JP7835563B2 - Method for predicting the quality of ready-mixed concrete, method for manufacturing ready-mixed concrete, and system for manufacturing ready-mixed concrete - Google Patents

Description

This disclosure relates to a method for predicting the quality of ready-mixed concrete, a method for manufacturing ready-mixed concrete, and a system for manufacturing ready-mixed concrete.

Patent Document 1 discloses a quality prediction method for predicting the quality of ready-mixed concrete. This method predicts the quality of ready-mixed concrete using multiple image data captured continuously inside a mixer while water and other concrete materials are being mixed, along with a prediction model.

Patent No. 6680936

This disclosure provides a method for predicting the quality of ready-mixed concrete, a method for manufacturing ready-mixed concrete, and a system for manufacturing ready-mixed concrete, all of which enable easy assessment of the concrete's quality.

A method for predicting the quality of ready-mixed concrete relating to one aspect of this disclosure includes an acquisition step and a prediction step. In the acquisition step, input information is acquired, including vibration information indicating the magnitude of vibrations caused by the dropping or flow of ready-mixed concrete produced by a mixer that mixes concrete materials. In the prediction step, the quality of the ready-mixed concrete is predicted based on a prediction model pre-built using machine learning to output quality information indicating the quality of the ready-mixed concrete in response to the input information, and the input information acquired in the acquisition step.

The above vibration information may include information indicating the magnitude of vibrations in two or more directions.

The above vibration information may also indicate the magnitude of vibration in a hopper, which is located below the mixer and capable of containing ready-mix concrete, when ready-mix concrete is discharged from the mixer.

The above vibration information may include information indicating the magnitude of hopper vibration in a direction intersecting the hopper's side wall.

The above vibration information may further include information indicating the magnitude of hopper vibration in the direction along the hopper's side wall.

The above vibration information may be the result of calculating at least one value selected from the group consisting of the maximum value, minimum value, mean value, and standard deviation of the moving average value of the hopper's acceleration during an evaluation period set to include the time when the ready-mixed concrete is discharged from the mixer to the hopper.

The above acquisition process may include the steps of: calculating the time at which the absolute value of the hopper acceleration is maximum in the time-series data of hopper acceleration obtained during a predetermined data acquisition period, in a direction intersecting the hopper's side wall or along the hopper's side wall, as the reference time; and setting the period from before the reference time to after the reference time within the data acquisition period as the evaluation period.

In the above prediction process, the amount of air contained in the fresh concrete may also be predicted.

In the above prediction process, at least one of the slump and slump flow of the ready-mix concrete may be predicted.

The above input information may further include at least one of the following: information indicating the concrete material mix design conditions, and information indicating the conditions or state of the concrete material during mixing in the mixer.

The above quality prediction method may further include a model building step for constructing a prediction model. The model building step may include: preparing learning input information including learning vibration information indicating the magnitude of vibrations caused by the dropping or flow of test ready-mix concrete produced by a mixer; actual performance information indicating the quality of the test ready-mix concrete associated with the learning input information; and constructing a prediction model using machine learning with a neural network based on the learning input information and the actual performance information.

A method for manufacturing ready-mixed concrete relating to one aspect of this disclosure includes a mixing step, an acquisition step, and a prediction step. In the mixing step, concrete materials are mixed in a mixer. In the acquisition step, input information including vibration information indicating the magnitude of vibrations caused by the dropping or flow of the ready-mixed concrete produced in the mixing step is acquired. In the prediction step, the quality of the ready-mixed concrete is predicted based on a prediction model pre-built using machine learning to output quality information indicating the quality of the ready-mixed concrete in response to the input information, and the input information acquired in the acquisition step.

A ready-mix concrete manufacturing apparatus according to one aspect of this disclosure comprises a mixer, a sensor, an information acquisition unit, a model holding unit, and a quality prediction unit. The mixer mixes the concrete materials. The sensor detects the magnitude of vibrations caused by the dropping or flow of the ready-mix concrete produced in the mixer. The information acquisition unit acquires input information, including vibration information indicating the magnitude of vibrations, based at least on the detection results from the sensor. The model holding unit stores a prediction model pre-built by machine learning to output quality information indicating the quality of the ready-mix concrete in response to the input information. The quality prediction unit predicts the quality of the ready-mix concrete based on the prediction model stored by the model holding unit and the input information acquired by the information acquisition unit.

This disclosure provides a method for predicting the quality of ready-mixed concrete, a method for manufacturing ready-mixed concrete, and a system for manufacturing ready-mixed concrete, all of which enable easy assessment of the concrete's quality.

Figure 1 is a schematic diagram showing an example of a ready-mix concrete manufacturing system. Figure 2(a) is a schematic side view showing an example of a loading hopper. Figure 2(b) is a schematic bottom view showing an example of a loading hopper. Figure 3 is a block diagram showing an example of the functional configuration of a control device. Figure 4(a) is a graph showing an example of the values detected by a vibration detection sensor. Figure 4(b) is a graph showing an example of the values detected when fresh concrete falls. Figure 5 is a graph showing an example of a moving average value of the detected values. Figure 6 is a block diagram showing an example of the hardware configuration of a control device. Figure 7 is a flowchart showing an example of a series of processes performed during the learning phase. Figure 8 is a flowchart showing an example of a series of processes performed during the evaluation phase. Figure 9 is a graph showing an example of the verification results of a prediction model using vibration information. Figure 10 is a graph showing an example of the verification results of a prediction model using vibration information. Figure 11 is a graph showing an example of the verification results of a prediction model using vibration information.

An embodiment will be described below with reference to the drawings. In this description, identical elements or elements with the same function will be denoted by the same reference numeral, and redundant descriptions will be omitted.

[Ready-mix concrete manufacturing system]
Figure 1 schematically shows a ready-mix concrete manufacturing system according to one embodiment. The manufacturing system 1 shown in Figure 1 is a system for manufacturing ready-mix concrete. The manufacturing system 1 produces ready-mix concrete by mixing concrete materials. The concrete materials include cement, admixtures, coarse aggregate (e.g., gravel), fine aggregate (e.g., sand), water, and other admixtures.

Manufacturing system 1 loads the manufactured ready-mix concrete onto transport vehicle C. After loading, transport vehicle C transports the ready-mix concrete to the site where it will be used (e.g., a construction site). Examples of transport vehicles C include agitator trucks (mixer trucks) and dump trucks. Manufacturing system 1 may also manufacture ready-mix concrete from concrete materials to meet the target quality (required quality) set for each site. Manufacturing system 1 comprises a manufacturing device 10, a vibration detection sensor 40 (sensor), and a control device 50.

(Equipment for manufacturing ready-mixed concrete)
The manufacturing apparatus 10 is a device that manufactures ready-mixed concrete based on operation instructions from the control device 50. The manufacturing apparatus 10 includes, for example, a storage jar 12, a measuring jar 14, a collection hopper 16, a mixer 20, and a loading hopper 30 (hopper). In Figure 1, the vertical direction is indicated by "direction D1" and "direction D2". Direction D1 is vertically upward, and direction D2 is vertically downward. In the manufacturing apparatus 10, the storage jar 12, measuring jar 14, collection hopper 16, mixer 20, and loading hopper 30 are arranged in this order from top to bottom.

The storage jar 12 temporarily stores various concrete materials. Various concrete materials are transported to the storage jar 12 from aggregate storage areas, cement silos, and water tanks via conveying equipment such as belt conveyors. The storage jar 12 is configured to store each type of concrete material individually. Hereafter, "concrete materials" may be simply referred to as "materials." The various materials stored in the storage jar 12 are supplied to the measuring jar 14 as needed.

