Control signal monitoring device, control signal monitoring method, and control signal monitoring program

Abstract

A model generation unit (17) generates, on the basis of change interval data obtained from log data stored in a log data storage unit (12), a change interval model obtained by modeling change interval data representing an interval between a start signal that causes operation of a device and a completion signal that appears as a result of the operation of the device. An abnormality degree calculation unit (18) calculates the degree of abnormality of the operation of the device from an actual measurement value, which is an interval between the start signal and completion signal obtained from the device, and a predicted value, which is an interval predicted using the change interval model generated by the model generation unit (17).

Description

Control signal monitoring device, control signal monitoring method, and control signal monitoring program

[0001] The present disclosure relates to techniques for monitoring control signals that control the operation of devices.

[0002] PLCs exchange control signals with various devices in manufacturing facilities. PLC stands for Programmable Logic Controller. Specific examples of devices include robots, conveyors, and cylinders. Control signals are characterized by being composed primarily of binary digital signals with ON and OFF states. The control signals input and output to the PLC include signals whose state changes when each device starts operating (hereafter referred to as start signals) and signals whose state changes when the device completes operating (hereafter referred to as completion signals). By appropriately monitoring the start and completion signals, it is possible to monitor the operation of the devices.

[0003] Patent Document 1 describes a method for detecting abnormal operation of a device by monitoring the time difference between the rising edge of one PLC signal and the falling edge of another PLC signal. In Patent Document 1, an allowable time width is set for the reference operation pattern, and a timing deviation within the normal range is not detected as abnormal operation.

[0004] Japanese Patent Application Publication No. 09-22308

[0005] The technology described in Patent Document 1 compares the time difference between changes in signals with a tolerance to determine whether a device is operating abnormally. As a result, normal operational fluctuations that occur over the long term due to temperature changes, such as air temperature, may be detected as abnormal. Furthermore, in a device with two different operating patterns, it may be necessary to set a tolerance that overlooks an abnormality in one of the operations. The present disclosure aims to enable appropriate device abnormality determination using a start signal and a completion signal.

[0006] The control signal monitoring device according to the present disclosure includes a model generation unit that generates a change interval model that models change interval data that represents the interval between a start signal that causes an operation of the device and a completion signal that appears as a result of the operation of the device, and an abnormality degree calculation unit that calculates the abnormality degree of the operation of the device using the change interval model generated by the model generation unit.

[0007] In the present disclosure, an abnormality in a device is monitored using a change interval model that models change interval data that represents the interval between a start signal and a completion signal. By using the change interval model, an abnormality in the device can be appropriately determined from the start signal and the completion signal.

[0008] 1 is a configuration diagram of a control signal monitoring system 100 according to a first embodiment. A functional configuration diagram of a control signal monitoring device 10 according to the first embodiment. A hardware configuration diagram of the control signal monitoring device 10 according to the first embodiment. A flowchart showing the processing flow of the control signal monitoring device 10 according to the first embodiment. An explanatory diagram of a reduction in the size of stored data according to the first embodiment. An explanatory diagram of a method for generating change interval data according to the first embodiment. An explanatory diagram of a learning method when generating a change interval model that models change interval data according to the first embodiment. A diagram showing an example display of anomaly degrees according to the first embodiment. An explanatory diagram of an abnormality in a device having two operation patterns according to the first embodiment. A flowchart showing the processing flow of the control signal monitoring device 10 according to a second embodiment. An explanatory diagram of a method for predicting the timing at which an abnormality will occur according to the second embodiment. A diagram showing an example display of the timing at which an abnormality will occur according to the second embodiment.

[0009] Embodiment 1. ***Description of Configuration***

[0010] The configuration of a control signal monitoring system 100 according to a first embodiment will be described with reference to Fig. 1. The control signal monitoring system 100 includes a control signal monitoring device 10, a PLC 20, and a manufacturing facility 30. The manufacturing facility 30 includes one or more devices 31. The control signal monitoring device 10 and the PLC 20 are connected via a transmission path 41. The PLC 20 and each of the one or more devices 31 included in the manufacturing facility 30 are connected via a transmission path 42. The PLC 20 exchanges control signals with each device 31 via the transmission path 42. The control signal monitoring device 10 acquires control signals or control signal logs from the PLC 20 via the transmission path 41.

[0011] The functional configuration of the control signal monitoring device 10 according to the first embodiment will be described with reference to Fig. 2. The control signal monitoring device 10 includes, as functional components, an acquisition unit 11, a log data storage unit 12, an input unit 13, a signal set storage unit 14, a change interval calculation unit 15, a change interval storage unit 16, a model generation unit 17, an anomaly degree calculation unit 18, and a display control unit 19.

