Mechatronics device maintenance system, mechatronics device maintenance server, mechatronics device maintenance method, and mechatronics device maintenance program

Abstract

To provide a technology that enables maintenance personnel to efficiently perform maintenance and inspection of mechatronics devices.SOLUTION: A mechatronics device maintenance system according to an embodiment, comprises: a mechatronics device whose mechanism is controlled by software; and a central server capable of processing data of the mechatronics device. The central server includes a log data acquisition unit that acquires log data including information on the operation of the mechatronics device, a designated log data acquisition unit that acquires designated log data from the log data that is necessary to reproduce the operating status, and a data conversion unit that converts the designated log data into a specified data file arranged in chronological order in order to reproduce the operating status of the designated log data.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] An embodiment of the present invention relates to a mechatronics device maintenance system, a mechatronics device maintenance server, a mechatronics device maintenance method, and a mechatronics device maintenance program. [Background technology]

[0002] The maintenance system for mechatronics equipment, including automated ticket gates, is generally carried out by dispatching maintenance personnel to the site in response to an on-call from the station where a malfunction has actually occurred. In recent years, the mechanisms and functions of mechatronics equipment have become more complex and sophisticated, making it generally difficult to grasp the trends in failure modes, and much of the problem relies on the experience and intuition of maintenance personnel. In response to this, in order to enable preventive condition monitoring maintenance that does not rely on on-call responses (remedial maintenance), maintenance inspection cycles have sometimes been reviewed from time to time.

[0003] For example, Patent Document 1 discloses a technique for continuously referencing the operating status and failure history of various sensors installed in station equipment (automated ticket gates) and manually determining the operating status of each piece of station equipment to extract station equipment that is likely to fail.The technique then discloses a technique for reviewing the maintenance and inspection cycle of the extracted station equipment, such as by bringing forward the timing of the maintenance and inspection.

[0004] However, it is difficult to completely prevent unexpected breakdowns, and so there is still a need to improve the efficiency of on-call maintenance and inspection work itself. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2004-038747 A Summary of the Invention [Problem to be solved by the invention]

[0006] Maintenance inspections are performed by maintenance personnel who go to the locations where mechatronic devices are installed. Each maintenance personnel performs maintenance inspections of multiple pieces of mechatronic devices in a day. This causes a problem that the maintenance personnel cannot see or know the status of the mechatronic devices before arriving at the site. There is also a problem that the maintenance personnel go to the site but are unable to reproduce the operation that caused the failure, and must take measures for the mechatronic devices based on guesswork. There is also a problem that the mechatronic devices cannot be visually inspected due to their structure or operation. There is also a problem that the behavior of the mechatronic devices can only be visually inspected at the actual operating speed, and the behavior cannot be stopped midway. There is also a problem that the maintenance personnel must carry out measures while remembering the events that require maintenance (i.e., relying on vague memories).

[0007] The present invention has been made in light of the above-mentioned circumstances, and has an object to provide a technique that enables maintenance personnel to efficiently perform maintenance and inspection of mechatronics equipment. [Means for solving the problem]

[0008] A mechatronics equipment maintenance system according to an embodiment comprises a mechatronics equipment whose mechanism operates under the control of software, and a central server capable of processing data of the mechatronics equipment, the central server comprising a log data acquisition unit that acquires log data including information on the operation of the mechatronics equipment, a designated log data acquisition unit that acquires designated log data from the log data that is necessary to reproduce the operating status, and a data conversion unit that converts the designated log data into a predetermined data file arranged in chronological order in order to reproduce the operating status of the designated log data. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram showing an example of a schematic configuration of a mechatronics equipment maintenance system according to the present embodiment. [Diagram 2]FIG. 2 is a block diagram showing an example of a hardware configuration of a mechatronics device and a software configuration associated with the hardware configuration. [Diagram 3] FIG. 3 is a block diagram showing an example of a hardware configuration of a central server and a software configuration associated with the hardware configuration. [Figure 4] FIG. 4 is a diagram illustrating an example of a log data generating operation by a mechatronics device. [Diagram 5] FIG. 5 is a diagram illustrating an example of a log data generating operation by a mechatronics device. [Figure 6] FIG. 6 is a diagram showing an example of log data when a paper transport operation is performed. [Figure 7] FIG. 7 is a flowchart showing an example of an animation data generating operation in the embodiment. [Figure 8] FIG. 8 is a diagram showing an example of a data file that can be read by the simulator. [Figure 9] FIG. 9 shows an example of an animation displayed on the maintenance staff terminal. [Figure 10] FIG. 10 is a block diagram showing an example of a hardware configuration of a central server and a software configuration associated with the hardware configuration. [Figure 11] FIG. 11 is a flowchart for creating a learning model using supervised data. [Figure 12] FIG. 12 is a block diagram showing an example of a hardware configuration of a central server and a software configuration associated with the hardware configuration. [Figure 13] FIG. 13 is a flowchart for creating a learning model using unsupervised learning data. [Figure 14] FIG. 14 is a diagram showing an example in which deviation degrees derived from log data of a given mechatronics device are arranged in chronological order. [Figure 15] FIG. 15 is a diagram showing an example of log data that is normally recorded when a paper transport operation is performed. [Figure 16] FIG. 16 is a diagram showing an example in which log data was not recorded normally when a paper transport operation was performed. [Figure 17] FIG. 17 is a diagram showing another example in which log data is not normally recorded when a paper transport operation is performed. [Figure 18] FIG. 18 is a block diagram showing an example of a hardware configuration of a central server and a software configuration associated with the hardware configuration. [Figure 19] FIG. 19 is a block diagram showing an example of a software configuration included in the data conversion unit. [Figure 20] FIG. 20 is a diagram showing an example of the transition of the sensor reaction state when a paper transport operation is performed. [Figure 21] FIG. 21 is a flowchart showing an example of a data file generation operation in the second modified example of the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, a mechatronics equipment maintenance system, a mechatronics equipment maintenance server, a mechatronics equipment maintenance method, and a mechatronics equipment maintenance program will be described in detail with reference to the drawings. In the following embodiments, parts with the same numbers perform the same operations, and will not be described again. For example, when there are multiple identical or similar elements, a common symbol may be used to describe each element without distinguishing between them, and a subnumber may be used in addition to the common symbol to describe each element with distinction between them.

[0011] [Embodiment] (composition) FIG. 1 is a diagram showing an example of a schematic configuration of a mechatronics equipment maintenance system according to the present embodiment. As shown in FIG. 1, the mechatronics equipment maintenance system includes a mechatronics equipment 1, a central server 2, and a maintenance technician terminal 3.

[0012] The mechatronics device 1 can be realized by one or more computers. The computer here also includes an embedded system including hardware such as a microcomputer chip, an LSI (Large Scale Integration), an ASIC (Application Specific Integrated Circuit), or an FPGA (Field-Programmable Gate Array). The mechatronics device 1 is any device whose internal mechanism operates under the control of control software. For example, the mechatronics device 1 includes station equipment including automatic ticket gates, copiers, printers, cameras, automobiles, air conditioners, industrial robots, automatic transport robots in warehouses, etc.

[0013] The mechatronics device 1 realizes its function by, for example, outputting appropriate operation instructions to actuators such as motors and solenoids in the mechanism at appropriate timing by the control software based on sensor information obtained from the mechanism. In addition, for the purpose of analyzing the operation and the cause of failure, the control software of the mechatronics device 1 may be configured to output information obtained from the sensors and instructions to the actuators together with the timing as log data to a storage device in the mechatronics device 1 and store the information in the storage device. The log data stored in the storage device is transferred to the central server 2 by a USB memory or the like. Alternatively, the mechatronics device 1 may be configured to be able to communicate with the central server 2 and output the log data to the central server 2. Here, the log data may include a record of abnormal operation such as a printer stopping due to a paper jam, information inspected or confirmed by a user by opening the cover of the mechatronics device 1, and the like. In the following description, for the sake of simplicity, the mechatronics device 1 is described as an automatic ticket gate.

[0014] The central server 2 can be realized by one or more computers, and operates as a mechatronics equipment maintenance server. The central server 2 can communicate with the mechatronics equipment 1 and the maintenance staff terminal 3. The central server 2 is a server that can monitor the operation status of the mechatronics equipment 1 based on the log data of the mechatronics equipment 1 and determine whether or not there is an abnormality. The central server 2 also generates CG animation data for reproducing the movement of the mechatronics equipment 1 based on the log data. Then, it communicates the generated animation data to the maintenance staff terminal 3.

