Event information collection device and computer-readable recording medium

Abstract

An event information collection device according to the present disclosure comprises: an event data acquisition unit that acquires event data related to an event including an operation event and / or an actuation event in an industrial machine; a context information acquisition unit that acquires context information related to the context influencing the event occurrence; an event information creation unit that creates event information in which the event data and the context have been associated with each other; and an output unit that outputs information related to the event information.

Description

Event Information Collection Device and Computer-Readable Recording Medium

[0001] The present disclosure relates to an event information collection device and a computer-readable recording medium.

[0002] In order to reduce the downtime of industrial machines (machine tools, robots, etc.), it is desirable to detect and classify machine abnormalities. Conventionally, in manufacturing sites such as factories, a method has been implemented in which data reflecting the state of a machine, such as an acceleration sensor, is collected, and machine abnormalities are detected and classified from the data (for example, Patent Document 1, etc.).

[0003] Japanese Patent Application Laid-Open No. 2017-120622

[0004] As described above, the state of industrial machines has been diagnosed using sensor data such as acceleration sensors. However, it may be difficult to diagnose the state and estimate the cause based only on these data, and it is not sufficient to assist in troubleshooting and maintenance work on industrial machines. Also, sensor data alone cannot be fully utilized to support subsequent work using industrial machines. In the field, a mechanism for collecting information used to support work using industrial machines is desired.

[0005] The event information collection device according to the present disclosure solves the above problems by collecting event information such as operations on a machine, their timings and operation sequences, and the operation timings of the machine so that it can be utilized in work using industrial machines.

[0006] One aspect of the present disclosure is an event information collection device including an event data acquisition unit that acquires event data related to an event including at least one of an operation event and an operation event in an industrial machine, a context information acquisition unit that acquires context information related to a context that affects the occurrence of the event, an event information creation unit that creates event information associating the event data and the context, and an output unit that outputs information related to the event information.

[0007] This is a hardware configuration diagram of the event information collection device according to the first embodiment. This is a block diagram showing the functions of the event information collection device according to the first embodiment. This is a block diagram showing the functions of the event information collection device according to the second embodiment. This is a diagram showing an example of an event sequence model. This is a block diagram showing the functions of the event information collection device according to the third embodiment. This is a diagram showing another example of an event sequence model. This is a block diagram showing the functions of the event information collection device according to the fourth embodiment. This is a block diagram showing the functions of the event information collection device according to the fifth embodiment. This is a block diagram showing the functions of the event information collection device according to the sixth embodiment.

[0008] The embodiments of this disclosure will be described below with reference to the drawings. [First Embodiment] Figure 1 is a schematic hardware configuration diagram showing the main parts of an event information collection device according to the first embodiment of this disclosure. The event information collection device 1 of this disclosure can be implemented, for example, as a control device for controlling industrial machinery. The event information collection device 1 of this disclosure can also be implemented on a personal computer attached to a control device for controlling industrial machinery, or on a personal computer, cell computer, fog computer 6, cloud server 7, or other computer connected to the control device via a wired / wireless network. In this embodiment, an example is shown in which the event information collection device 1 is implemented on a personal computer equipped with diagnostic functions that is connected to a control device for controlling industrial machinery via a network.

[0009] The CPU 11 in the event information collection device 1 according to this embodiment is a processor that controls the event information collection device 1 as a whole. The CPU 11 reads the system program stored in the ROM 12 via the bus 22 and controls the entire event information collection device 1 according to the system program. The RAM 13 temporarily stores temporary calculation data, display data, and various data input from external sources.

[0010] The non-volatile memory 14 is composed of, for example, a memory backed up by a battery (not shown) or an SSD (Solid State Drive), and its stored state is maintained even when the power to the event information collection device 1 is turned off. The non-volatile memory 14 stores programs and data read from external devices 72 via the interface 15, programs and data input via the input device 71, and programs and data acquired from the industrial machine 3. The data includes control data such as command position, command speed, position feedback value, and speed feedback value; sensor data such as motor current, voltage, temperature, vibration, and sound detected by sensors 4 attached to the industrial machine 3; setting data such as the program number of the control program and the tool number of the tool being used; operation event data including the type and timing of operations related to the industrial machine 3; and operation event data including the type and timing of operations in the industrial machine 3. The data stored in the non-volatile memory 14 may be expanded into the RAM 13 when executed / used. In addition, various system programs, such as known analysis programs, are pre-written in the ROM 12.