[0012] The hardware configuration of the control signal monitoring device 10 according to the first embodiment will be described with reference to Fig. 3. The control signal monitoring device 10 is a computer. The control signal monitoring device 10 includes the following hardware components: a processor 101, a memory 102, a storage 103, and a communication interface 104. The processor 101 is connected to other hardware components via signal lines and controls the other hardware components.

[0013] The processor 101 is an IC that performs processing. IC stands for Integrated Circuit. Specific examples of the processor 101 include a CPU, a DSP, and a GPU. CPU stands for Central Processing Unit. DSP stands for Digital Signal Processor. GPU stands for Graphics Processing Unit.

[0014] The memory 102 is a storage device that temporarily stores data. Specific examples of the memory 102 include SRAM and DRAM. SRAM stands for Static Random Access Memory. DRAM stands for Dynamic Random Access Memory.

[0015] The storage 103 is a storage device that stores data. A specific example of the storage 103 is an SSD. SSD is an abbreviation for Solid State Drive. The storage 103 may also be a portable recording medium such as an SD (registered trademark) memory card, CompactFlash (registered trademark), NAND flash, a flexible disk, an optical disk, a compact disk, a Blu-ray (registered trademark) disk, or a DVD. SD is an abbreviation for Secure Digital. DVD is an abbreviation for Digital Versatile Disk.

[0016] The communication interface 104 is an interface for communicating with external devices. Specific examples of the communication interface 104 include Ethernet (registered trademark), USB, and HDMI (registered trademark) ports. USB stands for Universal Serial Bus. HDMI stands for High-Definition Multimedia Interface.

[0017] The functions of the acquisition unit 11, input unit 13, change interval calculation unit 15, model generation unit 17, abnormality degree calculation unit 18, and display control unit 19 of the control signal monitoring device 10 are realized by software. A program that realizes these functions is stored in the storage 103. This program is read into the memory 102 by the processor 101 and executed by the processor 101. In this way, the functions of each functional component of the control signal monitoring device 10 are realized.

[0018] The functions of the log data storage unit 12, the signal set storage unit 14, and the change interval storage unit 16 of the control signal monitoring device 10 are realized by a storage device such as the memory 102 or the storage 103. Note that these functions may also be realized by a storage device external to the control signal monitoring device 10.

[0019] ***Description of Operation*** The operation of the control signal monitoring device 10 according to the first embodiment will be described with reference to Figures 4 to 8. The operation procedure of the control signal monitoring device 10 according to the first embodiment corresponds to the control signal monitoring method according to the first embodiment. Furthermore, the program that realizes the operation of the control signal monitoring device 10 according to the first embodiment corresponds to the control signal monitoring program according to the first embodiment.

[0020] Referring to FIG. 4, the processing flow of the control signal monitoring device 10 according to the first embodiment will be described. (Step S101: Control Signal Collection Process) The PLC 20 communicates a plurality of control signals with each device 31 included in the manufacturing facility 30. The acquisition unit 11 acquires the plurality of control signals communicated with each device 31 from the PLC 20. Here, the acquisition unit 11 is described as acquiring the plurality of control signals themselves, but it may also acquire a log of the plurality of control signals. In this case, the acquisition unit 11 acquires a plurality of control signals for a time period during which a series of operations of each device occurs at least two times. The plurality of control signals include a start signal and a completion signal. The start signal is a signal that triggers the operation of the device and changes state when the device operation starts. The completion signal is a signal that appears as a result of the operation of the device and changes state when the device operation is completed. All of the plurality of control signals are binary digital signals representing ON and OFF.

[0021] The acquisition unit 11 stores the acquired control signal as log data in the log data storage unit 12. At this time, the acquisition unit 11 does not store the control signal as a binary signal value as log data, but replaces it with change time data indicating the time when the value changed between ON and OFF and stores the log data. This is expected to reduce the size of the data to be stored.

[0022] Referring to FIG. 5, we will explain how to reduce the size of stored data. Control signals are binary digital signals whose state changes over time. Signal sequence d100 indicates the state of a single signal observed at regular intervals. If signal sequence d100 were stored as log data as is, binary values ​​(d101 to d10n) for each time would be stored. However, individual control signals only change approximately once every few seconds. Therefore, storing log data as time information, such as the signal rise time d201 and fall time d202, often results in less storage capacity being used. Furthermore, storing the log data as time information is expected to shorten the calculation time when calculating the time difference between changes between signals.