[0015] The maintenance staff terminal 3 can be realized by one or more computers. The maintenance staff terminal 3 is placed, for example, in a room where a maintenance staff who maintains the mechatronics device 1 waits. The maintenance staff terminal 3 may be any terminal such as a stationary or portable computer, a mobile terminal, a tablet terminal, a wearable terminal, or the like. The maintenance staff terminal 3 displays an animation that reproduces the movement of the mechatronics device 1 from the animation data received from the central server 2. The animation can be run at, for example, any operating speed. This makes it possible to find defects that would not be noticeable at the actual speed.

[0016] The central server 2 may only convert the log data into a data file that can be converted into animation data. The central server 2 may then transmit the data to an external device (e.g., a cloud server) that can generate animation data based on the data file. The external device generates animation data from the data file and transmits the generated animation data to the maintenance staff terminal 3.

[0017] FIG. 2 is a block diagram showing an example of a hardware configuration of the mechatronics device 1 and a software configuration associated with the hardware configuration. The mechatronics device 1 includes a control unit 11, a program storage unit 12, a data storage unit 13, a communication interface 14, and an input / output interface 15. The control unit 11, the program storage unit 12, the data storage unit 13, the communication interface 14, and the input / output interface 15 are communicatively connected to one another via a bus, a serial interface, etc. The input / output interface 15 is also communicatively connected to an input device 151, an output device 152, etc.

[0018] The control unit 11 of the mechatronics device 1 has a function of controlling the operation of the entire mechatronics device 1. The control unit 11 may include an internal cache and various interfaces. The control unit 11 realizes various processes by executing programs stored in advance in the internal cache or the program storage unit 12. For example, the control unit 11 is realized by a CPU (Central Processing Unit). The control unit 11 may also be realized by hardware such as an LSI (Large Scale Integration), an ASIC (Application Specific Integrated Circuit), or an FPGA (Field-Programmable Gate Array).

[0019] The program storage unit 12 can use, as a storage medium, a combination of a non-volatile memory that can be written and read at any time, such as an EPROM (Erasable Programmable Read Only Memory), a HDD (Hard Disk Drive), or an SSD (Solid State Drive), and a non-volatile memory such as a ROM (Read Only Memory). The program storage unit 12 stores programs necessary for executing various processes. That is, the control unit 11 can realize various controls and operations by reading and executing the programs stored in the program storage unit 12.

[0020] The data storage unit 13 is a storage that uses a combination of a non-volatile memory that can be written to and read from at any time, such as a HDD or a memory card, and a volatile memory, such as a RAM (Random Access Memory), as a storage medium. The data storage unit 13 is used to store data acquired and generated in the process of the control unit 11 executing a program and performing various processes.

[0021] The communication interface 14 includes one or more wired or wireless communication modules. For example, the communication interface 14 includes a communication module that is wiredly connected to the central server 2 via a HUB or the like. The communication interface 14 may also include a communication module that is wirelessly connected to the central server 2. That is, the communication interface 14 may be a general communication interface as long as it can communicate with the central server 2 under the control of the control unit 11 and transmit and receive various information. Note that when the mechatronics device 1 is not connected to the central server 2 or the like, that is, when the mechatronics device 1 is a stand-alone device, the mechatronics device 1 does not need to include the communication interface 14.

[0022] The input / output interface 15 is connected to the input device 151, the output device 152, the mechanism 153, etc. The input / output interface 15 is an interface that enables transmission and reception of information between the input device 151, the output device 152, the mechanism 153, etc. The input / output interface 15 may be integrated with the communication interface 14. For example, the input / output interface 15 and at least one of the input device 151, the output device 152, the mechanism 153, etc. are wirelessly connected using a short-range wireless technology or the like, and transmission and reception of various information may be performed using the short-range wireless technology.

[0023] The input device 151 includes, for example, a keyboard, a pointing device, and the like, through which a maintenance person or an administrator who manages the mechatronics device 1 inputs various information to the mechatronics device 1. The input device 151 may also include a reader for reading data to be stored in the program storage unit 12 or the data storage unit 13 from a memory medium such as a USB memory, or a disk device for reading such data from a disk medium.

[0024] The output device 152 includes a display for displaying an image to be displayed by the mechatronics device 1 to a maintenance person or a manager, a printer for printing information displayed on the display, etc. The output device 152 may include a writer for writing information stored in the data storage unit 13 to a memory medium such as a USB memory, and a disk device.

[0025] Here, for safety reasons, the mechatronics device 1 may not be able to communicate with the central server 2. In such a case, the communication interface 14 may be omitted. For example, the information stored in the data storage unit 13 of the mechatronics device 1 may be written by a maintenance person or an administrator to a memory medium such as a USB memory or a disk medium by, for example, the output device 152, and copied to the central server 2.

[0026] The mechanism 153 includes a sensor 1531, a motor 1532, a solenoid 1533, and a state detection sensor 1534. The mechanism 153 may include any device other than the sensor 1531, the motor 1532, the solenoid 1533, and the state detection sensor 1534 that can acquire log data.

[0027] The sensor 1531 is a variety of sensors provided in the mechatronics device 1. The sensor 1531 starts operation based on a control signal from the control unit 11, and during operation, outputs a sensor value or the like to the control unit at the timing of a change in the sensor value or the like. The control unit 11 stores the sensor value together with the timing (time) at which it was obtained in the log data storage unit 131 described later.

[0028] The motor 1532 and the solenoid 1533 operate as actuators. The control unit 11 stores the instructions output to the motor 1532 and the solenoid 1533 together with the timing (time) of the instructions in the log data storage unit 131, which will be described later.

[0029] The state detection sensor 1534 is a sensor for detecting the state of the motor 1532, the solenoid 1533, etc., and detects the occurrence of an abnormality in these mechanisms 153. Then, the control unit 11 stores the detected information together with the detection timing (time) in the log data storage unit 131.

[0030] Next, the software configuration of the control unit 11 will be described in more detail. The control unit 11 includes a mechanism information acquisition unit 111 , a log data generation unit 112 , and a communication control unit 113 .

[0031] The mechanism information acquisition unit 111 acquires which of the devices included in the mechanism 153 has operated and the timing (time) at which the device operated.

[0032] The log data generating unit 112 generates log data that collects information indicating operations and timings acquired by the mechanism information acquiring unit 111 from a series of operations by each device of the mechanism 153. The log data generating unit 112 may generate log data while the mechatronics device 1 is operating. In other words, the log data may be generated in real time.

[0033] The communication control unit 113 controls the log data generated by the log data generation unit 112 to be transmitted to the central server 2 via the communication interface 14. When the mechatronics device 1 is a stand-alone device, the control unit 11 does not need to include the communication control unit 113. On the other hand, when the mechatronics device 1 is connected to the central server 2, the log data generated by the log data generation unit 112 may be transmitted to the central server 2 in real time.

[0034] Next, the software configuration of the data storage unit 13 will be described in more detail. The data storage unit 13 includes a log data storage unit 131 . The log data storage unit 131 is used to store the log data generated by the log data generation unit 112. In addition, the log data stored in the log data storage unit 131 may be copied to a memory medium such as a USB memory by an administrator who manages the mechatronics device 1 or an administrator who manages the central server 2.

[0035] FIG. 3 is a block diagram showing an example of a hardware configuration of the central server 2 and a software configuration associated with the hardware configuration. The central server 2 includes a control unit 21, a program storage unit 22, a data storage unit 23, a communication interface 24, and an input / output interface 25. The control unit 21, the program storage unit 22, the data storage unit 23, the communication interface 24, and the input / output interface 25 are communicatively connected to one another via a bus. The input / output interface 25 is also communicatively connected to an input device 251, an output device 252, and the like.

[0036] The control unit 21 of the central server 2 has a function of controlling the overall operation of the central server 2. The control unit 21 may include an internal cache and various interfaces. The control unit 21 realizes various processes by executing programs stored in advance in the internal cache or the program storage unit 22. For example, the control unit 21 is realized by a CPU. Note that the control unit 21 may also be realized by hardware such as an LSI, an ASIC, or an FPGA.

[0037] The program storage unit 22 can use, as a storage medium, a combination of a non-volatile memory such as an EPROM, HDD, SSD, etc., which can be written and read at any time, and a non-volatile memory such as a ROM. The program storage unit 22 stores programs necessary for executing various processes. That is, the control unit 21 can realize various controls and operations by reading and executing the programs stored in the program storage unit 22.

[0038] The data storage unit 23 is a storage medium that uses a combination of a non-volatile memory that can be written to and read from at any time, such as a HDD or a memory card, and a volatile memory, such as a RAM. The data storage unit 23 is used to store data acquired and generated in the process of the control unit 21 executing a program and performing various processes.