The measuring bottle 14 is located below the storage bottle 12. The measuring bottle 14 operates based on operation instructions from the control device 50, individually measuring various materials. When the measuring bottle 14 detects the target amount of material instructed by the control device 50, it supplies that material to the collection hopper 16. When water is supplied to the measuring bottle 14, an admixture may be mixed into the water. The collection hopper 16 is located below the measuring bottle 14. The collection hopper 16 collects the various materials discharged from the measuring bottle 14 and supplies the collected materials to the mixer 20.

The mixer 20 is located below the collective hopper 16. The mixer 20 is a device that produces ready-mixed concrete by mixing aggregate, cement, water, and admixtures. The mixer 20 may be a tilting mixer, a horizontal single-axis mixer, a horizontal double-axis mixer, or a pan-type mixer. The mixer 20 includes, for example, two stirring members 21 and a mixer drive unit 22. In a tilting mixer, it is difficult to capture images inside the mixer; therefore, the vibration-based quality prediction method according to this embodiment can be suitably utilized.

The two agitators 21 are components that agitate the various materials supplied to the mixer 20. The two agitators 21 are arranged side-by-side inside the main body (container) of the mixer 20 and are rotatably mounted. Each of the two agitators 21 includes a rotation axis extending in one horizontal direction. The mixer drive unit 22 rotates the rotation axes of each of the two agitators 21 based on operation instructions from the control device 50. The mixer drive unit 22 includes a drive source, such as a motor, that provides driving force to the agitators 21. The bottom wall of the main body of the mixer 20 is provided with an opening for discharging the manufactured ready-mixed concrete into the loading hopper 30.

The loading hopper 30 is positioned below the mixer 20 and is capable of containing ready-mixed concrete. The loading hopper 30 receives the ready-mixed concrete discharged from the mixer 20 (the aforementioned opening/closing port of the mixer 20) and temporarily stores it. When ready-mixed concrete is discharged from the mixer 20 to the loading hopper 30, it falls into the loading hopper 30. For example, when discharged from the mixer 20, almost all of the ready-mixed concrete produced in the mixer 20 falls into the loading hopper 30. After temporarily storing the ready-mixed concrete produced in the mixer 20, the loading hopper 30 supplies the ready-mixed concrete to the transport vehicle C.

The loading hopper 30 may have any shape as long as it can accommodate ready-mix concrete. The loading hopper 30 may be a truncated cone or a truncated square pyramid. For example, the loading hopper 30 is a truncated square pyramid and, as shown in Figures 2(a) and 2(b), has a bottom wall 32 and side walls 34a, 34b, 34c, and 34d. The bottom wall 32 is horizontally positioned, and in plan view (viewed from vertically above), the outer edge of the bottom wall 32 is square. Each of the side walls 34a, 34b, 34c, and 34d is connected to one edge of the bottom wall 32 and is formed to extend in a vertically oblique direction. Each side wall is inclined vertically, moving away from the bottom wall 32 as it extends upwards.

The bottom wall 32 and side walls 34a, 34b, 34c, and 34d form the storage space within the loading hopper 30. The top of the storage space within the loading hopper 30 is open. Within the storage space within the loading hopper 30, the horizontal area decreases from the top towards the bottom wall 32. The bottom wall 32 is provided with an opening (not shown) for discharging ready-mix concrete to the transport vehicle C. Figure 2(b) illustrates the positional relationship between the loading hopper 30 and the opening 24 provided in the bottom wall of the main body of the mixer 20, when viewed from vertically below.

The opening 24 of the mixer 20 may be formed to extend along one horizontal direction. The opening 24 may be positioned between the pair of stirring members 21 described above and may be formed to extend along the rotation axis of the stirring members 21. The loading hopper 30 may be positioned such that the pair of outer edges of the bottom wall 32 are aligned with the extending direction of the opening 24. Side walls 34a and 34b are positioned to sandwich the bottom wall 32 in the extending direction of the opening 24. Side walls 34c and 34d are positioned to sandwich the bottom wall 32 and the opening 24 in one horizontal direction perpendicular to the extending direction of the opening 24.

In the extending direction of the opening/closing port 24, the length of the opening/closing port 24 may be longer than the bottom wall 32, and may be shorter than the length (shortest distance) between the upper end of the side wall 34a and the upper end of the side wall 34b. In this case, when fresh concrete is discharged from the opening/closing port 24, the fresh concrete may fall not only onto the upper surface of the bottom wall 32, but also onto the inner surfaces of the side walls 34a and 34b. The fresh concrete that falls onto the inner surfaces of the side walls 34a or 34b will flow down along these inner surfaces towards the bottom wall 32.

(Vibration detection sensor)
The vibration detection sensor 40 is a sensor that detects vibrations (magnitude of vibrations) caused by the dropping or flow of ready-mixed concrete produced by the mixer 20. Vibrations caused by dropping or flowing include vibrations caused by both dropping and flowing of the ready-mixed concrete. The vibration detection sensor 40 may be capable of detecting the magnitude of vibrations in two or more directions caused by the dropping or flowing of ready-mixed concrete. The vibration detection sensor 40 detects, for example, the magnitude of vibrations in the loading hopper 30. The vibration detection sensor 40 may be installed on either side wall of the loading hopper 30. The vibration detection sensor 40 may be installed on one of the side walls 34a and 34b from which ready-mixed concrete discharged from the mixer 20 can fall. In the example shown in Figures 2(a) and 2(b), the vibration detection sensor 40 is provided on the outer surface of side wall 34a.

The vibration detection sensor 40 may detect the magnitude of vibration of the loading hopper 30 in a direction intersecting the side wall 34a. In addition to vibrations intersecting the side wall 34a, the vibration detection sensor 40 may also detect the magnitude of vibration of the loading hopper 30 in a direction along the side wall 34a. For example, the vibration detection sensor 40 detects the magnitude of vibration of the loading hopper 30 in a direction perpendicular to the side wall 34a, and the magnitude of vibrations in two directions that are along the side wall 34a and perpendicular to each other. In this disclosure, the direction perpendicular to the side wall 34a is defined as the "Z-axis," and the horizontal and vertical directions when viewing the side wall 34a from the direction perpendicular to the side wall 34a are defined as the "X-axis" and "Y-axis," respectively.

The vibration detection sensor 40 may be of any type as long as it is capable of detecting the magnitude of vibration of the loading hopper 30. For example, the vibration detection sensor 40 is a sensor that detects the acceleration of the loading hopper 30. The vibration detection sensor 40 may detect the acceleration of the loading hopper 30 in the X-axis direction, the Y-axis direction, and the Z-axis direction.

The vibration detection sensor 40 may detect the acceleration of each axis at a predetermined sampling period. The sampling period may be approximately 0.1 seconds to 1.0 second. Instead of acceleration, the vibration detection sensor 40 may detect the speed, angular velocity, position information, displacement, or both acceleration and angular velocity of the loading hopper 30. The sensor detecting position information may be a GPS sensor. The vibration detection sensor 40 may continuously detect physical quantities indicating the magnitude of vibration, such as acceleration, while the manufacturing system 1 (manufacturing equipment 10) is in operation. The vibration detection sensor 40 outputs the detected values, such as acceleration, to the control device 50.

(Control device)
The control device 50 is a device that controls the manufacturing apparatus 10. The control device 50 is composed of one or more control computers. An input/output device 52 may be connected to the control device 50 (see Figure 3). The input/output device 52 is a device that inputs information indicating instructions from an operator, etc., to the control device 50 and outputs information from the control device 50 to the operator, etc. The input/output device 52 may include a keyboard, an operation panel, or a mouse as an input device, and may include a monitor (e.g., a liquid crystal display) as an output device. The input/output device 52 may be a touch panel in which the input and output devices are integrated. The control device 50 and the input/output device 52 may be integrated.

