including at least one of an amount of additive added to the liquid to be treated, a centrifugal force of the bowl, and a differential speed controlled by the differential speed generating device;
a learning unit that learns a learning model that infers a correlation between the input data and the output data by inputting a plurality of sets of the learning data sets;
A learned model storage unit that stores the learning model learned by the learning unit;
Machine learning device. The learning unit includes:
a first learning unit that learns a first learning model that infers a feature amount of the first image data by inputting the first image data in the learning dataset;
an eleventh learning unit that learns an eleventh learning model that infers features of the third image data by inputting the third image data in the learning dataset;
a twelfth learning unit that learns a twelfth learning model that infers the control parameter by inputting the feature amount of the first image data and the feature amount of the third image data;
The machine learning device according to claim 9 . The input data further includes at least one of a slurry concentration of the liquid to be treated, a moisture content of the dehydrated solid matter discharged from the discharge port, and a torque value of the screw conveyor.
The machine learning device according to claim 9 . The input data further includes at least one of a slurry concentration of the liquid to be treated, a moisture content of the dehydrated solid matter discharged from the discharge port, and a torque value of the screw conveyor;
The learning unit includes:
a first learning unit that learns a first learning model that infers a feature amount of the first image data by inputting the first image data in the learning dataset;
an eleventh learning unit that learns an eleventh learning model that infers features of the third image data by inputting the third image data in the learning dataset;
a fourteenth learning unit that learns a fourteenth learning model that infers the control parameter by inputting at least one of a slurry concentration of the liquid to be treated, a moisture content of the dewatered solid matter discharged from the discharge port, and a torque value of the screw conveyor, as well as a feature amount of the first image data and a feature amount of the third image data;
The machine learning device according to claim 9 . A machine learning device for a centrifugation system including: a bowl that applies centrifugal force to a liquid to be treated to separate the liquid into solids and a separated liquid; a screw conveyor that transports the solids in the bowl toward a discharge outlet; a drive motor that rotates the bowl; and a differential speed generating device that rotates the screw conveyor at a differential speed relative to the bowl,
a learning dataset storage unit that stores a plurality of learning datasets each including input data including first image data obtained by imaging the separated liquid discharged from within the bowl at a predetermined angle of view, second image data obtained by imaging the dehydrated solid matter discharged from the discharge port at a predetermined angle of view, and third image data obtained by imaging the liquid to be treated at a predetermined angle of view before being supplied to the bowl and after a predetermined additive has been added thereto, and output data including control parameters associated with the input data, wherein the control parameters include at least one of an amount of additive added to the liquid to be treated, a centrifugal force of the bowl, and a differential speed controlled by the differential speed generating device;
a learning unit that learns a learning model that infers a correlation between the input data and the output data by inputting a plurality of sets of the learning data sets;
A learned model storage unit that stores the learning model learned by the learning unit;
Machine learning device. The learning unit includes:
a first learning unit that learns a first learning model that infers a feature amount of the first image data by inputting the first image data in the learning dataset;
a sixth learning unit that learns a sixth learning model that infers features of the second image data by inputting the second image data in the learning dataset;
an eleventh learning unit that learns an eleventh learning model that infers features of the third image data by inputting the third image data in the learning dataset;
a sixteenth learning unit that learns a sixteenth learning model that infers the control parameter by inputting the feature amount of the first image data, the feature amount of the second image data, and the feature amount of the third image data;
The machine learning device according to claim 13. The input data further includes at least one of a slurry concentration of the liquid to be treated and a torque value of the screw conveyor.
The machine learning device according to claim 13. The input data further includes at least one of a slurry concentration of the liquid to be treated and a torque value of the screw conveyor,
The learning unit includes:
a first learning unit that learns a first learning model that infers a feature amount of the first image data by inputting the first image data in the learning dataset;
a sixth learning unit that learns a sixth learning model that infers features of the second image data by inputting the second image data in the learning dataset;
an eleventh learning unit that learns an eleventh learning model that infers features of the third image data by inputting the third image data in the learning dataset;
an 18th learning unit that learns an 18th learning model that infers the control parameter by inputting at least one of the slurry concentration of the liquid to be treated and the torque value of the screw conveyor, and the feature amounts of the first image data, the feature amounts of the second image data, and the feature amounts of the third image data;
The machine learning device according to claim 13. The control parameters further include a supply amount of the liquid to be treated.
