JP7003255B2 - Operation status measurement system and operation status measurement method - Google Patents

Description

The present invention relates to an operating state measurement system and an operating state measuring method, and is suitable for application to, for example, an operating state measuring system and an operating state measuring method for determining the operating state of a target device.

In recent years, we have been working on improving productivity by utilizing IoT (Internet of Things) in domestic and overseas production equipment. In order to grasp the operating rate of the production line, it is necessary to accurately estimate the condition of the manufacturing equipment. For example, there is disclosed a technique for performing failure diagnosis by FFT (Fast Fourier Transform) the frequency of data obtained from an acceleration sensor attached to equipment and performing efficient equipment operation (see Patent Document 1).

Here, the spectrum obtained by performing the FFT calculation includes one or a plurality of peaks, and the frequency and intensity of the peaks differ depending on the product manufactured in the manufacturing equipment. For example, since the SN (signal-noise) ratio differs depending on the vibration mode of the engine knock, it is necessary to change the determination level for each frequency. Further, the frequency with respect to the spectral intensity of the peak in each vibration mode varies depending on the operating conditions and the like. In this regard, it is disclosed to learn the frequency of the peak in each vibration mode in order to detect the engine knock more accurately (see Patent Document 2).

Japanese Unexamined Patent Publication No. 5-72026 Japanese Unexamined Patent Publication No. 6-2607

In the technique described in Patent Document 2 described above, the number of vibration modes is predetermined, and it is difficult to apply it to a system in which the number of vibration modes is not fixed, such as a manufacturing facility for manufacturing a wide variety of products. be.

The present invention has been made in consideration of the above points, and an object of the present invention is to propose an operating state measurement system or the like capable of grasping the operating state of various target devices.

In order to solve this problem, in the present invention, the data processing device that processes the data acquired from the sensor that measures the target device and the data processed by the data processing device are analyzed to determine the operating state of the target device. In an operating state measurement system including an analysis device, a communication means for connecting the data processing device and the analysis device by communication including wireless communication, the data processing device receives data acquired from the sensor. , The frequency division data is converted into the frequency division data divided for each frequency, the frequency division data is transmitted to the analysis device via the communication means, and the analysis device is based on the received frequency division data. The operating state of the data is determined, and the frequency of the frequency-divided data transmitted by the data processing device to the analysis device is determined and notified to the data processing device.

Further, in the present invention, in an operating state measurement system including a computer that processes data acquired from a sensor that measures the target device and determines the operating state of the target device, the computer uses the data acquired from the sensor. Is converted into frequency division data divided for each frequency, and from the frequency division data, the frequency of the operation state determination data for determining the operation state and the frequency of the frequency division data related to the operation state determination data. The learning data for determining the above is selected, the operating state of the target device is determined based on the operating state determination data, and the operation state determination data and the learning data are used as the basis for determining the operating state. The frequency of the frequency division data used as the operating state determination data is changed, and the cycle of referring to the learning data is longer than the cycle of referring to the operating state determination data.

According to the above configuration, since the frequency for determining the operating state is determined according to the target device, it is possible to grasp the operating state of various target devices.

According to the present invention, a highly reliable system can be realized.

It is a figure which shows an example of the configuration which concerns on the operation state measurement system by 1st Embodiment. It is a figure which shows an example of the main body which performs the operation state determination processing by 1st Embodiment. It is a figure for demonstrating the operating state (mode) in the operating state measuring system by 1st Embodiment. It is a figure which shows the image of the series of processing in the measuring apparatus by 1st Embodiment. It is a figure which shows an example of the data transmission method by 1st Embodiment. It is a figure which shows an example of the processing procedure which concerns on the operation state determination process by 1st Embodiment. An example of data for each frequency at a specific time according to the first embodiment is shown. It is a figure which shows an example of the frequency data of the division integral by 1st Embodiment. It is a figure which shows an example of the calculation result of the SN ratio by 1st Embodiment. It is a figure which shows an example of the discrimination frequency by 1st Embodiment. It is a figure which shows an example of the operation state data by 1st Embodiment. It is a figure for demonstrating the setting which concerns on the operation state determination process by 1st Embodiment. It is a figure which shows an example of the data transmission method by 2nd Embodiment. It is a figure which shows an example of the data transmission method by the 3rd Embodiment.

Hereinafter, one embodiment of the present invention will be described in detail with reference to the drawings. In this embodiment, a configuration capable of grasping the operating state of various target devices will be described. Various target devices will be described by exemplifying a case where they are provided in a predetermined facility.

The equipment includes manufacturing equipment for manufacturing goods, maintenance equipment for inspecting and maintaining goods, and the like. Hereinafter, the case where the equipment is a manufacturing equipment will be described as an example, but the present invention is not limited to the manufacturing equipment.