The control device 50 may control the manufacturing apparatus 10 according to predetermined operating conditions. At least a portion of the operating information may be determined by instructions from an operator or the like. In addition to controlling the manufacturing apparatus 10, the control device 50 may be configured to predict the quality of the ready-mixed concrete produced by the manufacturing apparatus 10. In this case, the control device 50 constitutes a quality prediction device that predicts the quality of the ready-mixed concrete. The control device 50 (quality prediction device) is configured to perform at least the following: acquire input information including vibration information indicating the magnitude of vibrations caused by the dropping or flow of ready-mixed concrete produced in the mixer 20; predict the quality of the ready-mixed concrete based on a prediction model pre-built by machine learning to output quality information indicating the quality of the ready-mixed concrete in response to the input information; and predict the quality of the ready-mixed concrete based on the acquired input information.

As shown in Figure 3, the control device 50 has the following functional components (hereinafter referred to as "functional modules"): an operating condition holding unit 62, an operating control unit 64, an operating data acquisition unit 66, a detection data acquisition unit 68, a vibration data calculation unit 70, a quality prediction unit 72, a model holding unit 74, a model construction unit 76, and a notification unit 78. The processing performed by these functional modules corresponds to the processing performed by the control device 50.

The operating condition holding unit 62 holds (stores) information indicating the operating conditions for operating the manufacturing apparatus 10. The operating conditions held by the operating condition holding unit 62 may be predetermined by the operator, etc. The operator, etc., may adjust (change) the operating conditions according to the target quality of the ready-mixed concrete. The operation control unit 64 controls the manufacturing apparatus 10 according to the operating conditions held by the operating condition holding unit 62. The operating conditions may include the concrete material mix conditions, the conditions of the materials used, and the conditions during the mixing of the concrete materials in the mixer 20.

The operation data acquisition unit 66 acquires operation information used to predict the quality of ready-mixed concrete. The operation information includes, for example, at least one of the following: condition information indicating the operation conditions for operating the manufacturing apparatus 10; state information indicating the state of the apparatus or materials when the manufacturing apparatus 10 is actually operating; and environmental information indicating the environment during the operation of the manufacturing apparatus 10. The operation information may also include at least one of the following: information indicating the concrete material mix conditions; and information indicating the conditions or state during the mixing of the concrete material in the mixer 20.

The operational information may include at least one of the following: information regarding the mixing conditions of various materials, information regarding the materials used, information regarding mixing in the mixer 20, and information regarding the environment (external environment). Information regarding the mixing conditions of various materials may include, for example, at least one of the following: nominal strength, mix strength, target slump, target slump flow, target air content, water-binder ratio, fine aggregate ratio, and the unit amount [kg/ ^m³ ] or unit volume [L/ ^m³ ] of various materials (water, cement, admixtures, fine aggregate, coarse aggregate, and admixtures, etc.).

Information regarding the materials used includes, for example, the type, brand, manufacturing date, and density [g/ ^cm³ ] of each material, as well as at least one of the following types of information regarding aggregates (fine aggregates and coarse aggregates): water absorption rate, moisture content, surface moisture content, actual volume, coarseness ratio, particle size distribution, and maximum dimensions. The conditions or state during the mixing of concrete materials in the mixer 20 include, for example, at least one of the following types of information: the amount of each material mixed (total amount), the set value of the mixing time, the measured value of the mixing time, and the power load value of the mixer 20.

The operational information may include, in place of or in addition to the information described above, at least one of the following: information regarding cement and information regarding cement clinker. Information regarding cement may include, for example, at least one of the following: cement type, chemical composition, mineral composition, wet formula f. CaO, ignition loss, Blaine specific surface area, particle size distribution, sieve test residue, color, mineralogical and crystallographic properties of each mineral contained in the cement, and the hemihydrated ratio of gypsum contained in the cement.

Information regarding cement clinker includes, for example, the chemical composition, hydraulic coefficient, sieve test residue, Blaine specific surface area (powderness), and ignition loss of the cement clinker's raw materials; the mineral composition, chemical composition, wet formula f. CaO (free lime), and volumetric gravity of the cement clinker; the crystallographic properties (lattice constant or crystallite size, etc.) of each mineral contained in the cement clinker; and the ratio of two or more mineral compositions contained in the cement clinker, with at least one type of information selected from this group.

The power load value of the mixer 20 may be measured in the mixer drive unit 22. The information indicating the power load value of the mixer 20 may be continuous time-series data during the operation of the mixer 20, statistical values obtained from the time-series data, or power load values at a representative number of points. The information indicating the power load value may be the initial power load value at the start of operation of the mixer 20 (immediately after start), the maximum power load value during the operation of the mixer 20, and the final power load value at the end of operation of the mixer 20 (immediately before end). The difference between the initial and maximum power load values of the mixer 20, and the final power load value of the mixer 20, are thought to be influenced by factors such as the hardness (softness) of the ready-mixed concrete.

Information regarding the environment (external environment) includes, for example, at least one of the following: ambient temperature, humidity, temperature inside the mixer 20, temperature of various materials, and temperature of the containers storing the various materials. From the viewpoint of predicting quality accuracy, one or more pieces of information selected from the group consisting of the density of various materials, set and measured surface moisture content of fine aggregate, nominal strength, mix strength, water-binder ratio, fine aggregate ratio, unit quantity of each material, mixing amount of materials, mixing time, initial, maximum, and ultimate values of power load, and ambient temperature may be used as operational information. The various operational information described above are merely examples; any information that can affect the quality of the ready-mixed concrete may be included.

The detection data acquisition unit 68 acquires detection information indicating the vibration detection result from the vibration detection sensor 40. For example, the detection data acquisition unit 68 acquires detection information from the vibration detection sensor 40 indicating the acceleration of the loading hopper 30 in the X, Y, and Z axes. The detection data acquisition unit 68 may continuously acquire detection information indicating acceleration from the vibration detection sensor 40 while the manufacturing apparatus 10 is in operation.

The vibration data calculation unit 70 calculates (acquires) vibration information indicating the magnitude of vibrations caused by the dropping or flow of ready-mixed concrete produced by the mixer 20, based on the detection information acquired by the detection data acquisition unit 68. For example, the vibration data calculation unit 70 calculates vibration information indicating the magnitude of vibrations in two or more directions caused by the dropping or flow of ready-mixed concrete, based on the above detection information. The vibration data calculation unit 70 may also calculate vibration information indicating the magnitude of vibrations in the loading hopper 30 when ready-mixed concrete is discharged from the mixer 20 to the loading hopper 30.

The vibration information calculated by the vibration data calculation unit 70 may include information indicating the magnitude of vibration of the loading hopper 30 in the Z-axis direction. In addition to the Z-axis direction, the vibration information calculated by the vibration data calculation unit 70 may also include information indicating the magnitude of vibration of the loading hopper 30 in at least one of the X-axis and Y-axis directions. If the vibration detection sensor 40 is an acceleration sensor, the vibration data calculation unit 70 may acquire information indicating the acceleration of the loading hopper 30 in the X-axis, Y-axis, and Z-axis directions from the detection information acquired by the detection data acquisition unit 68. The vibration data calculation unit 70 may calculate statistical values (representative values) as the above vibration information from the acceleration detection information detected by the vibration detection sensor 40 in each of the X-axis, Y-axis, and Z-axis directions.

The following describes an example of calculating statistical values as vibration information from acceleration detection data. In the example below, the control device 50 (vibration data calculation unit 70) calculates vibration information in a situation where it does not know the timing of the discharge of ready-mixed concrete from the mixer 20 to the loading hopper 30. A unit of ready-mixed concrete produced in one mixing cycle in the mixer 20 and loaded onto the transport vehicle C via the loading hopper 30 is defined as "1 batch." In one example, 1 to 3 batches of ready-mixed concrete are loaded onto one transport vehicle C. For example, if 2 batches of ready-mixed concrete are loaded onto one transport vehicle C, the mixing of the concrete material is performed twice in the mixer 20 at different timings (sequentially).