[0023] (Step S102: Signal Set Information Input Processing) The input unit 13 inputs signal set information for a device to be monitored. If there are multiple devices to be monitored, the input unit 13 inputs signal set information for each of the multiple devices to be monitored. Note that the input unit 13 may input multiple pieces of signal set information for one device. The input unit 13 stores the input signal set information in the signal set storage unit 14. The signal set information includes, for each monitoring target, at least a signal name of a start signal, a change trigger of the start signal, a signal name of a completion signal, and a change trigger of the completion signal. The signal name is a name that can identify the control signal. The change trigger is information indicating whether to monitor the rising or falling edge of the signal. The signal set information also includes supplemental information for strictly defining the correspondence between signals, for example, when a signal changes multiple times within a single operating cycle. The supplemental information indicates, for example, which change within the cycle to monitor. Various methods are possible for creating the signal set information, such as manual setting by a person, extraction from device design information, or automatic extraction from data. Any method may be used here.

[0024] (Step S103: Change interval calculation process) The change interval calculation unit 15 generates change interval data from the signal log data stored in step S101 and the signal set information stored in step S102. The change interval calculation unit 15 stores the change interval data in the change interval storage unit 16.

[0025] A method for generating change interval data will be described with reference to FIG. 6 . The change interval calculation unit 15 identifies the start signal d311 and the completion signal d321 specified in the signal set information d400 from among the signals included in the signal log data d300. The change interval calculation unit 15 identifies the time of change d312 specified by the change trigger in the identified start signal. The change interval calculation unit 15 also identifies the time of change d322 specified by the change trigger in the identified completion signal, which occurred immediately after change d312. The change interval calculation unit 15 then generates the difference between the time of change d312 and the time of change d322 as change interval data d301 and d302. At this time, if a count specification d423 or the like is included in the supplemental information in the signal set information, the change interval calculation unit 15 adjusts the object of calculation in accordance with the count specification d423 or the like. In this case, since the count specification for the completion signal is set to 2, the calculation is adjusted to look at the second change in the cycle (see d323).

[0026] (Step S104: Model Generation Process) The model generation unit 17 uses a regression model to generate a change interval model that models the change interval data stored in step S103. The regression model to be used may be an autoregressive model, a neural network model, or the like.

[0027] With reference to FIG. 7 , a learning method for generating a change interval model that models change interval data will be described. The change interval data d500 is time-series data having an order of occurrence. The change interval model uses past data as input data d511 and predicts data 512 ahead of the input data d511. At this time, multiple pieces of data d501 may be input instead of a single piece of data. The model generation unit 17 performs learning so as to minimize the difference d513 between the future data d512 predicted by the change interval model and the actual data. This enables the change interval model to calculate the expected value of a future change interval based on past change intervals.

[0028] (Step S105: Abnormality degree calculation process) The abnormality degree calculation unit 18 calculates the abnormality degree of the operation of the device to be monitored using the change interval model generated in step S104. Specifically, the abnormality degree calculation unit 18 calculates the abnormality degree of the device by comparing an actual measurement value, which is the interval between a start signal and a completion signal obtained from the device, with a predicted value, which is the interval predicted using the change interval model. As a method of calculating the abnormality degree, ｎ and the expected value P ｎ Using this, the abnormality level = (|P ｎ -L ｎ |) / P ｎ The actual measurement value here is the latest interval between the start signal and the completion signal specified in the signal set information input in step S102.

[0029] (Step S106: Display control process) The display control unit 19 displays the abnormality degree and the like calculated in step S105 on a display device connected via the communication interface 104. Specifically, the display control unit 19 may display the abnormality degree, expected value, and actual measurement value for the device to be monitored as the monitoring result.

[0030] An example of the display of the abnormality level will be described with reference to FIG. 8 . The display control unit 19 displays an abnormality level display screen d610 that indicates the abnormality level for each device. This allows the user to monitor the abnormality level of each device. At this time, the display control unit 19 may highlight devices whose abnormality level exceeds a threshold, for example by changing the color. If the user wishes to check the abnormality level of a specific device in detail, the user selects the device to be checked. The display control unit 19 then displays a change interval transition graph d600. The change interval transition graph d600 shows the transitions of the expected value d601 of the change interval calculated using the change interval model, the actual measured value d602 of the change interval, and the abnormality level d603 calculated by comparing these. By referring to the change interval transition graph d600, the user can visually confirm the details of the abnormality, such as the operation of the selected device being behind schedule or fluctuating irregularly.

[0031] The process of Fig. 4 is executed periodically. When the process of Fig. 4 is executed for the second time or later, the processes of steps S102 to S104 may be skipped. The process of step S102 need only be executed when the control signal to be monitored is changed. Furthermore, the processes of steps S103 and S104 need only be executed when the change interval model is regenerated.