Further, the manufacturing equipment includes an IoT manufacturing equipment and a legacy manufacturing equipment. In the case of IoT equipment, sensors are attached to movable parts, non-movable parts, etc., and control signals for use in analysis can be acquired. In addition, the acquired data can be analyzed collectively for a plurality of facilities or analyzed within the own facility. On the other hand, since the legacy manufacturing equipment was introduced before considering the utilization of IoT, it does not have such a function, and various devices are retrofitted to support IoT. Legacy manufacturing equipment includes, for example, mounter equipment for mounting components on a board.

The legacy manufacturing equipment does not have a built-in sensor for IoT, but a sensor is installed later. The data acquired by the sensor is transmitted to the analysis device via wireless communication for analysis (determination of the operating state, etc.), but generally, there are many manufacturing facilities at the manufacturing site. In addition, even one manufacturing facility has many operating points, and innumerable communication devices perform wireless communication at the same time. Therefore, in the present embodiment, a configuration will also be described in which the status of the manufacturing equipment is reliably transmitted to the analysis device while limiting the communication amount so that the communication band of wireless communication is not tight.

Further, when the signal strength at the frequency used for discriminating the operating state (discrimination frequency) is reduced, the discriminant frequency is relearned, but when applied to a system with a narrow communication band, a learning time of about one week is required. And. Therefore, there is a problem that the discrimination accuracy is lowered during re-learning. Therefore, in the present embodiment, a configuration for avoiding a decrease in discrimination accuracy during re-learning will also be described.

(1) First Embodiment In FIG. 1, 100 indicates an operating state measurement system according to the first embodiment as a whole. The operating state measurement system 100 includes equipment 111 provided in the factory 110, a measuring device 112, a gateway 113, an LTE terminal 114, and a server 121 provided in the data center 120.

The equipment 111 is a manufacturing equipment for manufacturing goods and the like, a maintenance equipment for inspecting and maintaining goods and the like, and the like. Hereinafter, the case where the equipment 111 is a manufacturing equipment will be described as an example, but the equipment 111 is not limited to the manufacturing equipment.

The equipment 111 is, for example, a legacy manufacturing equipment and is configured to include one or more manufacturing equipment 115. Information related to the operation of the manufacturing apparatus 115 is acquired by the acceleration sensor 116. In the present embodiment, the acceleration sensor 116 will be described as an example, but the present invention is not limited to the acceleration sensor 116, and may be a sound sensor. A plurality of acceleration sensors 116 may be provided in one manufacturing apparatus 115.

The acceleration sensor 116 is an acceleration sensor that measures the vibration state of the manufacturing apparatus 115 as an acceleration. As the mounting position, it may be mounted on the movable part of the manufacturing apparatus 115 to measure the vibration, it may be mounted on the non-movable part to measure the vibration, or it may be mounted on each of the movable part and the non-movable part to measure the vibration. May be measured. Further, the acceleration to be measured may be parallel acceleration of 1 to 3 axes, or rotation may be measured using a gyro. The measured data (measurement data) is output to the measuring device 112.

The measuring device 112 is a device that is connected to the acceleration sensor 116 by wire or wirelessly, processes measurement data, and transmits it by wireless communication. The measuring device 112 has a built-in processor, memory, power supply, antenna, and the like (all not shown). The reason for using wireless communication is to omit the communication wiring. The function of the measuring device 112 will be described later.

The gateway 113 includes a processor, a memory, an antenna, and the like (all not shown), receives data transmitted by wireless communication using a communication module mounted on the measuring device 112, and receives data transmitted by LTE (Long Term Evollution). It is transmitted to the server 121 via the terminal 114.

The server 121 is a server device, a computer, or the like provided with a processor, a memory, a storage device, and the like (all of which are not shown). Analysis software is installed in the server 121. The server 121 may be installed in a data center 120 (site) different from the factory 110 (manufacturing site), or may be installed in the factory 110. The server 121 stores the data transmitted from each measuring device 112 in the storage device, analyzes the stored data, and provides information for management such as efficiency improvement of the manufacturing site and maintenance of the equipment 111 (for example, operation described later). Status data) is output.

The functions of devices such as the measuring device 112, the gateway 113, and the server 121 may be realized by, for example, a processor reading a program into a memory and executing it (software), or by hardware such as a dedicated circuit. It may be realized by combining software and hardware. In addition, some of the functions of the device may be realized by another computer capable of communicating with the device.

Next, the processing in the operating state measurement system 100 will be described. In the operating state measurement system 100, the measurement data acquired by the acceleration sensor 116 is transmitted to the server 121 via the measuring device 112 or the like, and the operating rate of the production line (for example, the operating or non-operating of the manufacturing device 115) is analyzed. To. Narrow band wireless communication is performed for wireless communication from the acceleration sensor 116 and the measuring device 112 to the gateway 113. In order to realize communication with a large number of measuring devices 112, for example, the following processing is performed in the measuring device 112.

The measuring device 112 performs a fast Fourier transform (FFT) on the measurement data acquired from the acceleration sensor 116, converts the data into data of frequency and intensity, and integrates the intensity in a certain division of frequency (divisional integration). By such processing, the amount of data is reduced.