The machine learning device according to any one of claims 1 to 16. The control parameters further comprise a dam set diameter of the bowl.
The machine learning device according to any one of claims 1 to 17. A data processing system for use in a centrifuge system including a bowl for applying centrifugal force to a liquid to be treated to separate the liquid into solids and a separated liquid, a screw conveyor for transporting the solids in the bowl toward a discharge outlet, a drive motor for rotating the bowl, and a differential speed generating device for rotating the screw conveyor at a differential speed relative to the bowl, comprising:
a first image data acquisition unit for acquiring first image data obtained by capturing an image of the separated liquid discharged from within the bowl at a predetermined angle of view;
an inference unit that infers a control parameter of the centrifuge system by inputting data acquired by the first image data acquisition unit into a trained model generated by the machine learning device according to claim 1 or 2;
Data processing system. A data processing system for use in a centrifuge system including a bowl for applying centrifugal force to a liquid to be treated to separate the liquid into solids and a separated liquid, a screw conveyor for transporting the solids in the bowl toward a discharge outlet, a drive motor for rotating the bowl, and a differential speed generating device for rotating the screw conveyor at a differential speed relative to the bowl,
a first image data acquisition unit for acquiring first image data obtained by capturing an image of the separated liquid discharged from within the bowl at a predetermined angle of view;
an additional variable acquisition unit for acquiring at least one of a slurry concentration of the liquid to be treated, a moisture content of the dewatered solid matter discharged from the discharge port, and a torque value of the screw conveyor;
and an inference unit that infers a control parameter of the centrifuge system by inputting data acquired by the first image data acquisition unit and the additional variable acquisition unit into a trained model generated by the machine learning device according to claim 3 or 4.
Data processing system. A data processing system for use in a centrifuge system including a bowl for applying centrifugal force to a liquid to be treated to separate the liquid into solids and a separated liquid, a screw conveyor for transporting the solids in the bowl toward a discharge outlet, a drive motor for rotating the bowl, and a differential speed generating device for rotating the screw conveyor at a differential speed relative to the bowl, comprising:
a first image data acquisition unit for acquiring first image data obtained by capturing an image of the separated liquid discharged from within the bowl at a predetermined angle of view;
a second image data acquisition unit for acquiring second image data obtained by capturing an image of the dehydrated solid matter discharged from the discharge outlet from a predetermined angle of view;
and an inference unit that infers a control parameter of the centrifuge system by inputting data acquired by the first image data acquisition unit and the second image data acquisition unit into a trained model generated by the machine learning device according to claim 5 or 6.
Data processing system. A data processing system for use in a centrifuge system including a bowl for applying centrifugal force to a liquid to be treated to separate the liquid into solids and a separated liquid, a screw conveyor for transporting the solids in the bowl toward a discharge outlet, a drive motor for rotating the bowl, and a differential speed generating device for rotating the screw conveyor at a differential speed relative to the bowl, comprising:
a first image data acquisition unit for acquiring first image data obtained by capturing an image of the separated liquid discharged from within the bowl at a predetermined angle of view;
a second image data acquisition unit for acquiring second image data obtained by capturing an image of the dehydrated solid matter discharged from the discharge outlet from a predetermined angle of view;
an additional variable acquisition unit for acquiring at least one of a slurry concentration of the liquid to be treated and a torque value of the screw conveyor;
and an inference unit that infers a control parameter of the centrifuge system by inputting data acquired by the first image data acquisition unit, the second image data acquisition unit, and the additional variable acquisition unit into a trained model generated by the machine learning device according to claim 7 or 8.