The vibration data calculation unit 70 first acquires data showing the time change of acceleration obtained during a predetermined data acquisition period in the X, Y, and Z axis directions (hereinafter referred to as "time-series data") from the detection information obtained by the detection data acquisition unit 68. The predetermined data acquisition period is set in advance to include the timing of one batch of ready-mixed concrete discharged from the mixer 20. In one example, the predetermined data acquisition period is set to the period from the start of production of one batch of ready-mixed concrete until the completion of loading onto the transport vehicle C.

Figure 4(a) shows a graph illustrating an example of the time change in acceleration in the Z-axis direction when the manufacturing apparatus 10 continues to operate and produces and loads several dozen batches of ready-mixed concrete. In the graph of Figure 4(a), an example of the data acquisition period is indicated by "Ta". As ready-mixed concrete is discharged from the mixer 20 to the loading hopper 30, the concrete falls into the loading hopper 30, causing a large fluctuation in the acceleration in the Z-axis direction in the negative direction (towards the outside of the loading hopper 30). The vibration data calculation unit 70 calculates the time at which the acceleration in the Z-axis direction shows its minimum value (the absolute value of the acceleration in the Z-axis direction is maximum) in the time-series data of acceleration obtained during the data acquisition period as the reference time ts.

Next, the vibration data calculation unit 70 sets the evaluation period to the time before the reference time ts and the time after the reference time ts within the data acquisition period. The evaluation period is the period for calculating the statistical values of acceleration. The vibration data calculation unit 70 calculates the statistical values of acceleration based on the detected acceleration values during the evaluation period. The evaluation period is set to include the time when the ready-mixed concrete is discharged from the mixer 20 to the loading hopper 30 (the time of dropping). The method for setting the evaluation period is predetermined.

The vibration data calculation unit 70 extracts data for the evaluation period from the time-series data during the data acquisition period. Figure 4(b) shows a graph representing the time change of acceleration in the Z-axis direction during the evaluation period. In the example shown in Figure 4(b), the disclosure point of the evaluation period is set to 5 seconds before the reference time ts, and the end point of the evaluation period is set to 5 seconds after the reference time ts. 5 seconds is just an example; the evaluation period may be set to a period of several seconds to several tens of seconds around the reference time ts. The vibration data calculation unit 70 may also set the evaluation period such that the time between the reference time ts and the disclosure point is different from the time between the reference time ts and the end point.

The vibration data calculation unit 70 may calculate moving average values for the acceleration of the loading hopper 30 in the X, Y, and Z axes during the evaluation period. The interval and number of data points used to calculate one value of the moving average can be set in any way. The vibration data calculation unit 70 may also calculate the average of the acceleration in the current sampling period, the acceleration in the previous period, and the acceleration in the period two periods prior, for each sampling period. Depending on the sampling period, the vibration data calculation unit 70 may use detected acceleration values from outside the evaluation period to calculate the moving average value for the evaluation period.

Figure 5 shows a graph representing the time change of the moving average value during the evaluation period. The vibration data calculation unit 70 may obtain as vibration information the result of calculating at least one selected from the group consisting of a maximum value a1, a minimum value a2, a mean value μ, and a value based on the standard deviation σ for the moving average value of the acceleration of the loading hopper 30 during the evaluation period in each of the X, Y, and Z axis directions. The value based on the standard deviation σ includes the standard deviation σ itself and a value obtained by performing a predetermined operation on the standard deviation σ (for example, variance ^σ² , 3×σ). The vibration data calculation unit 70 may calculate as vibration information the maximum value a1, a minimum value a2, a mean value μ, and variance ^σ² of the moving average value of acceleration during the evaluation period in each of the three directions, as shown in Figure 5. In this case, the vibration information obtained in the production of one batch of ready-mixed concrete consists of 12 data points (3 axes × 4 types of data).

Returning to Figure 3, the quality prediction unit 72 predicts the quality of the ready-mixed concrete based on the input information, including the vibration information calculated by the vibration data calculation unit 70. The input information is the information used by the quality prediction unit 72 to predict the quality of the ready-mixed concrete. The quality prediction unit 72 may also acquire information including the vibration information calculated by the vibration data calculation unit 70 and the operation information acquired by the operation data acquisition unit 66 as the input information. In this case, the operation data acquisition unit 66 and the vibration data calculation unit 70 constitute an information acquisition unit that acquires input information including vibration information and operation information.

The quality of the ready-mixed concrete predicted by the quality prediction unit 72 includes, for example, at least one of the following: slump, slump flow, time to reach 50 cm flow, V-funnel flow time, air content, setting start time, setting end time, strength (nominal strength, compressive strength, and flexural strength, etc.), Young's modulus, chloride content, durability, and color tone. The quality of the ready-mixed concrete may also include the quality of the hardened body formed when the ready-mixed concrete hardens.

The quality prediction unit 72 may predict the amount of air contained in the ready-mixed concrete as a quality indicator of the ready-mixed concrete. Alternatively, the quality prediction unit 72 may predict at least one of the slump (slump value) and slump flow (slump flow value) of the ready-mixed concrete as a quality indicator of the ready-mixed concrete, either in addition to or instead of the air content. The strength, slump, slump flow, air content, and chloride content of the ready-mixed concrete can be measured by the test methods specified in "JIS A 5308:2019 (Ready-Mixed Concrete)".

The quality prediction unit 72 predicts the quality of the ready-mixed concrete based on a prediction model pre-built using machine learning to output quality information indicating the quality of the ready-mixed concrete in response to input information including vibration information, and the input information acquired by the information acquisition unit. The input information input to the prediction model may consist of multiple types of input data (numerical values), and the quality information output from the prediction model may consist of at least one type of quality data (numerical values). The quality prediction unit 72 may also select a model corresponding to the quality type to be predicted from multiple prediction models pre-built for multiple quality types, and then predict the quality of the ready-mixed concrete using the prediction model for each quality type.

Machine learning is a method in which a machine (computer) autonomously discovers laws or rules by iteratively learning based on given information. The predictive model described above can be constructed using algorithms and data structures. For example, a predictive model can be implemented using a neural network, which is an information processing model that mimics the structure of the human brain. The specific machine learning algorithm used to construct the predictive model is not particularly limited.

The model storage unit 74 stores the prediction model. The model construction unit 76 constructs the prediction model by performing machine learning based on the input information for learning to construct the prediction model. The model construction unit 76 may autonomously construct a prediction model to predict quality by performing machine learning using, for example, the data provided as input to machine learning and the ground truth data (measured quality values) from the output of machine learning. The input to machine learning is various datasets of input information. At least one piece of information contained in the input information differs between the various datasets of input information. The output of machine learning is data (numerical values) indicating the quality of ready-mixed concrete. The model construction unit 76 iteratively learns a prediction model that outputs quality information indicating the quality of ready-mixed concrete using multiple combinations of the input data datasets and the ground truth quality data.

When the model building unit 76 constructs a prediction model, it prepares learning input information, which includes learning vibration information indicating the magnitude of vibrations caused by the dropping or flow of test ready-mixed concrete produced by the mixer 20, and actual performance information indicating the quality of the test ready-mixed concrete, which is associated with the learning input information. When the prediction model is constructed, for example, multiple combinations (multiple datasets) of learning input information and actual performance information are input to the control device 50. The model building unit 76 may construct the prediction model using machine learning with a neural network based on the learning input information and actual performance information.