[0039] The communication interface 24 includes one or more wired or wireless communication modules. For example, the communication interface 24 includes a communication module that connects to the mechatronics device 1 and the maintenance staff terminal 3 by wire via a HUB or the like. The communication interface 24 may also include a communication module that connects wirelessly to at least one of the mechatronics device 1 and the maintenance staff terminal 3. In other words, the communication interface 24 may be a general communication interface as long as it can communicate with the central server 2 under the control of the control unit 21 and transmit and receive various information.

[0040] The input / output interface 25 is connected to the input device 251, the output device 252, etc. The input / output interface 25 is an interface that enables transmission and reception of information between the input device 251, the output device 252, etc. The input / output interface 25 may be integrated with the communication interface 24. For example, the input / output interface 25 and at least one of the input device 251, the output device 252, etc. are wirelessly connected using a short-range wireless technology or the like, and transmission and reception of various information may be performed using the short-range wireless technology.

[0041] The input device 251 includes, for example, a keyboard, a pointing device, and the like, which are used by an administrator who manages the central server 2 to input various information to the central server 2. The input device 251 may also include a reader for reading data to be stored in the program storage unit 22 or the data storage unit 23 from a memory medium such as a USB memory, or a disk device for reading such data from a disk medium.

[0042] The output device 252 includes a display for displaying images that the central server 2 should display to the administrator, and a printer for printing information displayed on the display.

[0043] Next, the software configuration of the control unit 21 will be described in more detail. The control unit 21 includes a log data acquisition unit 211 , an operating status selection unit 212 , a data conversion unit 213 , an animation data generation unit 214 , and a communication control unit 215 .

[0044] The log data acquisition unit 211 acquires log data. The log data acquisition unit 211 acquires log data transmitted from the mechatronics device 1 or stored in a memory medium such as a USB memory.

[0045] The operating status selecting unit 212 includes an event selecting unit 2121 and a user-specified log data acquiring unit 2122. The operating status selecting unit 212 is a selecting unit that selects log data to be converted into animation data for displaying animation on the maintenance staff terminal 3.

[0046] The event selection unit 2121 displays a list of events on the display of the output device 252. The administrator (user) of the central server 2 selects one of the events from the list of events. The event selection unit 2121 acquires information on which event the administrator has selected (event selection information). Here, the list of events displayed on the display includes, for example, in the case of an automatic ticket gate, events such as a ticket jam, a magnetic ticket reading error, a separation error when multiple tickets are inserted, and a station attendant opening the cover to check. Here, it is assumed that the list of events can be set arbitrarily depending on the type of mechatronics device 1.

[0047] The user-specified log data acquisition unit 2122 acquires user-specified log data. The user-specified log data acquisition unit 2122, which operates as a display control unit, displays the log data narrowed down by the event selection information and its attribute value. Here, the attribute value indicates to which event the log data belongs. In addition, the attribute value may also include data such as the date and time when the log data was acquired. This makes it possible to acquire data such as data from a time period when abnormalities frequently occurred. Then, the user-specified log data acquisition unit 2122 acquires information on which log data the administrator (user) designated (i.e., user-specified log data). For example, the user-specified log data is log data that is required to reproduce the operating status of the mechatronics device 1.

[0048] The data conversion unit 213 converts the user-specified log data into a data file. The data conversion unit 213 converts the log data into a data file that can be read by software (simulator) that converts the log data into animation data. The method of converting into a data file will be described in detail later. For example, the data file has a predetermined data format in which the user-specified log data is arranged in chronological order.

[0049] The animation data generating unit 214 converts the data file into animation data that can be used in the mechanism simulation software. A general method may be used to convert the data file into animation data. The animation data generating unit 214 may store the animation data and the data file used to generate the animation data in the converted data storage unit 233.

[0050] The communication control unit 215 transmits the animation data to the maintenance staff terminal 3. The control unit of the maintenance staff terminal 3 that has received the animation data performs control so as to display an animation on the display from the animation data using the above-mentioned mechanism simulation software.

[0051] The data storage unit 23 includes an event selection information storage unit 231 , a user-specified data storage unit 232 , and a converted data storage unit 233 .

[0052] The event selection information storage unit 231 is used to store event selection information selected by an administrator (user). The user-specified data storage unit 232 is used to store user-specified data indicating which log data is specified by the administrator (user). The conversion data storage unit 233 is used to store animation data and data files used to generate the animation data.

[0053] (operation) (1) Log data generation operation First, a method for generating log data generated by the operation of the mechatronics device 1 in this embodiment will be described. In this embodiment, for simplicity, the log data generation operation will be described using an example of a paper transport mechanism. That is, in the following description, the operation of the mechatronics device 1 is assumed to be the operation of the paper medium proceeding along the transport path, and log data is generated by the paper medium passing a sensor installed on the transport path.

[0054] 4 and 5 are diagrams showing an example of the log data generating operation by the mechatronics device 1. In FIG. 4 and 5, paper medium P advances along a transport path C (indicated by dotted lines in the examples of FIGS. 4 and 5) sandwiched between belts. In addition, a first photosensor S1, a second photosensor S2, and a third photosensor S3 are installed on transport path C. These sensors are off while light is reaching them, and as paper medium P advances along transport path C, the light is blocked and the sensors turn on, and when the paper medium has passed, light reaches them again and the sensors turn off. Then, when the paper medium has finished passing through transport path C, the series of operations in mechatronics device 1 is completed.

[0055] Figure 5 shows the state of the paper medium at the moment each sensor turns on: Figure 5(a) shows the moment the first photosensor S1 turns on, Figure 5(b) shows the moment the second photosensor S2 turns on, and Figure 5(c) shows the moment the third photosensor S3 turns on.

[0056] For example, the mechanism information acquisition unit 111 acquires the timing (time) when each sensor is turned on. That is, the mechanism information acquisition unit 111 acquires the time when the device operated in addition to the information indicating that the sensor included in the mechanism 153 operated. The log data generation unit 112 generates log data including the timing when each sensor was turned on, etc. That is, the log data generation unit 112 generates log data that collects information indicating the operation and timing acquired by the mechanism information acquisition unit 111 from a series of operations by each device of the mechanism 153.

[0057] FIG. 6 is a diagram showing an example of log data when a paper transport operation is performed. 6 indicates that the first photosensor S1 was turned on at time 29628 (e.g., the time elapsed since the controller 11 was started, expressed in milliseconds). Similarly, the second photosensor S2 was turned on at time 29634, and the third photosensor S3 was turned on at time 29650.

[0058] In the example of FIG. 6, the log data generating unit 112 generates log data that collects information on the turning on of each photo sensor (information indicating an operation) and the timing of the operation. Then, the log data generating unit 112 outputs the generated log data to the communication control unit 113 or stores the generated log data in the log data storage unit 131. As described above, for safety reasons of the mechatronics device 1, there are cases where the device is not connected to the central server 2. In such a case, the log data generating unit 112 stores the generated log data in the log data storage unit 131. On the other hand, when the mechatronics device 1 can communicate with the central server 2, the log data generating unit 112 outputs the generated log data to the communication control unit 113. The communication control unit 113 transmits the log data to the central server 2 through the communication interface 14. Furthermore, the log data generating unit 112 may output the log data to the communication control unit 113 and store it in the log data storage unit 131.

[0059] The above-described log data generation operation is merely an example, and the mechanism information acquisition unit 111 may acquire mechanism information from the sensor 1531, the motor 1532, the solenoid 1533, the state detection sensor 1534, etc., included in the mechanism 153. Then, the log data generation unit 112 may generate log data based on the acquired mechanism information.

[0060] (2) Animation data generation Next, an example of an animation data generating operation performed by the central server 2 in this embodiment will be described. The central server 2 may generate a data file capable of generating animation data from log data, instead of generating the animation data itself. An external device different from the central server 2 may generate the animation data based on the data file. FIG. 7 is a flowchart showing an example of an animation data generating operation in the embodiment. The operation of this flowchart is realized by reading out and executing a program stored in the internal cache of the control unit 21 of the central server 2 or in the program storage unit 12 .

[0061] First, the mechatronics device 1 generates log data by the above-mentioned operation and transmits the log data to the central server 2, or the flow chart is started when the administrator writes the log data to a memory medium such as a USB memory and inserts the memory medium such as a USB memory into the input device 251 of the central server 2. Note that, as described above, when the mechatronics device 1 is connected to the central server 2, the log data may be generated in real time and transmitted to the central server 2.