Data processing system. A data processing system for use in a centrifuge system including a bowl for applying centrifugal force to a liquid to be treated to separate the liquid into solids and a separated liquid, a screw conveyor for transporting the solids in the bowl toward a discharge outlet, a drive motor for rotating the bowl, and a differential speed generating device for rotating the screw conveyor at a differential speed relative to the bowl,
a first image data acquisition unit for acquiring first image data obtained by capturing an image of the separated liquid discharged from within the bowl at a predetermined angle of view;
a third image data acquisition unit for acquiring third image data obtained by capturing an image of the liquid to be treated from a predetermined angle of view before it is supplied to the bowl and after a predetermined additive has been added;
and an inference unit that infers a control parameter of the centrifuge system by inputting data acquired by the first image data acquisition unit and the third image data acquisition unit into a trained model generated by the machine learning device according to claim 9 or 10.
Data processing system. A data processing system for use in a centrifuge system including a bowl for applying centrifugal force to a liquid to be treated to separate the liquid into solids and a separated liquid, a screw conveyor for transporting the solids in the bowl toward a discharge outlet, a drive motor for rotating the bowl, and a differential speed generating device for rotating the screw conveyor at a differential speed relative to the bowl,
a first image data acquisition unit for acquiring first image data obtained by capturing an image of the separated liquid discharged from within the bowl at a predetermined angle of view;
a third image data acquisition unit for acquiring third image data obtained by capturing an image of the liquid to be treated from a predetermined angle of view before it is supplied to the bowl and after a predetermined additive has been added;
an additional variable acquisition unit for acquiring at least one of a slurry concentration of the liquid to be treated, a moisture content of the dewatered solid matter discharged from the discharge port, and a torque value of the screw conveyor;
an inference unit that infers a control parameter of the centrifuge system by inputting data acquired by the first image data acquisition unit, the third image data acquisition unit, and the additional variable acquisition unit into a trained model generated by the machine learning device according to claim 11 or 12;
Data processing system. A data processing system for use in a centrifuge system including a bowl for applying centrifugal force to a liquid to be treated to separate the liquid into solids and a separated liquid, a screw conveyor for transporting the solids in the bowl toward a discharge outlet, a drive motor for rotating the bowl, and a differential speed generating device for rotating the screw conveyor at a differential speed relative to the bowl, comprising:
a first image data acquisition unit for acquiring first image data obtained by capturing an image of the separated liquid discharged from within the bowl at a predetermined angle of view;
a second image data acquisition unit for acquiring second image data obtained by capturing an image of the dehydrated solid matter discharged from the discharge outlet from a predetermined angle of view;
a third image data acquisition unit for acquiring third image data obtained by capturing an image of the liquid to be treated from a predetermined angle of view before it is supplied to the bowl and after a predetermined additive has been added;
an inference unit that infers a control parameter of the centrifuge system by inputting data acquired by the first image data acquisition unit, the second image data acquisition unit, and the third image data acquisition unit into a trained model generated by the machine learning device according to claim 13 or 14;
Data processing system. A data processing system for use in a centrifuge system including a bowl for applying centrifugal force to a liquid to be treated to separate the liquid into solids and a separated liquid, a screw conveyor for transporting the solids in the bowl toward a discharge outlet, a drive motor for rotating the bowl, and a differential speed generating device for rotating the screw conveyor at a differential speed relative to the bowl,
a first image data acquisition unit for acquiring first image data obtained by capturing an image of the separated liquid discharged from within the bowl at a predetermined angle of view;
a second image data acquisition unit for acquiring second image data obtained by capturing an image of the dehydrated solid matter discharged from the discharge outlet from a predetermined angle of view;
a third image data acquisition unit for acquiring third image data obtained by capturing an image of the liquid to be treated from a predetermined angle of view before the liquid is supplied to the bowl and after a predetermined additive has been added;
an additional variable acquisition unit for acquiring at least one of a slurry concentration of the liquid to be treated and a torque value of the screw conveyor;
an inference unit that infers a control parameter of the centrifuge system by inputting data acquired by the first image data acquisition unit, the second image data acquisition unit, the third image data acquisition unit, and the additional variable acquisition unit into a trained model generated by the machine learning device according to claim 15 or 16;
Data processing system. A machine learning method using a computer for a centrifuge system including a bowl that applies centrifugal force to a liquid to be treated to separate the liquid into solids and a separated liquid, a screw conveyor that transports the solids in the bowl toward a discharge outlet, a drive motor that rotates the bowl, and a differential speed generating device that rotates the screw conveyor at a differential speed relative to the bowl, the method comprising:
a step of storing a plurality of sets of learning data sets including input data including image data obtained by capturing an image of the separated liquid discharged from the bowl at a predetermined angle of view, and output data including control parameters associated with the input data, the control parameters including at least one of an amount of additive added to the liquid to be treated, a centrifugal force of the bowl, and a differential speed controlled by the differential speed generating device;
learning a learning model that infers a correlation between the input data and the output data by inputting a plurality of sets of the learning data sets;
and storing the trained model.