The stage of autonomously generating the above prediction model corresponds to the learning phase. During the production phase or evaluation phase, the quality prediction unit 72 uses the prediction model stored in the model storage unit 74 to predict the quality of the ready-mixed concrete produced when the input information was obtained, based on input information where the quality is unknown. The above learning phase may be performed in the early stages of the production phase, where ready-mixed concrete is produced. The learned prediction model can be ported between computers. Therefore, the control device 50 may acquire a prediction model constructed on another computer and store that prediction model in the model storage unit 74. In this case, the control device 50 may execute the production phase (evaluation phase) without executing the learning phase.

The notification unit 78 notifies the operator of the prediction results from the quality prediction unit 72. The notification unit 78 outputs quality information indicating the predicted quality of the ready-mixed concrete to the input/output device 52. The notification unit 78 may also display the above quality information on the monitor of the input/output device 52. After confirming the quality information notified to the input/output device 52, the operator may determine whether the quality information meets the target quality. Alternatively, the control device 50 (for example, the quality prediction unit 72) may determine whether the predicted quality information meets the target quality, and then the notification unit 78 may output information indicating the determination result to the input/output device 52.

As shown in Figure 6, the control device 50 has a circuit 91. The circuit 91 includes one or more processors 92, a memory 93, a storage device 94, an input/output port 95, and a timer 96. The storage device 94 has a computer-readable storage medium, such as a non-volatile semiconductor memory. The storage device 94 stores a program that causes the control device 50 to execute a control procedure to control the manufacturing apparatus 10, the vibration detection sensor 40, and the input/output device 52. For example, the storage device 94 stores a program for configuring each of the above-mentioned functional modules.

Memory 93 temporarily stores the program loaded from the storage medium of storage 94 and the calculation results from processor 92. Processor 92, in cooperation with memory 93, executes the program to constitute each functional module of the control device 50. Input/output ports 95 perform input and output of electrical signals to the manufacturing equipment 10, vibration detection sensor 40, and input/output devices 52, etc., according to commands from processor 92. Timer 96 measures elapsed time, for example, by counting reference pulses of a fixed period. Note that the circuit 91 is not necessarily limited to those whose functions are constituted by a program. For example, circuit 91 may consist of at least some functions composed of dedicated logic circuits or an ASIC (Application Specific Integrated Circuit) that integrates them.

[Method of manufacturing ready-mixed concrete]
Next, an example of a ready-mix concrete manufacturing method performed in manufacturing system 1 will be described. The ready-mix concrete manufacturing method includes a process for manufacturing ready-mix concrete and a process for predicting the quality of the ready-mix concrete. The process for manufacturing ready-mix concrete (hereinafter referred to as the "manufacturing process") includes, for example, a transport process, a weighing process, a input process, a mixing process, a discharge process, and a loading process. In the manufacturing process, these processes may be executed sequentially for each batch of ready-mix concrete produced. While the input process, mixing process, discharge process, and loading process are being executed, the transport process and weighing process for the next batch may be executed.

In the conveying process, various concrete materials are transported to storage jars 12 by conveying equipment such as belt conveyors, and each material is supplied to the storage jars 12 individually. In the weighing process, each material is supplied individually from the storage jars 12 to weighing jars 14, where it is weighed. In the weighing process, when the measured amount of each material reaches the set amount for one batch, that material is discharged into the collection hopper 16. In the input process, after all types of materials have been collected in the collection hopper 16, one batch of material from the collection hopper 16 is input (supplied) into the mixer 20.

In the mixing process, the concrete materials are mixed in the mixer 20. During the mixing process, the control device 50 may control the mixer drive unit 22 to drive the two agitators 21 according to predetermined operating conditions. In the discharge process, after the mixing of the concrete materials in the mixer 20 is completed, the freshly made concrete is discharged from the mixer 20 to the loading hopper 30. For example, the control device 50 determines that mixing in the mixer 20 is complete when it determines that the power load value of the mixer 20 has stabilized (in one example, when the time change of the power load value falls below a predetermined value). In the loading process, after one batch of freshly made concrete is discharged from the mixer 20 to the loading hopper 30, the freshly made concrete is loaded from the loading hopper 30 into the transport vehicle C.

The process for predicting the quality of ready-mixed concrete (hereinafter referred to as the "quality prediction process") is performed during a period that overlaps with at least a portion of the execution period of the manufacturing process described above. The quality prediction process includes a model building process in the learning phase and a quality evaluation process in the production or evaluation phase. An example of the model building process and an example of the quality evaluation process are described below. Note that the explanation assumes that one quality evaluation is performed each time two batches of ready-mixed concrete are produced.

(Model building process)
Figure 7 is a flowchart showing an example of a series of processes performed in the model building process. This model building process is performed in parallel with the repeated production of test ready-mixed concrete in the manufacturing apparatus 10. The test ready-mixed concrete may be ready-mixed concrete that will not actually be shipped, or it may be ready-mixed concrete that is manufactured in the early stages of the production phase and will actually be shipped. While the model building process is being executed, the detection data acquisition unit 68 of the control device 50 continues to detect acceleration values from the vibration detection sensor 40 at predetermined sampling intervals.

In the model building process, the control device 50 first executes step S11. In step S11, for example, the control device 50 waits until a predetermined evaluation timing is reached. The predetermined evaluation timing may be set to the timing when two batches of ready-mixed concrete, manufactured to the same target quality, are loaded onto the transport vehicle C. In this case, data evaluation and collection for model building are performed for each production of two batches of test ready-mixed concrete.

Next, the control device 50 executes step S12. In step S12, for example, the vibration data calculation unit 70 acquires data (the above-mentioned time-series data) showing the time change of acceleration obtained during a predetermined data acquisition period in each of the X, Y, and Z axis directions, from the detection information obtained by the detection data acquisition unit 68, batch by batch. That is, the vibration data calculation unit 70 extracts the time-series data for the data acquisition period corresponding to the first batch and the time-series data for the data acquisition period corresponding to the second batch from the above-mentioned detection information.

Next, the control device 50 executes step S13. In step S13, for example, the vibration data calculation unit 70 calculates the vibration information from the time-series data for each batch. In one example, the vibration data calculation unit 70 calculates the time at which the acceleration in the Z-axis direction shows the minimum value in the time-series acceleration data obtained during the data acquisition period as the reference time ts. Then, the vibration data calculation unit 70 sets the period from before the reference time ts to after the reference time ts within the data acquisition period as the evaluation period. The evaluation period is set to be shorter than the data acquisition period.

After setting the evaluation period, the vibration data calculation unit 70 calculates moving average values for the acceleration of the loading hopper 30 during the evaluation period in the X, Y, and Z axis directions. The vibration data calculation unit 70 then calculates the maximum value a1, minimum value a2, mean value μ, and variance ^σ² as vibration information for the moving average values of the acceleration of the loading hopper 30 during the evaluation period in the X, Y, and Z axis directions. For each batch, the vibration data calculation unit 70 performs the calculation of the reference time ts, setting the evaluation period, calculating the moving average values, and calculating the maximum value a1, etc. In this case, the vibration information obtained in the production of two batches of ready-mixed concrete consists of 24 data points (3 axes × 4 types × 2 batches). The information obtained in step S13 corresponds to the vibration information for learning.

Next, the control device 50 executes step S14. In step S14, for example, the operation data acquisition unit 66 acquires the operation information described above. In one example, the operation data acquisition unit 66 acquires information regarding the blending conditions of various materials, information regarding the materials used, information regarding mixing in the mixer 20, and information regarding the environment (external environment) as the operation information. The operation data acquisition unit 66 may acquire information for two batches regarding mixing (mixing amount of materials, mixing time, and power load value of the mixer 20). The information obtained in step S14 corresponds to the operation information for learning.

Next, the control device 50 executes step S15. In step S15, for example, the model construction unit 76 acquires measured values of the ready-mixed concrete quality as ground truth data. In one example, after two batches of ready-mixed concrete are loaded onto the transport vehicle C, some of the concrete is extracted by workers. The slump, slump flow, and air content of the extracted concrete are then measured by the workers, and these measured values are input to the control device 50.