[0062] In step ST101, the log data acquisition unit 211 acquires log data. The log data acquisition unit 211 acquires log data transmitted from the mechatronics device 1 or stored in a memory medium such as a USB memory.

[0063] Here, log data usually includes a large amount of operation logs. Therefore, the administrator needs to select operation data to be converted into animation data from the log data. That is, it is necessary to narrow down the log data to the log data that the user thinks is necessary. Steps ST102 and ST103 described below are the operations of this narrowing down.

[0064] In step ST102, the event selection unit 2121 acquires event selection information. For example, the event selection unit 2121 operating as a display control unit causes a list of events to be displayed on the display of the output device 252. Then, the event selection unit 2121 acquires information on which event the administrator has selected (event selection information). The operation of step ST102 may be performed before acquiring log data. That is, the administrator may select the event selection information in advance. The selected event selection information is stored in the event selection information storage unit 231. Then, the event selection unit 2121 may read out the event selection information from the event selection information storage unit 231.

[0065] In step ST103, the user-specified log data acquisition unit 2122 acquires user-specified log data. The user-specified log data acquisition unit 2122 operating as a display control unit displays the log data narrowed down by the event selection information and its attribute value. Here, the attribute value indicates what event the log data belongs to. Then, the user-specified log data acquisition unit 2122 acquires user-specified log data based on user-specified data indicating which log data the administrator (user) has specified. Note that the operation of step ST103 may be performed before acquiring log data. For example, the user specifies a condition (user-specified data) for which log data to specify, and stores the specified data in the user-specified data storage unit 232. Then, the user-specified log data acquisition unit 2122 reads out the user-specified data from the user-specified data storage unit 232, and acquires the user-specified log data from the log data based on the specified data.

[0066] In step ST104, the data conversion unit 213 converts the user-specified log data into a data file that can be read by software (simulator) that converts the log data into animation data.

[0067] FIG. 8 is a diagram showing an example of a data file that can be read by the simulator. Fig. 8 shows an example in which log data of paper media P, paper1 and paper2, passing in order on a transport path C is converted into a data file. The top row of the table in Fig. 8 shows the time, the next row shows a flag indicating whether paper1 is on the transport path (True if it is on the transport path, False if it is not), the next row shows the position of paper1, the next row shows a flag indicating whether paper2 is on the transport path, and the last row shows the position of paper2. The values ​​103, 230, and 320 shown in the rows for the positions of paper1 and paper2 respectively indicate the positions of the first photosensor S1, the second photosensor S2, and the third photosensor S3.

[0068] 8, paper1 is on the transport path because it is True at times 29628, 29634, and 29650, and has passed the first photosensor S1, the second photosensor S2, and the third photosensor S3 at the respective times. Paper2 is on the transport path because it is True at times 30000 and 30521, and has passed the first photosensor S1 and the second photosensor S2 at the respective times.

[0069] 7, in step ST105, the animation data generating unit 214 converts the data file into animation data that can be used in the mechanism simulation software. Then, the animation data generating unit 214 may store the animation data and the data file used to generate the animation data in the converted data storage unit 233. Here, the mechanism simulation software may be, for example, a general product such as XVL kinematics suite or XVL Vmech simulator (https: / / www.lattice.co.jp / products / lineup / vmech / ) provided by Lattice Technology. Furthermore, as a technique for realizing such a mechanism simulator, for example, descriptions in patent documents (Japanese Patent No. 3940665, Japanese Patent No. 4564255, Japanese Patent No. 3866595), academic papers (K. Kondo and M. Yoshida: Use of hybrid models for testing and debugging control software for electromechanical systems, IEEE / ASME Trans. Mechatronics, Vol. 10, No. 3, 275-284 (2005 / 7)), etc. may be used. Such a simulator has a function of registering the states of the mechanism as point-by-point points, and when the time when each state occurs is specified, interpolating between those states to operate the mechanism and outputting the animation data. For example, the animation data generating unit 214 generates animation data collectively for the data file by batch processing.

[0070] In step ST106, the communication control unit 215 transmits the animation data to the maintenance staff terminal 3. The control unit of the maintenance staff terminal 3 that has received the animation data performs control so as to display an animation on the display from the animation data using the above-mentioned mechanism simulation software. Note that, as described above, when log data is received in real time, the animation displayed on the display of the maintenance staff terminal 3 can also be a real-time animation.

[0071] FIG. 9 shows an example of an animation displayed on the maintenance staff terminal. FIG. 9 shows an example of a printer as the mechatronics device 1. However, for example, in the case of maintenance of an automatic ticket gate, this diagram shows the mechanism of the automatic ticket gate. For example, the animation shown on the display can reproduce the behavior of the user-specified log data to be confirmed at any speed, such as slow playback, frame-by-frame playback, and pause. Furthermore, by using animation, maintenance personnel can check the behavior from a more flexible perspective, such as an angle that is easy to see, or the behavior of the operation when a cover is closed. Furthermore, maintenance personnel can take action while repeatedly watching the video. In addition, it is possible to leave a single event as a maintenance example (know-how), which can be utilized not only for current maintenance personnel but also for future maintenance personnel.

[0072] By displaying an animation created from log data on the maintenance personnel's terminal, the behavior of a failure can be reproduced 100% on the digital twin, enabling the maintenance personnel to take appropriate action according to the event.

[0073] The operations of steps ST105 and ST106 may not be performed by the central server 2, but may be performed by an external device connected to the central server 2. In this case, after the process of step ST104 is completed, the communication control unit 215 transmits the data file to the external device.

[0074] As described above, necessary data (user-specified log data) can be extracted from the log data, and CBM (Condition Based Maintenance) can be realized, in which maintenance is performed according to the state of the mechatronics device 1.

[0075] (Effects of the embodiment) According to the embodiment described above, the central server 2 creates a data file for converting the log data created by the mechatronics device 1 into animation. The central server 2 then converts the data file into animation data. Then, an animation using the animation data is displayed on a display, allowing a maintenance worker to determine whether or not maintenance is necessary. Also, by displaying the animation, it is possible to express operations that may not be reproducible on-site.

[0076] [First Modification of the Embodiment] In the embodiment, in order to extract user-specified log data, the administrator (user) must select an event and manually specify log data that is deemed necessary. However, the amount of log data is huge, and there is a problem that it takes too much time for the user to select the user-specified log data. Therefore, in the first modified example of the embodiment, a method of automatically specifying user-specified log data using knowledge engineering or machine learning will be described.

[0077] When introducing the technology of knowledge engineering or machine learning, it is necessary to define various feature quantities for the log data. The central server 2 calculates feature quantities for a large number of log data, and performs learning by various machine learning methods using the calculated data. The central server 2 then uses the trained model obtained as a result of the learning to evaluate each of the obtained log data, and uses the results to determine whether or not to make it a candidate for animation generation (i.e., user-specified log data). As feature quantities for knowledge engineering or machine learning, for example, in the case of an automatic ticket gate, the presence or absence of events such as ticket jamming, magnetic reading errors, separation errors when multiple tickets are inserted, and station staff opening the cover to check the inside can be considered. In addition, for the purpose of responding to a situation in which the operating speed of the mechanism 153 decreases due to deterioration, the feature quantity may be any value that can be calculated from the speed of the target object when each mechanism 153 of the mechatronics device 1 operates, the speed at which the mechanism 153 operates, the angular velocity, the acceleration, and the like.

[0078] For example, referring to FIG. 5, if transport is smooth in the section from the first photosensor S1 to the second photosensor S2, but the speed is slightly slower in the section from the second photosensor S2 to the third photosensor S3, it may be possible to detect signs of malfunction in the operation of mechanism 153 by calculating the ratio of the elapsed time for each section. (1) Automation through knowledge engineering First, an example of a method for automatically specifying user-specified log data by knowledge engineering will be described.

[0079] (composition) The hardware configuration of the mechatronics device 1 and the central server 2 and the software configuration associated with the hardware configuration are the same as those in the embodiment, and therefore a duplicated description will be omitted.

[0080] (operation) It is assumed that the above-mentioned parameters such as the presence or absence of an event or the ratio of elapsed time are organized under conditions such as when a parameter reaches a certain value or when multiple conditions are combined, maintenance and inspection should be dispatched to the site based on the experience of on-site maintenance personnel. Specifically, there are rules for determining the relationship between a certain value and a reference value using an inequality, rules for determining whether the number of occurrences of a specific event exceeds a reference value, and rules for detecting when either of these conditions occurs, or when they occur simultaneously. It is assumed that these rules are stored in, for example, the data storage unit 23.