Further, the measuring device 112 refers to a frequency band for discrimination for discriminating the operating state (operating or non-operating) among the data (frequency data) obtained by FFT and divisional integration (hereinafter, for convenience of explanation, the frequency and frequency band). The data (discrimination frequency data) of (referred to as) is selected, and the selected discrimination frequency data is transmitted to the server 121 via the gateway 113. By such processing, the amount of data is further reduced.

Here, the following problems may occur. Since the value (measurement data) of the acceleration sensor 116 attached to the manufacturing apparatus 115 fluctuates over time due to deterioration and drift of the package, acceleration sensor 116, etc., the signal strength at the frequency (discrimination frequency) determined to be transmitted is reduced. I have something to do. Therefore, in the present embodiment, the signal strength of each frequency is analyzed periodically or at a predetermined timing, and the discrimination frequency is changed. Here, when the discrimination frequency is changed, the server 121 needs a certain amount of spare frequency frequency data (partial frequency data) in order to maintain the desired accuracy. As described above, in narrow band wireless communication, it is difficult to transmit partial frequency data at one time because the communication band is limited. Therefore, we examined a method of transmitting partial frequency data according to the situation of the communication band.

FIG. 2 is a diagram showing an example of a main body that performs a process (operating state determination process) related to the determination (identification) of the operating state of the manufacturing apparatus 115. The operating state determination process is roughly divided into FFT, division integration, learning, and identification. Since the FFT and the segmented integral are highly related, they are performed on the same hardware, and since the learning and the discrimination are highly related, they are performed on the same hardware, but any combination is possible.

In addition, when applied when the device that performs FFT and division integration and the device that performs learning and identification are different, it has the effect of reducing communication between the two. However, the device that performs FFT and divisional integration and the device that performs learning and identification are not limited to a configuration in which they are separate devices, and may be configured to be the same device (one device).

Case 210 of the record "1" is a case where the measuring device 112 performs FFT and divisional integration, and the server 121 performs learning and identification. Case 220 of the record "2" is a case where the measuring device 112 performs FFT and divisional integration, and the gateway 113 performs learning and identification. Case 230 of the record "3" is a case where the gateway 113 performs FFT and divisional integration, and the server 121 performs learning and identification.

In the following, the case 210 of the record “1” will be described as an example. In case 230 of the record "3", wireless communication is performed between the gateway 113 and the LTE terminal 114.

Subsequently, the operating state determination process will be described in detail with reference to FIGS. 3 to 5.

FIG. 3 is a diagram for explaining an operating state (mode) in the operating state measuring system 100. The operating state measurement system 100 is provided with an initial setting mode 310 and a measurement mode 320. After the initial setting is performed in the initial setting mode 310, measurement (determination of the operating state of the manufacturing apparatus 115, etc.) is performed in the measurement mode 320.

More specifically, in the initial setting mode 310, the measuring device 112 time-divides the FFT of all frequencies and the result of the divisional integration (all frequency data) for the measurement data from the acceleration sensor 116 and transmits the measurement data to the server 121. The server 121 selects the optimum frequency (discrimination frequency) for discriminating the operating state of the manufacturing apparatus 115 based on the received all frequency data, and transmits the frequency to the measuring apparatus 112.

In the measurement mode 320, the measuring device 112 transmits the discrimination frequency data out of the FFT and the divisionally integrated total frequency data of the measurement data from the acceleration sensor 116 to the server 121 at regular intervals. At this time, the measuring device 112 intermittently transmits partial frequency data (frequency data obtained by excluding the discrimination frequency data from all frequency data) to the server 121 within a range that does not impose a tight communication band. The server 121 determines the operating state of the manufacturing apparatus 115 based on the determination frequency data. Further, the server 121 learns (collects partial frequency data) about a spare frequency to be used next as a discrimination frequency based on the partial frequency data. Further, the server 121 determines whether or not the discrimination frequency needs to be reviewed (detects a decrease in the signal strength of the discrimination frequency data), and selects a new discrimination frequency from the learned spare frequencies. do.

FIG. 4A is a diagram showing an image of a series of processes from acquisition of measurement data of the acceleration sensor 116 in the measuring device 112 to transmission of all frequency data in the initial setting mode 310. FIG. 5A is a diagram showing an image of all frequency data (an example of a data transmission method) transmitted from the measuring device 112 to the server 121 in the initial setting mode 310.

In the initial setting mode 310, the measuring device 112 performs an FFT on the measurement data acquired by the acceleration sensor 116 while the manufacturing device 115 is operating or not operating, and calculates the intensity for each predetermined frequency. Then, the measuring device 112 transmits the total frequency data obtained by dividing and integrating this to the server 121. The data acquisition time is the same for all frequencies, but transmission may be divided into multiple times. FIG. 5A shows an example in which the frequency is divided into bands of “15”, and the frequency is divided into three times for each of the five bands and transmitted to the server 121. Further, the data acquisition interval may be intermittent as shown in FIG. 4A, or may be continuous in time.