The measured quality values of the ready-mixed concrete obtained in step S15 correspond to actual performance data. The model construction unit 76 stores the learning input information, including the learning vibration information obtained in step S13 and the learning motion information obtained in step S14, after associating it with the actual performance data (quality measurement values) obtained in step S15.

Next, the control device 50 executes step S16. In step S16, for example, the control device 50 determines whether a predetermined number of data sets have been acquired. If it is determined in step S16 that a predetermined number of data sets have not been acquired (step S16: NO), the control device 50 returns to step S11, and the control device 50 repeats steps S11 to S16. This process of repeating steps S11 to S16 continues until the number of data sets, which are associated with learning input information and performance information, reaches a predetermined number.

If it is determined in step S16 that a predetermined number of datasets have been acquired (step S16: YES), the control device 50 proceeds to step S17. In step S17, for example, the model building unit 76 constructs a prediction model (e.g., a regression prediction model) using machine learning with a neural network, based on multiple combinations of input information for learning and performance information. The neural network has at least an input layer and an output layer. The neural network may also include one or more intermediate layers. Including intermediate layers allows for the construction of a more complex prediction model, improving the accuracy of quality predictions. The model building unit 76 may construct a prediction model for each type of quality. The model storage unit 74 stores the prediction model constructed in step S17.

(Quality evaluation process)
Figure 8 is a flowchart showing an example of a series of processes performed in the quality evaluation process. The quality evaluation process is performed in parallel with the production of ready-mixed concrete to be actually shipped, which is repeatedly carried out in the manufacturing apparatus 10, after the model construction process described above. While the production of ready-mixed concrete to be actually shipped continues, the detection data acquisition unit 68 of the control device 50 may continue to detect acceleration values from the vibration detection sensor 40 at predetermined sampling cycles.

In the quality evaluation process, the control device 50 executes steps S21, S22, S23, and S24, similar to steps S11, S12, S13, and S14. In step S23, evaluation vibration information is calculated, and in step S24, evaluation operation information is acquired. That is, by executing steps S23 and S24, evaluation input information, including evaluation vibration information and evaluation operation information, is acquired. "For evaluation" means that the quality of the ready-mixed concrete is unknown.

Next, the control device 50 executes step S25. In step S25, for example, the quality prediction unit 72 inputs the evaluation input information obtained in steps S23 and S24 into the prediction model held by the model holding unit 74. The quality prediction unit 72 then acquires the quality information output from the prediction model, thereby predicting the quality of the ready-mixed concrete to be evaluated. Unlike step S15 in the model construction process, the quality of the ready-mixed concrete is not measured in step S25. The quality prediction unit 72 may predict the quality for each quality type using a prediction model appropriate to that type.

The quality prediction unit 72 predicts the slump of the ready-mixed concrete to be evaluated using the input information for evaluation and a prediction model for slump. The quality prediction unit 72 predicts the slump flow of the ready-mixed concrete to be evaluated using the input information for evaluation and a prediction model for slump flow. The quality prediction unit 72 predicts the air content of the ready-mixed concrete to be evaluated using the input information for evaluation and a prediction model for air content.

Next, the control device 50 executes step S26. In step S26, for example, the notification unit 78 outputs the quality prediction result obtained in step S25 to the input/output device 52. This completes one quality evaluation process performed every two batches. The control device 50 may execute the series of processes from steps S21 to S26 each time two batches of ready-mixed concrete are manufactured (loaded).

[Variations]
The series of processes shown in Figures 7 and 8 are examples and can be modified as appropriate. In the above series of processes, the control device 50 may execute one step and the next step in parallel, or it may execute each step in a different order than the example described above. The control device 50 may also execute steps with different content than the example described above.

Unlike the example above, if one batch of ready-mixed concrete is loaded onto one transport vehicle C, the control device 50 may perform quality prediction for each batch of ready-mixed concrete produced. Furthermore, even if two or more batches of ready-mixed concrete are loaded onto transport vehicle C, the control device 50 may perform quality prediction for each batch produced (loaded). Regardless of the amount loaded onto transport vehicle C, the control device 50 may perform quality prediction for each batch of ready-mixed concrete produced. If ready-mixed concrete with different compositions or physical properties is produced, quality prediction may be performed using data from a larger number of batches to improve prediction accuracy.

In the example above, the point in time when the absolute value of acceleration in the Z-axis is maximum is calculated as the reference time ts, but the method for calculating the reference time ts is not limited to this. As ready-mixed concrete is discharged from the mixer 20 to the loading hopper 30, the concrete falls into the loading hopper 30, causing large fluctuations in acceleration in the X and Y axes. The vibration data calculation unit 70 may calculate the time when the absolute value of acceleration in the X-axis or Y-axis direction is maximum in the time-series acceleration data obtained during the data acquisition period as the reference time ts.

In the example above, the evaluation period is set based on the reference time ts, but the method of setting the evaluation period is not limited to this. The timing of the opening of the mixer 20's discharge port and the timing of the significant vibration of the loading hopper 30 approximately coincide. Therefore, if the control device 50 can detect the timing of the discharge of ready-mixed concrete from the mixer 20, the vibration data calculation unit 70 may set the evaluation period to include the period before and after the detected discharge timing. The control device 50 may also set the evaluation period to include the period before and after the timing of sending a command to open the mixer 20's discharge port, or the timing of receiving a signal indicating that the mixer 20's discharge port has opened. If a sensor for detecting the presence of ready-mixed concrete is provided in the loading hopper 30, the control device 50 may set the evaluation period to include the period before and after the timing when the sensor detects the presence of ready-mixed concrete.

The vibration data calculation unit 70 may calculate at least one of the median, mode, and peak count as vibration information, instead of or in addition to the values based on the maximum, minimum, mean, and standard deviation of the moving average value of acceleration during the evaluation period. Alternatively, the vibration data calculation unit 70 may calculate the statistical value of the acceleration itself during the evaluation period as vibration information, instead of the moving average value of acceleration. The vibration data calculation unit 70 may also calculate vibration information as the result of calculating at least one of the values based on the maximum, minimum, mean, and standard deviation of acceleration during the evaluation period. The input information may include vibration information but not motion information. In this case, a prediction model is constructed from the vibration information, and quality is predicted based on the vibration information and the prediction model.

The vibration data calculation unit 70 may calculate vibration information based on the frequency spectrum obtained by frequency transformation (e.g., Fast Fourier Transform) of the data acquired by the vibration detection sensor 40. The vibration data calculation unit 70 may also calculate vibration information using frequency information, feature quantities, or statistics obtained from the frequency spectrum. The control device 50 may acquire image data of a graph (waveform) showing the measured values acquired by the vibration detection sensor 40 as vibration information.

A separate computer (quality prediction device) from the control device 50 for controlling the manufacturing apparatus 10 may predict the quality of the ready-mixed concrete using a prediction model. In this case, instead of the control device 50, the separate computer may have a functional module that performs quality prediction processing, such as a quality prediction unit 72. The vibration information used for quality prediction may be information indicating vibrations caused by the dropping or flow of ready-mixed concrete in locations other than the loading hopper 30.

[Verification of prediction models using vibration information]
To verify quality prediction using vibration information, we compared the prediction results of a comparative model without vibration information with the prediction results of the above prediction model with vibration information. Specifically, we prepared 83 to 100 training datasets and 10 validation datasets. The training datasets were prepared by associating training input information (training vibration information and motion information) with measured values of slump, slump flow, and air volume.