[0081] The basic operation is similar to that described with reference to Fig. 7. However, instead of steps ST102 and ST103, the operation described below is carried out.

[0082] The operating status selection unit 212 calculates the above-mentioned feature amount from the log data obtained from the log data acquisition unit 211. Then, the operating status selection unit 212 judges whether or not the feature amount satisfies a rule (a predefined rule) stored in the data storage unit 23. If the condition is satisfied, the operating status selection unit 212 outputs the log data to the data conversion unit 213 as user-specified log data (i.e., as data to be animated).

[0083] (2) Automation through supervised machine learning Next, we will explain an example of automation through machine learning using supervised data. Assume that past failure history and log data from the time of failure and a certain period immediately preceding it are managed in association with each other. In such a case, it is possible to collect log data corresponding to each failure mode of interest (for example, ticket jamming, magnetic reading errors, separation errors when multiple tickets are inserted, etc.). In supervised data, learning is performed using this log data. (composition) The hardware configuration of the mechatronics device 1 and the software configuration associated with the hardware configuration are the same as those in the embodiment, and therefore a duplicated description will be omitted.

[0084] FIG. 10 is a block diagram showing an example of a hardware configuration of the central server 2 and a software configuration associated with the hardware configuration. The hardware configuration of the central server 2 is the same as that in the embodiment, so a duplicated description will be omitted.

[0085] This differs from the embodiment in that the operating status selection unit 212 of the control unit 21 includes a teacher data creation unit 2123, a learning unit 2124, and a discrimination unit 2125, and the data storage unit 23 includes a log data database 234, a fault record database 235, and a learned model storage unit 236.

[0086] The teacher data creation unit 2123 acquires failure information from a failure record database 235 described later, and acquires log data corresponding to the failure information from a log data database 234 described later. Then, the teacher data creation unit 2123 creates log data having the failure content included in the failure information as a label, as teacher data.

[0087] The learning unit 2124 performs a learning process using the teacher data created by the teacher data creation unit 2123, and creates a learning model (determines parameters). Note that the method of learning the supervised data is not limited to a specific method, and any method such as decision tree learning (e.g., ID3, C4.5, random forest, etc.), clustering, support vector machine, deep learning, etc. may be used.

[0088] The discrimination unit 2125 extracts user-specified log data from the log data by using the learning model created by the learning unit 2124. For example, instead of acquiring user-specified log data based on event selection information and user-specified data as described in the embodiment, the discrimination unit 2125 automatically acquires user-specified log data from the log data acquired by the log data acquisition unit 211.

[0089] The log data database 234 is a database that accumulates past log data. It is assumed that the log data stored in the log data database 234 is a sufficient number of log data for creating a learning model.

[0090] The fault record database 235 is used to store past fault records. For example, the fault record includes information such as a fault description indicating what kind of fault occurred and when the fault occurred.

[0091] (operation) Next, a method for automatically extracting user-specified log data by the operational status selection unit 212 using supervised learning will be described. First, before the operation described below is performed, it is assumed that the log data acquisition unit 211 has stored (accumulated) the log data in the log data database 234. It is also assumed that past failure records have been stored in the failure record database 235.

[0092] FIG. 11 is a flowchart for creating a learning model using supervised data. The operation of this flowchart is realized by reading and executing a program stored in the internal cache of the control unit 21 of the central server 2 or in the program storage unit 12. This flowchart is started in accordance with, for example, an instruction from an administrator (user).

[0093] In step ST201, the teacher data creation unit 2123 creates teacher data. The teacher data creation unit 2123 acquires failure information from the failure record database 235, and acquires log data corresponding to the failure information from the log data database 234. Then, the teacher data creation unit 2123 creates log data, which has the failure content included in the failure information as a label, as teacher data.

[0094] In step ST202, the learning unit 2124 creates a learning model. The learning unit 2124 performs a learning process using the teacher data created in step ST201, and creates a learning model (determines parameters). As described above, the method of learning the teacher data may be a general method, and is not limited to a specific method.

[0095] In step ST203, the learning unit 2124 stores the learning model in the learned model storage unit 236.

[0096] For example, instead of step ST102 and step ST103 described with reference to FIG. 7, the discrimination unit 2125 extracts user-specified log data from the log data by using the learning model learned by the above processing.

[0097] (2) Automation through unsupervised machine learning Finally, we will explain an example of automation through machine learning using unsupervised data. When past failure history and log data are not managed in an associated manner, or when failures occur infrequently and there is insufficient accumulated data, unsupervised learning techniques are used. Unsupervised learning is used for data that does not have a history label. It is said to be a technique in which the algorithm itself identifies the meaning of the data, explores the data, and finds some kind of structure within it. When applying unsupervised learning to support maintenance work, as much operational data as possible is collected from the period when the target mechatronics equipment 1 is thought to be operating normally, and learning is performed using that data to generate a learning model.

[0098] (composition) The hardware configuration of the mechatronics device 1 and the software configuration associated with the hardware configuration are the same as those in the embodiment, and therefore a duplicated description will be omitted.

[0099] FIG. 12 is a block diagram showing an example of a hardware configuration of the central server 2 and a software configuration associated with the hardware configuration. The hardware configuration of the central server 2 is the same as that in the embodiment, so a duplicated description will be omitted.

[0100] The operating status selection unit 212 of the control unit 21 differs from the embodiment in that it includes a normal period extraction unit 2126, a log data extraction unit 2127, a learning unit 2128, and a discrimination unit 2129.

[0101] The normal period extraction unit 2126 extracts a period that does not correspond to any fault record recorded in the fault record database 235, that is, a period during which the mechatronics device 1 was operating normally (normal period).

[0102] The log data extraction unit 2127 extracts log data for a period during which the mechatronics device 1 was operating normally (normal period) from the log data database 234, and creates a set of log data.

[0103] The learning unit 2128 creates a learning model. The learning unit 2128 creates a learning model (determines parameters) using a collection of log data for a normal period. The method of creating an unsupervised learning model is not limited to a specific method and may be any method, such as a k-nearest-neighbor method, a one-class-support-vector-machine, a classification method based on Mahalanobis distance, or a deep learning Auto-Encoder. The discrimination unit 2129 extracts user-specified log data from the log data by using the learning model learned by the learning unit 2128. For example, instead of acquiring user-specified log data based on event selection information and user-specified data as described in the embodiment, the discrimination unit 2129 automatically acquires user-specified log data from the log data acquired by the log data acquisition unit 211.

[0104] (operation) Next, a method for automatically extracting user-specified log data by the operational status selection unit 212 using unsupervised learning will be described. First, before the operation described below is performed, it is assumed that log data is stored (accumulated) in the log data database 234 by the log data acquisition unit 211. It is also assumed that past failure records are stored in the failure record database 235. Here, it is assumed that the number of failure records stored in the failure record database 235 is too small to perform supervised learning.

[0105] FIG. 13 is a flowchart for creating a learning model using unsupervised learning data. The operation of this flowchart is realized by reading and executing a program stored in the internal cache of the control unit 21 of the central server 2 or in the program storage unit 12. This flowchart is started in accordance with, for example, an instruction from an administrator (user).

[0106] In step ST301, the normal period extraction unit 2126 extracts a normal period. The normal period extraction unit 2126 extracts a period that does not correspond to a failure record recorded in the failure record database 235, that is, a period during which the mechatronics device 1 was operating normally (normal period).

[0107] In step ST302, the log data extraction unit 2127 extracts log data for a normal period. For example, the log data extraction unit 2127 extracts log data for a period during which the mechatronics device 1 was operating normally (normal period) from the log data database 234, and creates a set of log data.

[0108] In step ST303, the learning unit 2128 creates a learning model. The learning unit 2128 creates the learning model (determines parameters) using a set of log data for a normal period. The method of creating the unsupervised learning model may be a general method as described above, and is not limited to a specific method.

[0109] In step ST304, the learning unit 2128 stores the trained learning model in the trained model storage unit 236.

[0110] For example, instead of step ST102 and step ST103 described in FIG. 7, the discrimination unit 2129 extracts user-specified log data from log data using the learning model learned by the above processing. For example, the learning model obtained by unsupervised learning outputs a smaller value for data that is closer to the learned data set, and outputs a larger value for data that deviates from the learning data set. Therefore, the discrimination unit 2129 acquires a deviation from the learned model (a certain type of learning data set). For example, in the case of log data that is operating normally, the deviation output using the learned model is small. On the other hand, in the case of log data that is not operating normally (different from normal), the deviation output using the learned model is large.