[0032] ***Effects of First Embodiment*** As described above, the control signal monitoring device 10 according to the first embodiment monitors an abnormality in the device using a change interval model that models change interval data that represents the interval between the start signal and the completion signal. By using the change interval model, it becomes possible to appropriately determine an abnormality in the device.

[0033] By using the change interval model, it is possible to prevent normal operation fluctuations that occur over the long term due to temperature changes such as air temperature from being detected as an abnormality, as in the technology described in Patent Document 1. Furthermore, by using the change interval model, it is possible to avoid having to set a tolerance that overlooks an abnormality in one of the operations of a device that has two different operation patterns, as in the technology described in Patent Document 1.

[0034] Referring to FIG. 9 , an abnormality in a device with two motion patterns will be specifically described. Assume that the robot 50 has two motion patterns: one for moving an object from point A to point C, and the other for moving an object from point B to point C. When the motion time from point A is longer than the motion time from point B, using the technique of Patent Document 1 requires that the allowable time width be set longer than the motion time width from point A. As a result, even if an abnormality occurs during the movement from point B, causing the motion to slow down, the allowable time width may fall within the allowable time width, potentially overlooking the abnormality. However, the control signal monitoring device 10 according to the first embodiment calculates the degree of abnormality of the change interval of a specified signal using a change interval model. Therefore, it is possible to calculate the degree of abnormality using separate change interval models for the movement from point A and the movement from point B. This allows for appropriate detection of abnormalities in a device with two motion patterns.

[0035] Furthermore, the control signal monitoring device 10 according to the first embodiment calculates the degree of abnormality using the interval between the start signal and the completion signal specified in the signal set information. Therefore, there is no need to perform calculations using the interval between unnecessary signals. The technology described in Patent Document 1 does not provide a method for narrowing down the signals connected to the PLC to be monitored. Therefore, there may be a large amount of data to be monitored. If the number of signals connected to the PLC is n, the maximum number of ordered combinations of ON and OFF states of each signal is P ２ｎ，２ = 2n × (2n - 1) possible monitoring targets. In contrast, the control signal monitoring device 10 according to the first embodiment monitors only the necessary number of targets, thereby reducing the processing load.

[0036] ***Other Configurations*** <Modification 1> In the first embodiment, each functional component is realized by software. However, in Modification 1, each functional component may be realized by hardware. The differences between Modification 1 and the first embodiment will be described below.

[0037] When each functional component is realized by hardware, the control signal monitoring device 10 includes an electronic circuit instead of the processor 101, the memory 102, and the storage 103. The electronic circuit is a dedicated circuit that realizes the functions of each functional component, the memory 102, and the storage 103.

[0038] Possible electronic circuits include a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, a logic IC, a GA, an ASIC, and an FPGA. GA stands for Gate Array. ASIC stands for Application Specific Integrated Circuit. FPGA stands for Field-Programmable Gate Array. Each functional component may be realized by a single electronic circuit, or each functional component may be distributed across multiple electronic circuits.

[0039] <Modification 2> As a modification 2, some of the functional components may be realized by hardware, and other functional components may be realized by software.

[0040] The processor 101, memory 102, storage 103, and electronic circuitry are collectively referred to as a processing circuit. In other words, the functions of the respective functional components are realized by the processing circuit.

[0041] Furthermore, the term "unit" in the above description may be read as a "circuit," "step," "procedure," "process," or "processing circuit."

[0042] Embodiment 2. Embodiment 2 differs from embodiment 1 in that it predicts the timing at which an abnormality occurs. In embodiment 2, this difference will be explained, and explanation of the same points will be omitted.

[0043] ***Description of Operation*** The processing flow of the control signal monitoring device 10 according to embodiment 2 will be described with reference to Fig. 10. The processing from step S201 to step S203 is the same as the processing from step S101 to step S103 in Fig. 4.

[0044] (Step S204: Model Generation Processing) The model generation unit 17 uses a regression model to generate a change interval model that models the change interval data stored in step S203. The regression model used in the second embodiment is capable of trend analysis like an autoregressive model. The regression model breaks down the change interval data into fluctuation factors such as trend fluctuation, seasonal fluctuation, and irregular fluctuation, and stores the data as parameters that enable data reproduction and prediction.

[0045] (Step S205: Abnormality Degree Prediction Processing) The abnormality degree calculation unit 18 predicts the timing at which an abnormality will occur using the change interval model generated in step S204. Specifically, the abnormality degree calculation unit 18 predicts, as the timing at which an abnormality will occur, the timing at which a predicted value, which is an interval predicted using the change interval model, reaches a threshold value used when determining whether or not there is an abnormality in the operation of the device. The threshold value may be set manually, or may be set by multiplying the average value over a certain period of time in the past by a coefficient.