As operational information among the input data, data was prepared regarding the density of various materials, the set and measured surface moisture content of fine aggregate, nominal strength, mix strength, water-binder ratio, fine aggregate ratio, unit quantity of each material, mixing amount of materials for two batches, mixing time for two batches, initial, maximum, and ultimate values of power load for two batches, and ambient temperature. As vibration information among the input data, data was prepared regarding the maximum value a1, minimum value a2, mean value μ, and variance ^σ² of the moving average values of acceleration for two batches for the X, Y, and Z axes, respectively.

In the training dataset, the above comparison model was constructed using motion information and quality measurements, without using vibration information. In the training dataset, the above prediction model was constructed using vibration information, motion information, and quality measurements. Comparison and prediction models were constructed for slump, slump flow, and air volume. In other words, the following six models were created.
Comparative Model 1: A model that predicts slump without using vibration information Prediction Model 1: A model that predicts slump using vibration information Comparative Model 2: A model that predicts slump flow without using vibration information Prediction Model 2: A model that predicts slump flow using vibration information Comparative Model 3: A model that predicts air volume without using vibration information Prediction Model 3: A model that predicts air volume using vibration information

A first-order regression prediction model was constructed as both a comparison model and a prediction model, using various input data as variables to output a single quality (metric value). The first-order regression prediction model is shown by equation (1) below.

In equation (1), "y" is the output representing the quality index value, and represents slump, slump flow, or air volume. "i" is a natural number between 2 and N, and "N" represents the number of input data points. "w" is the weight (coefficient), and "b" is the bias term (coefficient). Using the training dataset, a regression prediction model was constructed for each quality type by determining the weight w and bias term b in equation (1). The value of weight w was determined by calculating the squared error between the output y value when a certain value is input to weight w and the measured quality value, which is the ground truth data, and updating the value of weight w using mini-batch gradient descent.

Note that the regression prediction model shown in equation (1) is a model constructed for validation purposes, and the prediction models in this disclosure are not limited to the regression prediction model shown above. When a prediction model (regression prediction model) is constructed using a neural network including an intermediate layer, the regression prediction model is shown, for example, by equation (2) below. In equation (2), f() represents a predetermined function.

After constructing the regression prediction model using equation (1) above, we compared the predicted values with the measured values using 10 validation datasets for both the comparison model without vibration information and the prediction model with vibration information. We compared the predicted and measured values for slump, slump flow, and air volume.

Figure 9 is a graph showing the relationship between predicted and measured slump values. In the graph shown in Figure 9, the horizontal axis represents the measured slump value, and the vertical axis represents the predicted slump value. Prediction results from comparison model 1 (without vibration information) are plotted as triangles (△), and prediction results from prediction model 1 (with vibration information) are plotted as circles (〇). A line is drawn where the measured and predicted values coincide; the closer the plotted result is to this line, the higher the prediction accuracy. Regarding the slump, the arithmetic mean of the absolute difference between the measured and predicted values (hereinafter referred to as "average absolute difference") was calculated. The average absolute difference without vibration information was 1.1 cm, and the average absolute difference with vibration information was also 1.1 cm. This indicates that even when using vibration information, a prediction accuracy of a similar quality to that without vibration information can be obtained.

Figure 10 is a graph showing the relationship between predicted and measured slump flow values. In the graph shown in Figure 10, the horizontal axis represents the measured slump flow value, and the vertical axis represents the predicted slump flow value. Prediction results from comparison model 2 (without vibration information) are plotted with triangles (△), and prediction results from prediction model 2 (with vibration information) are plotted with circles (〇). Regarding slump flow, the absolute average difference was calculated to be 4.0 cm without vibration information and 4.2 cm with vibration information. Since slump flow is measured in 0.5 cm increments, it can be seen that even with vibration information, a prediction accuracy of a similar quality to that without vibration information can be obtained.

Figure 11 is a graph showing the relationship between predicted and measured air volume. In the graph shown in Figure 11, the horizontal axis represents the measured air volume, and the vertical axis represents the predicted air volume. Prediction results from comparative model 3 (without vibration information) are shown with triangles (△), and prediction results from prediction model 3 (with vibration information) are plotted with circles (〇). When the absolute average difference for air volume was calculated, the absolute average difference without vibration information was 0.53%, and the absolute average difference with vibration information was 0.27%. The absolute average difference decreased when vibration information was used.

Regarding air volume, the impact of using vibration information was evaluated from another perspective. In the graph shown in Figure 11, an upper limit line shifted vertically by +0.5% from the line where the measured value and predicted value coincide, and a lower limit line shifted vertically by -0.5% are drawn. Predicted results within the region between the upper and lower limits were defined as correct, and the accuracy rate of 10 prediction results was evaluated. The allowable error for air volume according to JIS standards is ±1.5%, and considering that quality may change during transportation to the site, the correct line was set at ±0.5%. The accuracy rate without vibration information was 60%, while the accuracy rate with vibration information was 80%. From the above, it can be seen that the prediction accuracy of air volume is improved when vibration information is used compared to when it is not used.

[Effects of the Embodiment]
The ready-mixed concrete quality prediction method described above includes an acquisition step and a prediction step. In the acquisition step, input information is acquired that includes vibration information indicating the magnitude of vibrations caused by the dropping or flow of ready-mixed concrete produced by the mixer 20 that mixes the concrete materials. In the prediction step, the quality of the ready-mixed concrete is predicted based on a prediction model pre-built by machine learning to output quality information indicating the quality of the ready-mixed concrete in response to the input information, and the input information acquired in the acquisition step. In this quality prediction method, information including vibration information is acquired and the quality is predicted using the prediction model without actually measuring the quality of the ready-mixed concrete produced. Therefore, it is possible to easily grasp the quality of ready-mixed concrete.

In the quality prediction method described above, the vibration information may include information indicating vibrations in two or more directions. Depending on the quality of the ready-mixed concrete, the state of vibration caused by the dropping or flow of the concrete may vary. In this method, since quality is predicted based on vibration information indicating the magnitude of vibrations in two or more directions, it is possible to improve the prediction accuracy.

The vibration information described above may also indicate the magnitude of vibration in a hopper, which is located below the mixer and capable of holding ready-mix concrete, when the ready-mix concrete is discharged from the mixer. In this case, vibration information can be easily acquired by installing a vibration-detecting sensor in the hopper. Therefore, it becomes possible to understand the quality of the ready-mix concrete even more easily.

The above vibration information may include information indicating the magnitude of hopper vibration in the direction intersecting the hopper's side wall. At the side wall of a hopper located below the mixer, vibration in the direction intersecting the side wall can fluctuate depending on the hardness or weight of the prepared concrete. Therefore, by obtaining vibration information from the hopper vibration in the direction intersecting the side wall, the quality of the prepared concrete can be predicted with high accuracy. Thus, it becomes possible to easily grasp the quality while improving the accuracy of concrete quality prediction.

The above vibration information may further include information indicating the magnitude of hopper vibration in the direction along the hopper's side wall. At the side wall of the hopper located below the mixer, vibrations in the direction along the side wall may also vary depending on the hardness or weight of the prepared concrete. By obtaining vibration information from both vibrations intersecting and along the side wall, it becomes possible to predict quality with high accuracy.

The above vibration information may be the result of calculating at least one value selected from the group consisting of the maximum value, minimum value, mean value, and standard deviation (e.g., variance) of the moving average value of the hopper's acceleration over an evaluation period set to include the time when the ready-mix concrete is discharged from the mixer to the hopper. Calculating the moving average value of acceleration reduces noise caused by the sensor itself, and using the statistical value of the moving average value simplifies data processing. Therefore, it becomes possible to easily grasp quality while improving the accuracy of quality prediction.