[0111] FIG. 14 is a diagram showing an example of a table in which the deviations of log data of a predetermined mechatronics device 1 are arranged in chronological order. The X-axis in Fig. 14 represents the date and time when the log data was created (for simplicity, all are shown as 20XX-XX in Fig. 14), and the Y-axis represents the degree of deviation. As shown in Fig. 14, the degree of deviation is high when there is dirt in the XX part and when there is an abnormality in YY. For example, when the degree of deviation is higher than a predetermined threshold, the discrimination unit 2129 determines that there is a high possibility that an abnormal operation is occurring in the mechatronics device 1, and therefore sets such log data as user-specified log data. In other words, it picks up such log data as a target for inspection.

[0112] The learning operation described above with reference to Fig. 11 and Fig. 13 may be performed periodically. For example, it may be performed again when a certain amount of logs have been accumulated. In this case, the newly learned learning model may be stored in the learned model storage unit 236.

[0113] (Functions and Effects of the First Modification of the Embodiment) According to the modified example of the embodiment described above, the operating status selection unit 212 automatically extracts user-specified log data using a trained model. This eliminates the need for the administrator (user) to select user-specified log data, thereby reducing the administrator's workload.

[0114] Furthermore, it is possible to realize CBM that automatically extracts necessary data (user-specified log data) from the log data and executes maintenance according to the state of the mechatronics device 1.

[0115] [Second Modification of the Embodiment] When the mechanism 153 of the mechatronics device 1 does not operate normally and operates in an abnormal state, the log data becomes different from normal, that is, the log data has low reliability. First, a problem that occurs when such log data has low reliability will be described. In this embodiment, it is assumed that log data is generated by the mechatronics device 1 as described with reference to Figures 4 and 5. In the following description, the log data refers to the log data when the paper conveying operation of Figures 4 and 5 is performed.

[0116] FIG. 15 is a diagram showing an example of log data that is normally recorded when a paper transport operation is performed. 15 indicates the time elapsed in milliseconds since control unit 11 was started. The part surrounded by [_ _] indicates which mechanism 153 operated, and the right side indicates the operation result of each mechanism 153 or an instruction from control unit 11.

[0117] For example, at time 57502, the motor M1 (not shown in FIGS. 4 and 5) that operates the belt of the transport path C is turned on by an instruction output from the control unit 11 to the motor M1. As a result, the paper medium starts to move on the belt on the transport path C.

[0118] Furthermore, at time 57512, the first photosensor S1 is on, and the second photosensor S2 and the third photosensor S3 are off. Similarly, at time 57522, the first photosensor S1 is off, the second photosensor S2 is on, and the third photosensor S3 is off. At time 57532, the first photosensor S1 is off, the second photosensor S2 is off, and the third photosensor S3 is on. At time 57542, all the photosensors are off.

[0119] Furthermore, at time 57523, the control unit 11 determines the operating status based on information obtained from the photosensor, and indicates that a predefined status has been reached. At time 57524, an instruction is given to the solenoid L1 (not shown in FIGS. 4 and 5) that switches the transport path C. At time 57553, the control unit 11 determines the size of the paper medium that has been inserted.

[0120] These operations occur at arbitrary timings without being synchronized with each other in the operating conditions that change from moment to moment. Therefore, depending on the load condition of the control unit 11 and the operating environment surrounding the mechatronics device 1, when the operations are recorded as log data, there may be inconsistencies in the recording order, or errors or omissions may occur in the recorded information. In addition, if any of the mechanisms 153 such as the sensor 1531 malfunctions, an abnormal operation is recorded at a timing different from the normal state.

[0121] FIG. 16 is a diagram showing an example in which log data was not recorded normally when a paper transport operation was performed. The example in Fig. 16 shows an example where the solenoid L1 is recorded in an abnormal state. Compared to Fig. 15, the example in Fig. 16 shows that the instruction to the solenoid L1 acting to switch the transport path C at time 57524 is not "L1 OFF" but is erroneously recorded as meaningless data.

[0122] FIG. 17 is a diagram showing another example in which log data is not normally recorded when a paper transport operation is performed. The example of Fig. 17 shows an example where the third photosensor S3 is recorded in an abnormal state. Compared with Fig. 15, the example of Fig. 19 shows that at time 57522, the first photosensor S1 is off, the second photosensor S2 is on, and the third photosensor S3 is on. That is, it can be seen that the timing at which the third photosensor S3 turns on is early. Also, at time 57542, the third photosensor S3 continues to be on, indicating that an abnormal operation in which the third photosensor S3 operates at a timing different from the normal state is recorded.

[0123] When converting the log data recorded under such a condition with low reliability into a data file that reproduces the operating status, it is necessary to implement exception processing for abnormal operations in the central server 2 in addition to the main conversion processing for normal operations. That is, it is necessary to implement exception processing in the central server 2 each time according to the variation of the abnormal operation, which is very costly. Furthermore, there is a problem that the quality of the reproduction of the operating status obtained as a result deteriorates, such as failure to switch the transport route C, and the reproduction may result in an operation that is significantly different from the operation of the mechatronics device 1 at the time the log was generated.

[0124] In a second variant of the embodiment, a method is described for converting log data containing information on the operation of the mechatronics equipment 1 into data that can reproduce the operating status with high quality, even if the log data contains low-reliability data, when reproducing the operating status of the mechatronics equipment 1 from the log data.

[0125] (composition) The hardware configuration of the mechatronics device 1 and the software configuration associated with the hardware configuration are the same as those in the embodiment, and therefore a duplicated description will be omitted.

[0126] FIG. 18 is a block diagram showing an example of a hardware configuration of the central server 2 and a software configuration associated with the hardware configuration. The hardware configuration of the central server 2 is the same as that in the embodiment, so a duplicated description will be omitted.

[0127] The second modified example of the embodiment differs from the embodiment in that the data storage unit 23 of the central server 2 further includes a log data database 234. In addition, as will be described later, the second modified example of the embodiment also differs from the embodiment in that the data conversion unit 213 further includes a software configuration.

[0128] The log data database 234 is a database that accumulates past log data, similar to the above. When the mechatronics device 1 and the central server 2 are connected, the log data stored in the log data database 234 may be updated at any time or at regular intervals. Alternatively, even when the mechatronics device 1 and the central server 2 are not connected, the log data may be updated at regular intervals. Note that the log data database 234 in this embodiment stores a sufficient amount of log data for calculating the reliability of the log data described below.

[0129] FIG. 19 is a block diagram showing an example of a software configuration included in data conversion unit 213. As shown in FIG. The data conversion unit 213 includes a log data reliability calculation unit 2131, a log data sorting unit 2132, and a behavior data generation unit 2133. The behavior data generation unit 2133 further includes a summary behavior data generation unit 21331 and a detailed behavior data generation unit 21332.

[0130] The log data reliability calculation unit 2131 is a calculation unit that calculates log data reliability from the log data stored in the log data database 234. For example, the log data reliability calculation unit 2131 calculates the occurrence probability of what state will transition from the log data, and generates the log data reliability based on the occurrence probability. Note that a detailed method of generating the log data reliability will be described later.

[0131] The log data sorting unit 2132 sorts each line of the user-specified log data into data with high reliability that is normally recorded and data with low reliability due to the occurrence of an abnormality, based on the user-specified log data and the log data reliability. Furthermore, the log data sorting unit 2132 extracts data of the sensor 1531 from the user-specified log data to generate user-specified sensor data. Then, the log data sorting unit 2132 generates sensor reliability information, which is the sorting result corresponding to the user-specified sensor data. Here, the sensor reliability information is information indicating the reliability of the behavior of each sensor in the user-specified log data.

[0132] The behavior data generating unit 2133 converts the user-specified log data into a data file in such a way that the influence of data with low reliability is smaller than that of data with high reliability.

[0133] The general behavior data generating unit 21331 generates general behavior data based on the user-specified sensor data and the sensor reliability information. For example, the general behavior data generating unit 21331 refers to the sensor reliability information for the acquired user-specified sensor data, and generates general behavior data that reduces the influence of data with low reliability. The details of the generation method of the general behavior data generating unit will be described later.

[0134] The detailed behavior data generating unit 21332 generates a data file. The detailed behavior data generating unit 21332 converts the general behavior data and the user-specified data (excluding the sensor data) into a data file. The method of converting into a data file will be described later in detail.

[0135] (operation) (1) Reliability calculation operation First, the operation of calculating the reliability of the log data will be described by taking as an example the transition of the sensor reaction state of the photosensor during the above-mentioned paper transport operation.