Machine learning methods. A machine learning method using a computer for a centrifuge system including a bowl that applies centrifugal force to a liquid to be treated to separate the liquid into solids and a separated liquid, a screw conveyor that transports the solids in the bowl toward a discharge outlet, a drive motor that rotates the bowl, and a differential speed generating device that rotates the screw conveyor at a differential speed relative to the bowl, the method comprising:
a step of storing a plurality of sets of learning data sets including input data including first image data obtained by imaging the separated liquid discharged from the bowl at a predetermined angle of view and second image data obtained by imaging the dehydrated solid matter discharged from the discharge outlet at a predetermined angle of view, and output data including control parameters associated with the input data, the control parameters including at least one of an amount of additive added to the liquid to be treated, a centrifugal force of the bowl, and a differential speed controlled by the differential speed generating device;
learning a learning model that infers a correlation between the input data and the output data by inputting a plurality of sets of the learning data sets;
and storing the trained model.
Machine learning methods. A machine learning method using a computer for a centrifuge system including a bowl that applies centrifugal force to a liquid to be treated to separate the liquid into solids and a separated liquid, a screw conveyor that transports the solids in the bowl toward a discharge outlet, a drive motor that rotates the bowl, and a differential speed generating device that rotates the screw conveyor at a differential speed relative to the bowl, the method comprising:
a step of storing a plurality of sets of learning data sets including input data including first image data obtained by capturing an image of the separated liquid discharged from the bowl at a predetermined angle of view and third image data obtained by capturing an image of the liquid to be treated from a predetermined angle of view before being supplied to the bowl and after a predetermined additive has been added thereto, and output data including control parameters associated with the input data, the control parameters including at least one of an amount of additive added to the liquid to be treated, a centrifugal force of the bowl, and a differential speed controlled by the differential speed generating device;
learning a learning model that infers a correlation between the input data and the output data by inputting a plurality of sets of the learning data sets;
and storing the trained model.
Machine learning methods. A machine learning method using a computer for a centrifuge system including a bowl that applies centrifugal force to a liquid to be treated to separate the liquid into solids and a separated liquid, a screw conveyor that transports the solids in the bowl toward a discharge outlet, a drive motor that rotates the bowl, and a differential speed generating device that rotates the screw conveyor at a differential speed relative to the bowl, the method comprising:
a step of storing a plurality of sets of learning data sets including input data including first image data obtained by imaging the separated liquid discharged from within the bowl at a predetermined angle of view, second image data obtained by imaging the dehydrated solid matter discharged from the discharge port at a predetermined angle of view, and third image data obtained by imaging the liquid to be treated at a predetermined angle of view before being supplied to the bowl and after a predetermined additive has been added, and output data including control parameters associated with the input data, the control parameters including at least one of an amount of additive added to the liquid to be treated, a centrifugal force of the bowl, and a differential speed controlled by the differential speed generating device;
learning a learning model that infers a correlation between the input data and the output data by inputting a plurality of sets of the learning data sets;
and storing the trained model.
Machine learning methods.