The server 121 receives all frequency data and selects a frequency at which the operating state can be discriminated as the discriminating frequency. Here, if the manufacturing apparatus 115 operates, the movable part moves, and even if it is a non-movable part, it vibrates due to the influence of the movable part, the acceleration data of the acceleration sensor 116 becomes a large value, and if the movable part does not move, the acceleration data is It will be a small value. This is common regardless of frequency. Therefore, the server 121 sets a threshold value for each frequency data of all frequency data, and determines operation or non-operation for each frequency. At this time, the server 121 calculates the SN (signal-noise) ratio. For example, the server 121 has a large acceleration (intensity) during operation, S (signal) when it is larger than the threshold value, and N (noise) when it is smaller than the threshold value. ) To calculate the SN ratio. Further, for example, the server 121 may determine that a sufficiently large value is S and a sufficiently small value is N without using a threshold value. If the acceleration can be measured accurately, the SN ratio becomes large and it can be judged that the frequency is good. The server 121 selects a frequency having a large SN ratio as a discrimination frequency based on the SN ratio.

When the measurement mode 320 is entered, data is transmitted as shown in the images shown in FIGS. 4 (B) and 5 (B). It is desirable that the acquisition interval of the measurement data of the accelerometer 116 is short and continuous in time. It is the same as the initial setting mode 310 that the measuring device 112 performs FFT and divisional integration on the acquired measurement data. The measuring device 112 continuously (in a short cycle) transmits the discrimination frequency data to the server 121. The server 121 receives the discrimination frequency data and determines whether the manufacturing apparatus 115 is operating or not operating. More specifically, the server 121 compares the intensity of the discrimination frequency data with the threshold value, and if it is larger than the threshold value, it determines that the manufacturing apparatus 115 is operating, and if it is smaller than the threshold value. , It is determined that the manufacturing apparatus 115 is not operating (not operating), and time series data (operating status data) of operating and / or non-operating is created.

Further, as shown in FIG. 5B, the measuring device 112 intermittently (long-period) transmits partial frequency data as learning data separately from the discrimination frequency data. The frequency to be transmitted changes over time. The reason why the learning data is acquired even in the measurement mode 320 is that the SN ratio decreases with the aging of the acceleration sensor 116, and the accuracy of determining the operating state decreases. In this case, the server 121 newly selects the discrimination frequency using the learning data and continues the measurement.

The server 121 calculates the SN ratio for each frequency of the acquired learning data. In the learning data for one measurement of each frequency data, only one of S or N is known, but by using the learning data for multiple measurements, S and N can be obtained, and the SN ratio can be determined. Can be calculated. Then, the server 121 sets the frequency having a large SN ratio as the next discrimination frequency.

Further, the server 121 also calculates the SN ratio for the discrimination frequency data, and when this is lowered, it is determined that the signal strength of the discrimination frequency data is reduced. The server 121 may determine the decrease in the SN ratio by comparing it with a predetermined value or by comparing it with the SN ratio of another frequency. When the server 121 finds deterioration, secular variation, etc. of the acceleration sensor 116, the server 121 changes the discrimination frequency. For example, in FIG. 5B, the server 121 used the frequency “2” from the time “0” as the discrimination frequency, but since the SN ratio decreased, the server 121 changed to the spare frequency “4” at the time TC. It is shown to be switching.

The server 121 selects a new discrimination frequency based on the learning data acquired in the measurement mode 320. The server 121 calculates a threshold value from the discrimination frequency data, determines whether the data is operating or not, and determines whether the learning data acquired at the same time is S or N based on the threshold value. Then, the server 121 determines the SN ratio for each frequency by using the learning data acquired a plurality of times in the same frequency range. The server 121 selects a frequency having a higher SN ratio as the next discrimination frequency.

In this way, in the measurement mode 320, deterioration of the sensor 116, secular variation, and the like can be detected, and another frequency can be appropriately set as the discrimination frequency. The data shown in FIG. 4 is data during wireless communication, and the measuring device 112 may handle more data than this.

FIG. 6 is a diagram showing an example of a processing procedure related to the operating state determination process. In FIG. 6, the positions in the vertical direction are summarized for each operation subject of each process.

In step S601, the measuring device 112 acquires measurement data (accelerometer data) from the acceleration sensor 116, performs FFT on the acquired measurement data, and converts the acquired measurement data into data of frequency and intensity. FIG. 7 shows an example of data for each frequency at a specific time (for example, a waveform when acceleration data sampled at 1 KHz is accumulated for a predetermined time and FFTed). The configuration is not limited to the configuration in which the acceleration data is accumulated for a predetermined time and FFTed, and the acceleration data at one time may be FFTed without accumulating the acceleration data.