[0136] FIG. 20 is a diagram showing an example of the transition of the sensor reaction state when a paper transport operation is performed. A set of reaction results of the first photosensor S1, the second photosensor S2, and the third photosensor S3 recorded in one line of the log data is regarded as a state (st1, st2, st3, st4, st31 in FIG. 20), and the state transition in the time series of the log data is shown. For example, in the case of normal operation recorded as in FIG. 15, the state transitions are t1, t2, t3, and t4. In general, such state transitions due to normal operation account for the majority of the entire paper transport. Therefore, the state transition probability of this state transition is high. On the other hand, in the case of abnormal operation recorded as in FIG. 17, the state transitions are t1, t21, t31, and t41. Such state transitions due to abnormal operation occur less frequently in the entire paper transport. Therefore, the state transition probability of this state transition is low.

[0137] 20, when there are multiple state transition destinations, comparing st3 reached via a transition to t2 with a high state transition probability with st31 reached via a transition to t21 with a low state transition probability, there is a difference in the value of the third photosensor S3. In other words, it can be determined that the reaction of the third photosensor S3 is the cause of the transition to t21 with a low state transition probability, and that the reliability of the operation of the third photosensor S3 has decreased.

[0138] By applying the concept of occurrence probability, reliability can be evaluated not only when the sensor operation is not normal, but also when the log data is not recorded normally. In other words, the instructions to the solenoid L1 that acts to switch the transport path, "L1 OFF" shown in Figure 15, and its opposite instruction, "L1 ON", account for the majority of the overall paper transport, and their occurrence probability is high. On the other hand, there are few opportunities for instructions based on meaningless data, as shown in Figure 16, and their occurrence probability is low. Therefore, information indicating operations such as those in Figure 16 can be judged to have low reliability.

[0139] For example, the log data database 234 stores log data for a certain period or the entire period, as described above. The log data reliability calculation unit 2131 calculates the occurrence probability of information indicating an operation from the log data stored in the log data database 234. Here, the information indicating an operation includes a change in the value of the sensor 1531 provided in the mechanism 153 in the mechatronics device 1, an instruction output by the control software to the motor 1532 and the solenoid 1533 operating as actuators, a change in state by the state detection sensor 1534, and the like. The information indicating an operation also includes a result of the control unit 11 determining the operating status based on information obtained from the mechanism 153 including the sensor 1531, and further includes an error operation occurring during operation.

[0140] The log data reliability calculation unit 2131 calculates the occurrence probability itself as the reliability. For example, if the occurrence probability is higher than a predetermined threshold, the log data reliability calculation unit 2131 determines that the reliability of the log data is high, and if the occurrence probability is equal to or lower than the predetermined threshold, the log data reliability calculation unit 2131 determines that the reliability of the log data is low.

[0141] Alternatively, the reliability is calculated by ranking into a prescribed level according to the magnitude of the occurrence probability. For example, the log data reliability calculation unit 2131 may perform ranking for each power of 10. For example, the log data reliability calculation unit 2131 may perform ranking so that data with an occurrence probability of 10^(-1) is ranked 1 and data with an occurrence probability of 10^(-2) is ranked 2. For example, data that occurs with an occurrence probability of 10^(-7) is too rare to be used to determine the presence or absence of maintenance. Therefore, the log data reliability calculation unit 2131 may determine data that occurs with rank 1 as high reliability and calculate data that occurs between ranks 2 and 5 as low reliability.

[0142] Then, the log data reliability calculation unit 2131 stores the calculated reliability in the log data database 234 in association with the log data. (2) Data file generation Next, a description will be given of the operation of generating a data file for generating the animation data described in the above embodiment.

[0143] FIG. 21 is a flowchart showing an example of a data file generation operation in the second modified example of the embodiment. The operation of this flowchart is realized by reading out and executing a program stored in the internal cache of the control unit 21 of the central server 2 or in the program storage unit 12 .

[0144] First, as described above, it is assumed that the log data database 234 stores log data and the log data reliability associated with the log data.

[0145] This operation may start from the flowchart after the operations in steps ST101 to ST102 described with reference to FIG.

[0146] In step ST401, the user-specified log data acquisition unit 2122 acquires user-specified log data. The user-specified log data is acquired in the same manner as in step ST103 described with reference to FIG. 7 in the embodiment. Alternatively, the mechatronics device 1 may transmit the user-specified log data to the central server 2 or acquire the user-specified log data from a memory medium or the like in response to an instruction from an administrator (user) of the central server 2 or the like. The log data acquisition unit 211 outputs the acquired user-specified log data to the log data allocation unit 2132.

[0147] In step ST402, the log data sorting unit 2132 acquires the log data reliability. The log data sorting unit 2132 outputs the user-specified log data to the log data reliability calculation unit 2131. The log data reliability calculation unit 2131 refers to the log data stored in the log data database 234 and extracts the log data reliability corresponding to each row of the user-specified log data. Then, the log data reliability calculation unit 2131 outputs the extracted log data reliability to the log data sorting unit 2132. Alternatively, the log data reliability calculation unit 2131 may extract all log data with high reliability corresponding to the user-specified log data from the log data database 234 and output it to the log data sorting unit 2132.

[0148] Alternatively, the log data sorting unit 2132 may directly access the log data database 234 and obtain the log data reliability corresponding to the user-specified log data.

[0149] In step ST403, the log data sorting unit 2132 generates sensor reliability information. For example, the log data sorting unit 2132 sorts each line of the user-specified log data into data that is normally recorded and has high reliability, and data that has low reliability due to an abnormality, based on the user-specified log data and the log data reliability. Next, the log data sorting unit 2132 extracts user-specified sensor data from the user-specified log data. It generates sensor reliability information that is the sorting result (high reliability or low reliability) corresponding to the user-specified sensor data. Then, the log data sorting unit 2132 outputs the user-specified sensor data and the sensor reliability information to the outline behavior data generating unit 21331, and outputs the user-specified log data excluding the sensor data to the detailed behavior data generating unit 21332.

[0150] Here, if the log data reliability of a specific line of the user-specified log data is lower than a predetermined threshold, the log data sorting unit 2132 may replace the information of the specific line based on all of the log data with high reliability received from the log data reliability calculation unit 2131.

[0151] In step ST404, the general behavior data generating unit 21331 generates general behavior data based on the user-specified sensor data and the sensor reliability information. For example, the general behavior data generating unit 21331 refers to the sensor reliability information for the acquired user-specified sensor data, and processes the data so that the influence of data with low reliability is smaller than that of data with high reliability.

[0152] For example, the general behavior data generating unit 21331 checks whether or not a contradiction occurs in the reproduction of the movement of the paper medium using a normal algorithm when low-reliability data is taken into consideration. Specifically, the general behavior data generating unit 21331 first checks whether or not the movement of the entire paper medium can be reproduced while ignoring low-reliability data. If the movement of the paper medium can be reproduced even when low-reliability data is ignored, the general behavior data generating unit 21331 adopts this as general behavior data. On the other hand, if it is difficult to reproduce the movement of the entire paper medium when low-reliability data is ignored, the general behavior data generating unit 21331 adopts data with a high probability of transition from the previous state instead of the low-reliability data. Next, the general behavior data generating unit 21331 checks whether or not the movement of the entire paper medium can be reproduced using the adopted data. Then, if the movement of the paper medium can be reproduced, the general behavior data generating unit 21331 adopts the data as general behavior data. When there are multiple pieces of data with a high probability of transitioning from the previous state, the general behavior data generating unit 21331 analyzes parameters required to reproduce the movement of the paper medium, such as the type and number of tickets inserted, and whether the ticket was inserted with the front or back side facing up, for the low reliability data.The general behavior data generating unit 21331 then selects, through this analysis, data with a high probability of transitioning from the previous state instead of the low reliability data, and calculates the movement of the paper medium based on this selected data.In this way, the general behavior data generating unit 21331 performs control so as to reduce the influence of the low reliability data.

[0153] In this way, the general behavior data generating unit 21331 generates general behavior data based on the analysis result, and outputs the generated general behavior data to the detailed behavior data generating unit 21332.

[0154] In step ST405, the detailed behavior data generating unit 21332 generates a data file. The detailed behavior data generating unit 21332 converts the data into a data file based on the outline behavior data and the user-specified log data. For example, the detailed behavior data generating unit 21332 associates each sensor response with a paper medium from the outline behavior data, determines which paper medium is present at which position on the transport path at what timing, and generates a data file required to reproduce the operation of the paper medium proceeding along the transport path. In addition, the detailed behavior data generating unit 21332 generates data required to reproduce operations other than the operation of the paper medium proceeding along the transport path, such as a transport path switching operation, from the user-specified log data including data other than the sensor data. Then, the detailed behavior data generating unit 21332 outputs the data file generated in a format that can be converted into animation data to the animation data generating unit 214.