[0011] Interface 15 is an interface for connecting the CPU 11 of the event information collection device 1 to an external device 72 such as a USB device. From the external device 72, for example, pre-stored control programs and data related to the operation of each industrial machine 3 can be read. In addition, control programs and setting data edited within the event information collection device 1 can be stored in an external storage means via the external device 72.

[0012] Interface 20 is an interface for connecting the CPU 11 of the event information collection device 1 to a wired or wireless network 5. Industrial machinery 3, fog computers 6, cloud servers 7, etc., are connected to the network 5, and they exchange data with the event information collection device 1.

[0013] The display device 70 displays data, programs, and other data obtained as a result of their execution, which are loaded into memory, and output via the interface 17. The input device 71, consisting of a keyboard and pointing device, transmits commands and data based on operator operations to the CPU 11 via the interface 18.

[0014] Industrial machine 3 is a machine such as a machine tool, injection molding machine, or robot. Industrial machine 3 is equipped with sensors 4 that detect predetermined physical quantities related to the operation of at least each part of the industrial machine 3. The event information collection device 1 acquires predetermined command values ​​and signal values, predetermined physical quantities detected by sensors 4, setting data such as the program number of the control program being executed and the tool number of the tool being used, operation event data related to user operations, and operation event data related to the operation of industrial machine 3 from industrial machine 3 via the network 5 and interface 20. The acquired predetermined data is stored in RAM 13 or non-volatile memory 14 and processed by CPU 11.

[0015] Figure 2 is a schematic block diagram showing the functions of the event information collection device 1 according to the first embodiment of this disclosure. Each function of the event information collection device 1 according to this embodiment is realized by the CPU 11 of the event information collection device 1 shown in Figure 1 executing a system program and controlling the operation of each part of the event information collection device 1.

[0016] The event information collection device 1 of this embodiment includes an event data acquisition unit 100, a context information acquisition unit 105, an event information creation unit 110, and an output unit 195. In addition, an event information storage unit 200, which is an area for storing event information created by the event information creation unit 110, is pre-prepared on the RAM 13 to non-volatile memory 14 of the event information collection device 1.

[0017] The event data acquisition unit 100 acquires event data relating to the type and timing of events that occur in the industrial machine 3. The event data includes at least one of the following: operation event data relating to operation events that occur when a user operates the control panel of the industrial machine 3, etc., and operation event data relating to operation events that occur when predetermined operations or processes are performed in the industrial machine 3, such as starting or ending the execution of a machining program or switching to maintenance mode. Each event data includes at least information relating to the type of event and information relating to the timing in which the event occurred. For example, operation event data may include information that the operation performed by the user was the operation of the automatic operation start button and information relating to the timing in which the automatic operation start button was operated. Operation event data may include information relating to the automatic operation start operation in the industrial machine 3 based on the user's operation of the automatic operation start button and information relating to the timing in which the automatic operation start operation occurred. Operation event data may include information relating to the occurrence of operation events in the industrial machine 3 determined from control data, sensor data, setting data, etc. However, it should be noted that operation event data is not the control data, sensor data, or setting data itself. The event data acquisition unit 100 outputs the acquired event data to the event information creation unit 110.

[0018] The context information acquisition unit 105 acquires context information relating to the context that influences the occurrence of events in the industrial machine 3. Context information is information indicating the circumstances under which events such as operation events and motion events occurred in the industrial machine 3 (processing status, operating status, environmental conditions, etc.). The context information acquisition unit 105 acquires context information at least when the event data acquisition unit 100 acquires event data. Examples of context information include the time of day when the event occurred (morning, afternoon, daytime, nighttime, etc.), the day of the week, the season, the ambient temperature of the industrial machine 3, information relating to the location where the industrial machine 3 is installed, information relating to the user operating the industrial machine 3, information regarding the delivery destination of the workpiece being processed, and information indicating the purpose of the processing being performed at that time (rough machining, finishing machining, etc.). The context information acquisition unit 105 may use data acquired from the industrial machine 3 as context information. Alternatively, the context information acquisition unit 105 may create predetermined context information based on data acquired from the industrial machine 3. For example, it may create context information by identifying the time period in which the event occurred based on the time acquired from the industrial machine 3. In this case, the rules for creating context information based on what data should be prepared in advance using tables or functions. The context information acquisition unit 105 outputs the acquired context information to the event information creation unit 110.