The data shown in FIG. 7 show data during operation. In the non-operating data, there is no peak and the low background value spreads throughout. The acceleration data reflects the vibration of the equipment in addition to the acceleration of the moving part. The huge peak near the frequency of 0 Hz is due to the movement of the moving part, and multiple peaks due to vibration can be seen. Since the peak appears at the position of the natural frequency in the frequency direction (horizontal direction), the operating state of the manufacturing apparatus 115 can be accurately determined by applying the data of the frequency near the peak. If a frequency other than the natural frequency is used as the discrimination frequency, vibrations caused by other than the operation of the manufacturing apparatus 115 will be picked up, which may reduce the accuracy.

In step S602, the measuring device 112 performs a divisional integral with respect to the frequency and intensity data. FIG. 8 shows an example of frequency data after divisional integration. The peak is shown thicker and clearer than before the partial integration, and noise is suppressed. It should be noted that the data can be made smaller by performing the divisional integration, but it becomes impossible to separate the frequency with a large noise from the frequency with a small noise when the frequency division is made large.

In step S603, the measuring device 112 determines whether or not the initial setting mode 310 is set. When the measuring device 112 determines that the initial setting mode 310 is set, the process is transferred to step S604, and when it is determined that the mode is not the initial setting mode 310 (determined to be the measurement mode 320), the measuring device 112 shifts the process to step S612.

In step S604, the measuring device 112 transmits all the frequency data generated in step S602 to the gateway 113. All frequency data may be sent once or may be sent in multiple times.

In step S605, the gateway 113 transfers all frequency data received from the measuring device 112 to the server 121 via the LTE terminal 114.

In step S606, the server 121 estimates the signal-to-noise ratio based on the received all frequency data. The server 121 sets a threshold value for each frequency data of all frequency data, and determines operation or non-operation for each frequency. As the threshold value, for example, when the operating time and the non-operating time are the same or about the same, the average value of the intensities of each frequency data may be used, or the operating time and the non-operating time may be used. If there is a bias, the average value when the frequency data is extracted in consideration of the time may be used, or the value calculated by another method may be used. Subsequently, the server 121 calculates the SN ratio.

FIG. 9 is a diagram showing an example of the calculation result of the SN ratio. The server 121 uses the signal-to-noise ratio of the ratio of the average value 901 of the intensity during operation, which is larger than the threshold value, to the average value 902 of the intensity during non-operation, which is smaller than the threshold value, for the frequency data collected in a certain period. Calculated as 903.

In step S607, the server 121 selects the discrimination frequency. The frequency having the best SN ratio calculated in step S606 is selected as the discrimination frequency. The server 121 notifies the gateway 113 of the selected discrimination frequency via the LTE terminal 114, and performs the process of step S609.

FIG. 10 shows an example of the selected discrimination frequency. In FIG. 10, it is shown that the frequency 1010 having the highest SN ratio is selected as the discrimination frequency.

In step S608, the gateway 113 notifies the measuring device 112 of the selected discriminant frequency.

In step S609, the server 121 calculates a threshold value used for determining the operating state. In step S606, a threshold value was used to determine the operating state, but since the threshold value is a tentative value, a more accurate value is calculated. For example, the server 121 calculates for each frequency a value intermediate between the average value of the intensity during operation and the average value of the intensity during non-operation as a threshold value. The calculation of the threshold value is not limited to the above-mentioned calculation method. For example, a value intermediate between the median strength during operation and the median strength during non-operation may be used as the threshold value, or the threshold value may be calculated by another calculation method.

In step S610, the server 121 sets the threshold value calculated in step S609 and sets the measurement mode 320. More specifically, the server 121 writes the threshold value calculated in step S609 to the setting file, and transmits the setting file including the instruction to shift to the measurement mode 320 to the gateway 113 via the LTE terminal 114.

In step S611, the gateway 113 transmits an instruction to shift to the measurement mode 320 to the measuring device 112. By such processing, in the operating state measurement system 100, the mode shifts from the initial setting mode 310 to the measurement mode 320.

In step S612, the measuring device 112 transmits temporally continuous discriminant frequency data and intermittently transmits partial frequency data to the gateway 113 in a random or predetermined order.

In step S613, the gateway 113 transfers the frequency data transmitted from the measuring device 112 to the server 121 via the LTE terminal 114.

In step S614, the server 121 learns (accumulates) the received partial frequency data.

In step S615, the server 121 calculates the signal-to-noise ratio of the received discrimination frequency data.

In step S616, the server 121 determines whether or not the SN ratio calculated in step S615 is equal to or greater than a predetermined value (whether or not the determination frequency needs to be reviewed). When the server 121 determines that the value is equal to or higher than the predetermined value (review is not necessary), the server 121 shifts the process to step S617, and when it determines that the value is less than the predetermined value (review is necessary), the server 121 performs the process in step S606. Move.

Here, when the SN ratio is less than a predetermined value, it is necessary to review the discrimination frequency. In this case, the server 121 performs the SN ratio estimation (step S606) and the frequency selection (step S607) again, and notifies the measuring device 112 of the reselected determination frequency. After that, the frequency data of the reselected determination frequency is transmitted from the measuring device 112, and the server 121 performs the operation state determination (step S617).