[0155] The subsequent processing may be the same as that in step ST105 described with reference to Fig. 7. Therefore, a duplicated description will be omitted here.

[0156] (Functions and Effects of the Second Modification of the Embodiment) According to the second modified example of the embodiment described above, in the process of converting the log data created by the mechatronics device 1 into a data file for generating an animation in the central server 2, even if the log data includes data with low reliability, the central server 2 can convert the data into data that can reproduce the operating status with good quality. As a result, even if the log data includes data with low reliability, it is possible to convert the data into data that can reproduce the operating status with good quality while suppressing the implementation cost of the data conversion software.

[0157] [Other embodiments] It should be noted that the present invention is not limited to the above-described embodiment. For example, the embodiment and the modified example of the embodiment can be combined. For example, the user-specified log data extracted using the learning model may be displayed to the administrator (user), and only the data required by the user may be converted into animation data. Also, log data with a deviation degree equal to or greater than a certain value may be presented as candidates for user-specified log data, and the administrator (user) may select user-specified log data from among them. In this case, the log data may be sorted in descending order of deviation degree and presented to the administrator (user).

[0158] In short, this invention is not limited to the above-mentioned embodiment, and various modifications can be made in the implementation stage without departing from the gist of the invention. Moreover, each embodiment may be implemented in combination as appropriate as possible, in which case the combined effect can be obtained. Furthermore, the above-mentioned embodiment includes inventions at various stages, and various inventions can be extracted by appropriate combinations of the disclosed constituent elements. [Explanation of symbols]

[0159] 1. Mechatronics equipment 11...Control section 111… Mechanism information acquisition department 112...Log data generation unit 113...Communication control unit 131...Log data storage unit 12...Program memory section 13...Data storage unit 14...Communication interface 15... Input / Output Interface 151...Input device 152...Output device 153…mechanism 1531…Sensor 1532…Motor 1533…Solenoid 1534…Status detection sensor 2. Central server 21...Control unit 2121…Event selection section 2122...User-specified log data acquisition unit 2123…Teacher Data Creation Department 2124…Learning Department 2125...Discrimination section 2126…Normal period extraction unit 2127…Log data extraction unit 2128…Learning Department 2129…Discrimination unit 211…Log data acquisition unit 212…Operation status selection section 213...Data conversion unit 2131...Log data reliability calculation unit 2132...Log data distribution unit 2133...Behavior data generation unit 21331…Outline behavior data generation unit 21332…Detailed behavior data generation section 214…Animation data generation unit 215…Communication control unit 22...Program memory section 23...Data storage unit 231...Event selection information storage unit 232...user-specified data storage unit 233...conversion data storage unit 234…Log data database 235…Fault record database 236…Trained model memory section 24…Communication interface 25... Input / Output Interface 251...Input device 252...Output device 3...Maintenance personnel terminal S1, S2, S3...Photo sensors

Claims

1. A mechatronics device whose mechanism is controlled by software; A central server capable of processing data of the mechatronics equipment; The central server comprises: a log data acquisition unit that acquires log data including information on the operation of the mechatronics device; a designated log data acquisition unit for acquiring designated log data that is required to reproduce an operating status from among the log data; a data conversion unit that converts the designated log data into a predetermined data file arranged in chronological order in order to reproduce the operating status of the designated log data; A mechatronics equipment maintenance system comprising:

2. the central server and the mechatronics device are connected via a network, the log data includes log data during operation of the mechatronics device, and the data conversion unit converts the designated log data into the data file in real time. The mechatronics equipment maintenance system according to claim 1 .

3. The mechatronics device is a media handling system including a paper transport mechanism and a control unit that controls the paper transport mechanism. The mechatronics equipment maintenance system according to claim 1 .

4. The media handling system is station service equipment. The mechatronics equipment maintenance system according to claim 3 .

5. A transmission control unit for transmitting the data file is further provided. The mechatronics equipment maintenance system according to claim 1 .

6. The data file is in a format readable by animation generation software. The mechatronics equipment maintenance system according to claim 1 .

7. The data file is data representing the location of each mechanism of the mechatronics device and which mechanism was operating at what time. The mechatronics equipment maintenance system according to claim 1 .

8. The specified log data is data selected according to an event specified by an administrator who manages the mechatronics equipment maintenance system. The mechatronics equipment maintenance system according to claim 1 .

9. 9. The mechatronics equipment maintenance system according to claim 8, wherein the specified log data is data selected further in accordance with conditions specified by the administrator.

10. The central server comprises: an operation status selection unit that calculates a feature amount of the mechatronics device from the log data, determines whether the feature amount satisfies a predetermined rule, and sets the log data in which the feature amount satisfies the predetermined rule as the designated log data, The mechatronics equipment maintenance system according to claim 1 .

11. The designated log data acquisition unit acquires the designated log data using a learning model trained using supervised data. The mechatronics equipment maintenance system according to claim 1 .

12. The central server comprises: a log data database that stores the log data; A failure record database that stores past failure information; a learning unit that acquires failure log data corresponding to the failure information, and learns the learning model using the failure log data, which is labeled with a failure content included in the failure information, as training data; The mechatronics equipment maintenance system according to claim 11 , further comprising:

13. The designated log data acquisition unit acquires the designated log data using a learning model trained using unsupervised data. The mechatronics equipment maintenance system according to claim 1 .

14. The central server comprises: a log data database that stores the log data; A failure record database that stores past failure information; a learning unit that acquires normal period log data during a period during which the mechatronics device is operating normally based on the failure information, and learns the learning model based on the normal period log data; The mechatronics equipment maintenance system according to claim 13, further comprising:

15. The designated log data acquisition unit acquires, as the designated log data, log data whose deviation degree outputted using a learning model exceeds a predetermined threshold value. The mechatronics equipment maintenance system according to claim 13.

16. The central server further includes an animation data generating unit that generates animation data based on the data file, and the animation data is used to display an animation that reproduces the operation status of the designated log data. The mechatronics equipment maintenance system according to claim 1 .

17. the animation data generating unit generates the animation data collectively by batch processing the data file; The mechatronics equipment maintenance system according to claim 16.

18. The central server further comprises a log data database for storing the log data; The data conversion unit is a log data reliability calculation unit that calculates the reliability of each operation of the log data; a log data sorting unit that sorts each line of the specified log data into high reliability data and low reliability data based on the reliability; a behavior data generating unit that generates the data file so that the influence of the data with low reliability is smaller than that of the data with high reliability; The mechatronics equipment maintenance system according to claim 1 , further comprising:

19. the log data reliability calculation unit calculates an occurrence probability of information indicating an action from the log data, and determines whether the reliability of each line of the log data is high or low based on the calculated occurrence probability. The mechatronics equipment maintenance system according to claim 18.

20. the log data sorting unit extracts designated sensor data indicating a behavior of a sensor from the designated log data, generates sensor reliability information including a sorting result corresponding to the designated sensor data, and outputs the designated sensor data and the sensor reliability information to the behavior data generating unit. The mechatronics equipment maintenance system according to claim 18.

21. The behavior data generation unit is a summary behavior data generating unit that generates summary behavior data representing a behavior of the sensor based on the designated sensor data and the sensor reliability information; a detailed behavior data generating unit that generates the data file, which is detailed behavior data, based on the outline behavior data and the designated log data; The mechatronic equipment maintenance system of claim 20, further comprising:

22. a log data acquisition unit that acquires log data including information on the operation of a mechatronics device whose mechanism is controlled by software; a designated log data acquisition unit for acquiring designated log data that is required to reproduce an operating status from among the log data; a data conversion unit that converts the designated log data into a predetermined data file arranged in chronological order in order to reproduce the operating status of the designated log data; A mechatronics equipment maintenance server comprising:

23. Acquiring log data including information on the operation of a mechatronics device that is controlled by software; acquiring designated log data from the log data, the designated log data being required to reproduce an operating status; converting the designated log data into a predetermined data file arranged in chronological order in order to reproduce the operating status of the designated log data; A mechatronics equipment maintenance method comprising:

24. A mechatronics equipment maintenance program comprising instructions to be executed by a processor of a mechatronics equipment maintenance server, Acquiring log data including information on the operation of a mechatronics device that is controlled by software; acquiring designated log data from the log data, the designated log data being required to reproduce an operating status; converting the designated log data into a predetermined data file arranged in chronological order in order to reproduce the operating status of the designated log data; Mechatronics equipment maintenance program.

Images