The above acquisition process may include the steps of: calculating the time at which the absolute value of the hopper's acceleration is maximum in the time-series data of hopper acceleration obtained during a predetermined data acquisition period, in a direction intersecting or along the hopper's side wall, as the reference time; and setting the period from before the reference time to after the reference time within the data acquisition period as the evaluation period. When fresh concrete falls into the hopper, the absolute value of the hopper's acceleration tends to be larger than during other periods. Therefore, vibration information at the time the fresh concrete falls contains a lot of information related to the quality of the fresh concrete. With the above method, even if the time at which the fresh concrete fell into the hopper is not known, the time at which the fresh concrete fell can be identified from the vibration information. Therefore, it becomes possible to easily grasp the quality while improving the accuracy of quality prediction.

In the above prediction process, the amount of air contained in the fresh concrete may also be predicted. The amount of air in the fresh concrete can affect the vibration caused by the falling or flow of the concrete. This method allows for easy determination of the air content by using vibration information and a prediction model.

In the above prediction process, at least one of the slump and slump flow of the ready-mix concrete may be predicted. In this case, by using input information including vibration information and a prediction model, it becomes possible to easily determine at least one of the slump and slump flow.

The above input information may further include at least one of the following: information indicating the mix design conditions for the concrete material, and information indicating the conditions or state of the concrete material during mixing in the mixer. Mix design conditions and the conditions or state during mixing affect the quality of the ready-mix concrete. Further use of information indicating mix design conditions, etc., makes it possible to improve the accuracy of quality prediction.

The above quality prediction method may further include a model building step for constructing a prediction model. The model building step may include: preparing learning input information, including learning vibration information indicating the magnitude of vibrations caused by the dropping or flow of test ready-mix concrete produced by a mixer; actual performance information indicating the quality of the test ready-mix concrete associated with the learning input information; and constructing a prediction model using machine learning with a neural network based on the learning input information and the actual performance information. In this case, since the model is constructed using machine learning with a neural network based on the actual performance information associated with the input information, the predicted values from the model can be brought closer to the measured values. Therefore, it becomes possible to improve the prediction accuracy when predicting the quality of ready-mix concrete.

1…Ready-mix concrete manufacturing system, 10…Ready-mix concrete manufacturing equipment, 20…Mixer, 30…Loading hopper, 40…Vibration detection sensor, 50…Control device, 66…Operation data acquisition unit, 70…Vibration data calculation unit, 72…Quality prediction unit, 74…Model holding unit, 76…Model construction unit.

Claims (9)

An acquisition step that acquires input information including vibration information indicating the magnitude of vibration caused by the dropping or flow of ready-mixed concrete produced by a mixer for mixing concrete materials , and motion information ,
The system includes a predictive model pre-built by machine learning to output quality information indicating the air content of the ready-mixed concrete in response to the input of the aforementioned input information, and a prediction step that predicts the air content of the ready-mixed concrete based on the input information acquired in the acquisition step .
The vibration information is information indicating the magnitude of vibration of the hopper, which is located below the mixer and capable of containing the ready-mixed concrete, when the ready-mixed concrete is discharged from the mixer.
The vibration information is the result of calculating at least one value selected from the group consisting of the maximum value, minimum value, mean value, and standard deviation of the moving average value of the acceleration of the hopper during an evaluation period set to include the time when the ready-mixed concrete is discharged from the mixer to the hopper.
A method for predicting the quality of ready-mixed concrete, wherein the operation information includes at least one type of information from among condition information indicating the conditions for operating a manufacturing apparatus that produces the ready-mixed concrete using the mixer, state information indicating the state of the apparatus or materials when the manufacturing apparatus is actually in operation, and environmental information indicating the environment when the manufacturing apparatus is in operation . The method for predicting the quality of ready-mixed concrete according to claim 1, wherein the vibration information includes information indicating the magnitude of vibrations in two or more directions. The method for predicting the quality of ready-mixed concrete according to claim 1 or 2 , wherein the vibration information includes information indicating the magnitude of vibration of the hopper in a direction intersecting the side wall of the hopper. The method for predicting the quality of ready-mixed concrete according to claim 3 , wherein the vibration information further includes information indicating the magnitude of vibration of the hopper in a direction along the side wall of the hopper. The acquisition process described above is:
A step of calculating, as a reference time, the time at which the absolute value of the acceleration of the hopper is maximum in the time-series data of the acceleration of the hopper obtained during a predetermined data acquisition period, in a direction intersecting the side wall of the hopper or in a direction along the side wall of the hopper;
A method for predicting the quality of ready-mixed concrete according to any one of claims 1 to 4, comprising the step of setting the period from a time before the reference time to a time after the reference time within the data acquisition period as the evaluation period. The method for predicting the quality of ready-mixed concrete according to any one of claims 1 to 5 , wherein the operation information includes at least one of information indicating the mixing conditions of the concrete material and information indicating the conditions or state when the concrete material is mixed in the mixer. The process further includes a model building step for constructing the aforementioned prediction model,
The aforementioned model building process is:
A step of preparing learning input information including learning vibration information indicating the magnitude of vibration caused by the dropping or flow of test ready-mixed concrete produced by the mixer, and actual information indicating the amount of air in the test ready-mixed concrete associated with the learning input information,
A method for predicting the quality of ready-mixed concrete according to any one of claims 1 to 6 , comprising the step of constructing the prediction model by machine learning using a neural network based on the input information for learning and the performance information. The mixing process involves mixing concrete materials in a mixer,
An acquisition step that acquires input information including vibration information indicating the magnitude of vibration caused by the dropping or flow of the ready-mixed concrete produced in the mixing step , and motion information .
The system includes a predictive model pre-built by machine learning to output quality information indicating the air content of the ready-mixed concrete in response to the input of the aforementioned input information, and a prediction step that predicts the air content of the ready-mixed concrete based on the input information acquired in the acquisition step .
The vibration information is information indicating the magnitude of vibration of the hopper, which is located below the mixer and capable of containing the ready-mixed concrete, when the ready-mixed concrete is discharged from the mixer.
The vibration information is the result of calculating at least one value selected from the group consisting of the maximum value, minimum value, mean value, and standard deviation of the moving average value of the acceleration of the hopper during an evaluation period set to include the time when the ready-mixed concrete is discharged from the mixer to the hopper.
A method for manufacturing ready-mixed concrete, wherein the operation information includes at least one of the following types of information: condition information indicating the conditions for operating a manufacturing apparatus that manufactures ready-mixed concrete using the mixer; state information indicating the state of the apparatus or materials when the manufacturing apparatus is actually in operation; and environmental information indicating the environment when the manufacturing apparatus is in operation. A manufacturing apparatus comprising a mixer for mixing concrete materials, and a hopper positioned below the mixer and capable of containing the ready-mixed concrete produced in the mixer,
A sensor for detecting the magnitude of vibrations caused by the falling or flow of the ready-mixed concrete,
An information acquisition unit that acquires input information including vibration information indicating the magnitude of the vibration and operation information , based at least on the detection results from the sensor,
A model holding unit stores a predictive model that has been pre-built using machine learning to output quality information indicating the air content of the ready-mixed concrete in response to the input of the aforementioned input information.
The system includes a quality prediction unit that predicts the air content of the fresh concrete based on the prediction model stored by the model holding unit and the input information acquired by the information acquisition unit ,
The vibration information is information indicating the magnitude of vibration of the hopper when the ready-mixed concrete is discharged from the mixer to the hopper.
The vibration information is the result of calculating at least one value selected from the group consisting of the maximum value, minimum value, mean value, and standard deviation of the moving average value of the acceleration of the hopper during an evaluation period set to include the time when the ready-mixed concrete is discharged from the mixer to the hopper.
A ready-mix concrete manufacturing system, wherein the operation information includes at least one type of information from among condition information indicating the conditions for operating the manufacturing apparatus, state information indicating the state of the apparatus or materials when the manufacturing apparatus is actually operating, and environmental information indicating the environment when the manufacturing apparatus is operating .