In the frequency selection (step S607) for reviewing the discrimination frequency, the frequency is selected based on the learning (step S614) in the measurement mode 320, which is different from the initial setting mode 310. The server 121 calculates the SN ratio for each frequency based on the accumulated partial frequency data. It is determined whether the data related to one measurement of each frequency is operating "S" or non-operating "N", and the SN ratio is calculated by using the data related to multiple measurements. The server 121 sets the frequency having the largest SN ratio as a new discrimination frequency.

Further, the server 121 performs the threshold value calculation (step S609) again, and resets the calculated threshold value as the threshold value used for the operating state determination (step S617).

In step S617, the server 121 determines the operating state of the manufacturing apparatus 115 by the threshold value and creates the operating state data. The server 121 sets "1" (operating) when the threshold value is exceeded, and sets "0" (non-operating) when the threshold value is exceeded.

FIG. 11 is a diagram showing an example of operating state data (operating data 1110 and non-operating data 1120). The operating state data may be output to an output device (display, printer, etc.) connected to the server 121, may be output as a file, or may be output to a recording medium.

As described above, the operating state determination process is performed. Subsequently, the setting related to the operating state determination process will be described with reference to FIG.

The upper part of FIG. 12 shows a command table 1210 for explaining commands for the server 121 to make various settings on the measuring device 112. In the command table 1210, a command name 1211 indicating the name of the command, a range 1212 in which the value of the command can be taken, and a description 1213 of the command are shown.

As the command, a command 1214 for designating the range of the division integration, a command 1215 for designating the number of divisions of the frequency, and a command 1216 for describing the division number of the frequency are provided. The user uses command 1214 (command FD) to specify in what Hz the integration range of the FFT calculation result is integrated. The number of frequency divisions is the number of frequencies selected by the server 121 by learning, and is specified by the user using command 1215 (command N). The frequency division numbers from frequencies 1 to N are determined by the server 121 in (step S607) of FIG. 6, and the discrimination frequency is specified by the command 1216 (command FN).

In the first embodiment, the case where N is "1" is described. As an additional note, the case where N is plural (“3”) will be described in the second embodiment.

The lower part of FIG. 12 shows an example of a setting screen (setting screen 1220) for confirming the contents of a command. The setting screen 1220 includes an integration range 1221 designated by the user, a frequency division number 1222, and an automatically set frequency 1223 (frequency 1 to N automatically set by the server 121).

The setting screen 1220 may be output to an output device (display, printer, etc.) connected to the server 121, may be output as setting information, or may be output to a recording medium as setting information.

As described above, for example, the operating state measurement system is a data processing device that processes data (for example, measurement data) acquired from a sensor (for example, acceleration sensor 116) that measures the target device (for example, manufacturing device 115). (For example, the measuring device 112 and the gateway 113; see FIG. 2.) and the analysis device (for example, the gateway 113 and the server 121) that analyze the data processed by the data processing device and determine the operating state of the target device. 2) and a communication means (for example, an antenna of the measuring device 112 and an antenna of the gateway 113) for connecting the data processing device and the analysis device by communication including wireless communication, and the data processing device is provided. , The data acquired from the sensor is converted into frequency division data (for example, frequency data) divided for each frequency, and the frequency division data is transmitted to the analysis device via the communication means, and the analysis device is used. Determines the operating state of the target device based on the received frequency division data, and determines the frequency (for example, discrimination frequency) of the frequency division data transmitted by the data processing device to the analysis device. It is characterized by notifying the data processing device.

Further, for example, the operating state measurement system processes data (for example, measurement data) acquired from a sensor (for example, acceleration sensor 116) that measures the target device (for example, manufacturing device 115), and operates the target device. A computer for discriminating the state (a computer that performs FFT, divisional integration, learning, and identification) is provided, and the computer divides the data acquired from the sensor into frequency-divided data (for example, frequency data) for each frequency. From the frequency division data, the frequency of the operation state determination data (for example, discrimination frequency data) for determining the operation state and the frequency of the frequency division data related to the operation state determination data (for example, determination). The learning data (for example, partial frequency data) for determining (frequency) is selected, the operating state of the target device is determined based on the operating state determination data, and the operating state determination data is used. The frequency of the frequency division data used as the operating state determination data is changed based on the learning data, and the cycle of referring to the learning data is longer than the cycle of referring to the operating state determination data. It is characterized by.

According to the first and second characteristic configurations described above, since the frequency for discrimination is determined according to the target device, it is possible to grasp the operating state of various target devices.

Further, according to the first characteristic configuration described above, since the frequency data for discrimination is transmitted instead of all the frequency data, the operating state of the target device can be checked without tightening the communication band of the wireless communication. You will be able to distinguish.

Further, according to the second characteristic configuration described above, the frequency of the frequency division data used as the operating state determination data is changed based on the operating state determination data and the learning data. The operating state can be identified stably and accurately.