[0046] A method for predicting the timing at which an abnormality will occur will be described with reference to Fig. 11. A predicted value d702 for each time is obtained using a change interval model obtained from the change interval d701 of the device. An intersection between the predicted value d702 and a threshold d703, which is determined to be an abnormality when exceeded, is identified. The time of the identified intersection becomes the predicted time d704 at which an abnormality will occur. Furthermore, if there is no intersection or if there is no threshold, no abnormality is predicted to occur.

[0047] (Step S206: Display control process) The display control unit 19 displays the timing at which the abnormality predicted in step S205 will occur, etc., on a display device connected via the communication interface 104. Specifically, as shown in Fig. 12, the display control unit 19 displays the time at which the abnormality is predicted to occur for each device.

[0048] ***Effects of Embodiment 2*** As described above, the control signal monitoring device 10 according to Embodiment 2 identifies the timing at which a predicted value, calculated using a change interval model, reaches a threshold value as the timing at which an abnormality will occur. This allows users to plan repair parts or maintenance for the device in advance, thereby improving the availability of equipment.

[0049] The embodiments and modifications of the present disclosure have been described above. Some of these embodiments and modifications may be combined and implemented. Furthermore, one or more of them may be implemented partially. Note that the present disclosure is not limited to the above embodiments and modifications, and various modifications are possible as needed.

[0050] 100 Control signal monitoring system, 10 Control signal monitoring device, 20 PLC, 30 Manufacturing equipment, 31 Device, 41 Transmission path, 42 Transmission path, 11 Acquisition unit, 12 Log data storage unit, 13 Input unit, 14 Signal set storage unit, 15 Change interval calculation unit, 16 Change interval storage unit, 17 Model generation unit, 18 Anomaly degree calculation unit, 19 Display control unit, 101 Processor, 102 Memory, 103 Storage, 104 Communication interface.

Claims

1. A control signal monitoring device comprising: a model generation unit that generates a change interval model that models change interval data that represents the interval between a start signal that causes a device to operate and a completion signal that appears as a result of the device's operation; and an abnormality degree calculation unit that calculates the abnormality degree of the device's operation using the change interval model generated by the model generation unit.

2. The control signal monitoring device according to claim 1, wherein the abnormality degree calculation unit calculates the abnormality degree of the operation of the device by comparing an actual measured value, which is the interval between the start signal and the completion signal obtained from the device, with a predicted value, which is the interval predicted using the change interval model.

3. The control signal monitoring device according to claim 1 or 2, further comprising an acquisition unit that acquires the start signal and stores it in a log data storage unit as log data of the start signal, and acquires the completion signal and stores it in the log data storage unit as log data of the completion signal, and the model generation unit generates the change interval model based on the log data stored in the log data storage unit by the acquisition unit.

4. The control signal monitoring device according to claim 3, wherein the start signal and the completion signal are each a binary signal, and the acquisition unit replaces the start signal with the time when the value of the start signal changed, and stores this as log data of the start signal, and replaces the completion signal with the time when the value of the completion signal changed, and stores this as log data of the completion signal.

5. The control signal monitoring device according to claim 3 or 4, further comprising a change interval calculation unit that identifies the start signal and the completion signal specified in the signal set information from the log data and generates the change interval data, and the model generation unit models the change interval data generated by the change interval calculation unit to generate the change interval model.

6. A control signal monitoring device according to claim 5, wherein the log data storage unit stores log data for each of one or more of the start signals and one or more of the completion signals, and the signal set information includes a signal name that can identify the start signal and a signal name that can identify the completion signal.

7. A control signal monitoring device as claimed in any one of claims 1 to 6, wherein the anomaly degree calculation unit predicts the timing at which an anomaly will occur using a predicted value, which is an interval predicted using the change interval model, and a threshold value used when determining whether or not there is an anomaly in the operation of the device.

8. A control signal monitoring method in which a computer generates a change interval model that models change interval data that represents the interval between a start signal that causes a device to operate and a completion signal that appears as a result of the device's operation, and the computer calculates the degree of abnormality in the device's operation using the change interval model.

9. A control signal monitoring program that causes a computer to function as a control signal monitoring device that performs a model generation process that generates a change interval model that models change interval data that represents the interval between a start signal that causes a device to operate and a completion signal that appears as a result of the device's operation, and an abnormality degree calculation process that calculates the abnormality degree of the device's operation using the change interval model generated by the model generation process.

Images