Further, for example, in the first characteristic configuration, the frequency division data transmitted from the data processing device to the analysis device includes operating state determination data (for example, discrimination frequency) for determining the operating state. Data) and learning data (for example, partial frequency data) for determining the frequency of the frequency division data transmitted by the data processing device, and the learning data is based on the operating state determination data. Also, the measurement interval of the sensor is long (see FIG. 5B). According to this configuration, the frequency of the frequency division data used as the operating state determination data is changed based on the operating state determination data and the learning data, so that the operating state of the target device can be identified stably and accurately. can do. In addition, since the measurement interval of the sensor is longer in the learning data than in the operating state determination data, the learning data is accumulated within the range that does not overwhelm the communication band of wireless communication while constantly determining the operating state of the target device. You will be able to.

Further, for example, the learning data is characterized in that the frequency is changed more frequently than the operating state determination data (see FIG. 5B). Since the frequency of the learning data is changed more frequently than the operating state determination data, it becomes possible to select from a plurality of candidates when switching the frequency of the operating state determination data.

(2) Second Embodiment In the present embodiment, the difference from the first embodiment in that the operating state of the manufacturing apparatus is discriminated using a plurality of discriminating frequencies will be mainly described. do. In the present embodiment, the same reference numerals are used for the same configurations as those in the first embodiment, and the description thereof will be omitted as appropriate.

FIG. 13 is a diagram showing an example of a data transmission method in the measurement mode 320. In the example shown in FIG. 13, the frequency data 1310 having the discrimination frequency 5 and the frequency data 1320 and the frequency data 1330 having the 9th frequency data 1330 are constantly transmitted, and the other frequency data 1340 is intermittently transmitted. It is shown that.

In the present embodiment, the server 121 selects the discrimination frequency in step S607. At this time, the server 121 selects the upper N frequencies among the frequencies having a good SN ratio calculated in step S606 as the discrimination frequencies. N is a value specified by the user using the command 1215 (command N), and is "3" in this example. Then, the server 121 notifies the measuring device 112 of the selected discrimination frequency via the gateway 113.

Further, the server 121 determines the operating state of the manufacturing apparatus 115 in step S617. At this time, the server 121 determines whether or not the intensity of the frequency data exceeds the threshold value for each of the plurality of discrimination frequencies, requests operation or non-operation, and determines the operating state of the manufacturing apparatus 115 by a majority vote. do. For example, when the server 121 is determined to be operating by the frequency data 1310 of No. 5, is determined to be operating by the frequency data 1320 of No. 9, and is determined to be non-operating by the frequency data 1330 of No. 12, the operating state of the manufacturing apparatus 115 is determined. Will be decided as an operation. It should be noted that the configuration is not limited to the one determined by majority vote, and when it is determined that the system is operating or not operating, the operating state may be determined as unknown.

In the present embodiment, a plurality of discrimination frequencies can be selected when there is a margin in consideration of the capacity of wireless communication.

In this way, by using a plurality of discrimination frequencies in consideration of the capacity of wireless communication, it is possible to improve the accuracy of discrimination of the operating state of the manufacturing apparatus without tightening the communication band of wireless communication.

(3) Third Embodiment In the present embodiment, the point that the learning target of the preliminary frequency is narrowed down is different from the first embodiment, and the difference is mainly described. In the present embodiment, the same reference numerals are used for the same configurations as those in the first embodiment, and the description thereof will be omitted as appropriate.

FIG. 14 is a diagram showing an example of a data transmission method in the measurement mode 320. In the example of FIG. 14, frequency data 1410 having a discrimination frequency of No. 5 is constantly transmitted, and frequency data 1420 before and after No. 5 is intermittently transmitted. This is because the candidate of the discrimination frequency is set, and it indicates that the candidate frequency data is selected and transmitted.

The method of narrowing down the learning target of the preliminary frequency is not limited to the configuration of narrowing down to the frequencies before and after the discrimination frequency, and may be a configuration of narrowing down to a frequency having a higher SN ratio or another configuration.

In the present embodiment, the server 121 sets a spare frequency in addition to the discrimination frequency in step S607, and notifies the measuring device 112. In response to such notification, in step S612, the measuring device 112 constantly transmits the frequency data of the discrimination frequency, and intermittently transmits the frequency data of the spare frequency.

By narrowing down the candidates for the discrimination frequency in this way, the minimum period required for learning can be shortened.

As described above, since the operating states of various target devices can be accurately and quickly identified, a highly reliable system can be realized.

(4) Other Embodiments In the above-described embodiment, the case where the present invention is applied to the operating state measurement system 100 has been described, but the present invention is not limited to this, and various other operations are described. It can be widely applied to state measurement systems and operating state measurement methods.

Further, in the above-described embodiment, the case where the data processing device performs FFT and divisional integration and transmits the frequency data to the analysis device has been described, but the present invention is not limited to this, and the data processing device uses the FFT. The frequency data may be transmitted to the analyzer without performing the divisional integration.

Further, in the above-described embodiment, the case where the LTE terminal 114 is provided between the gateway 113 and the server 121 has been described, but the present invention is not limited to this, and the LTE terminal 114 is provided between the gateway 113 and the server 121. May not be provided, or other network devices may be provided.

Further, in the above-described embodiment, various data have been described using the XX table and the XX file for convenience of explanation, but the data structure is not limited and may be expressed as XX information or the like.

In addition, the information such as programs, tables, and files that realize each function in the above description includes a memory, a hard disk, a storage device such as an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD. Can be placed in.

Further, the above-mentioned configuration may be appropriately changed, rearranged, combined, or omitted as long as it does not exceed the gist of the present invention.

100 …… Operating condition measurement system, 110 …… Factory, 111 …… Equipment, 112 …… Measuring device, 113 …… Gateway, 114 …… LTE terminal, 115 …… Manufacturing equipment, 116 …… Accelerometer, 120 …… Data center, 121 …… Server.

Claims (11)

A data processing device that processes the data acquired from the sensor that measures the target device, and
An analysis device that analyzes the data processed by the data processing device and determines the operating state of the target device, and
A communication means for connecting the data processing device and the analysis device by communication including wireless communication, and
In the operating condition measurement system equipped with
The data processing device is
The data acquired from the sensor is converted into frequency division data divided for each frequency, and the data is converted into frequency division data.
The frequency division data is transmitted to the analysis device via the communication means, and the frequency division data is transmitted to the analysis device.
The analyzer is
Based on the received frequency division data, the operating state of the target device is determined, and the frequency of the frequency division data transmitted by the data processing device to the analysis device is determined and notified to the data processing device. Operation status measurement system. In claim 1,
The frequency division data transmitted from the data processing device to the analysis device includes the operation state determination data for determining the operation state and the frequency of the frequency division data transmitted by the data processing device. Contains training data,
The learning data is an operating state measurement system characterized in that the measurement interval of the sensor is longer than that of the operating state determination data. In claim 2,
The learning data is an operating state measurement system characterized in that the frequency is changed more frequently than the operating state determination data. In claim 2,
An operating state measurement system characterized in that the frequency of the operating state determination data transmitted to the analysis device is determined by the analysis device based on the learning data. In claim 4,
The analysis device determines whether or not the frequency of the operating state determination data needs to be changed based on the SN ratio of the operating state determination data, and the operating state determination data is based on the SN ratio of the learning data. An operating condition measurement system characterized by determining the frequency to which the frequency is changed. In claim 2,
An operating state measurement system characterized in that a plurality of frequencies of the operating state determination data are set. In claim 2,
An operating state measurement system characterized in that the frequency of the learning data and the frequency of the operating state determination data transmitted to the analysis device are determined by the analysis device based on the learning data. In claim 1,
The operating state measurement system characterized in that the frequency division data is generated by FFT (Fast Fourier Transform). A data processing device that processes data acquired from a sensor that measures the target device, an analysis device that analyzes the data processed by the data processing device, and determines the operating state of the target device.
A communication means for connecting the data processing device and the analysis device by communication including wireless communication, and
In the operating state measurement method using
The data processing device is
The data acquired from the sensor is converted into frequency division data divided for each frequency, and the data is converted into frequency division data.
The frequency division data is transmitted to the analysis device via the communication means, and the frequency division data is transmitted to the analysis device.
The analyzer is
Based on the received frequency division data, the operating state of the target device is determined, and the frequency of the frequency division data transmitted by the data processing device to the analysis device is determined and notified to the data processing device. Operation status measurement method. In an operating state measurement system equipped with a computer that processes data acquired from a sensor that measures the target device and determines the operating state of the target device.
The calculator
The data acquired from the sensor is converted into frequency division data divided for each frequency, and the data is converted into frequency division data.
From the frequency division data, the operation state determination data for determining the operation state and the learning data for determining the frequency of the frequency division data related to the operation state determination data are selected.
The operating state of the target device is determined based on the operating state determination data, and the frequency of the frequency division data to be the operating state determination data is determined based on the operating state determination data and the learning data. change,
An operating state measurement system characterized in that the cycle for referencing the learning data is longer than the cycle for referencing the operating state determination data. In the operating state measurement method using a computer that processes the data acquired from the sensor that measures the target device and determines the operating state of the target device.
The calculator
The data acquired from the sensor is converted into frequency division data divided for each frequency, and the data is converted into frequency division data.
From the frequency division data, the operation state determination data for determining the operation state and the learning data for determining the frequency of the frequency division data related to the operation state determination data are selected.
The operating state of the target device is determined based on the operating state determination data, and the frequency of the frequency division data to be the operating state determination data is determined based on the operating state determination data and the learning data. change,
A method for measuring an operating state, characterized in that the cycle for referencing the learning data is longer than the cycle for referencing the operating state determination data.

Images