[0019] The event information creation unit 110 creates event information that associates event data with context information based on the event data acquired by the event data acquisition unit 100 and the context information acquired by the context information acquisition unit 105. The event information includes at least information that associates the event data with the context information at the time the event occurred. The sequence of this event information arranged in chronological order will be a sequence of operation events performed on the industrial machine 3, a sequence of action events that occurred, or a combination of these sequences. The event information creation unit 110 stores the created event information in the event information storage unit 200.

[0020] The output unit 195 refers to the event information storage unit 200 and outputs information related to the event information stored in the event information storage unit 200. The output unit 195 may also display and output the event information stored in the event information storage unit 200 to the display device 70 based on user operations, for example. At this time, the output unit 195 may accept the user's specification of the industrial machine 3, the installation location of the industrial machine 3, the date and time and time range, user information, customer information, processing purpose, time range, and other context information, and narrow down the event information to output based on the content of the received specification. In addition, the output unit 195 may transmit and output event information via the network 5 in response to requests from other computers such as the fog computer 6 and the cloud server 7.

[0021] The event information collection device 1 according to this embodiment, having the above configuration, collects data related to events that occur in the industrial machine 3. It also creates event information associated with context information indicating the circumstances under which the event occurred, and stores the created event information. Then, by outputting the desired event information upon request, the user can understand in chronological order what kind of event occurred under what circumstances. The user can reduce the effort required for setting up, maintaining, and repairing the industrial machine 3 by referring to the referenced event information. For example, by referring to the sequence of operation events related to operations performed before the failure occurred, it is possible to check whether any unusual operations were performed, or by referring to the sequence of operation events, it is possible to more easily understand if the timing of operations is different from usual.

[0022] [Second Embodiment] The following describes an event information collection device according to a second embodiment. The event information collection device 1 according to this embodiment has the same hardware configuration as the event information collection device 1 according to the first embodiment.

[0023] Figure 3 is a schematic block diagram showing the functions of the event information collection device 1 according to the second embodiment of this disclosure. Each function of the event information collection device 1 according to this embodiment is realized by the CPU 11 of the event information collection device 1 shown in Figure 1 executing a system program and controlling the operation of each part of the event information collection device 1.

[0024] The event information collection device 1 of this embodiment includes an event data acquisition unit 100, a context information acquisition unit 105, an event information creation unit 110, an output unit 195, and an event sequence model creation unit 115. Furthermore, the RAM 13 to non-volatile memory 14 of the event information collection device 1 is pre-configured with an event information storage unit 200, which is an area for storing event information created by the event information creation unit 110, and an event sequence model storage unit 210, which is an area for storing event sequence models created based on the event information.

[0025] The event data acquisition unit 100, context information acquisition unit 105, and event information creation unit 110 in this embodiment each have the same functions as the event data acquisition unit 100, context information acquisition unit 105, and event information creation unit 110 in the first embodiment.

[0026] The event sequence model creation unit 115 creates an event sequence model based on the event information stored in the event information storage unit 200. An event sequence model is a model that represents at least one of the order in which events occur and the time interval between events for multiple event information. For example, a Hidden Semi-Markov Model (HSMM) can be used as the event sequence model. This model can represent the transition from one state to another using transition probabilities, as well as the distribution of the duration of each state. Therefore, by considering events that occur in the industrial machine 3 as states and defining the time taken until the next event occurs as the state duration, the order in which events occur and the time interval between events can be represented using a Hidden Semi-Markov Model. Note that if there is no particular need to represent the time interval, a Hidden Markov Model (HMM) can also be used as the event sequence model.

[0027] The event sequence model creation unit 115 creates an event sequence model based on the accumulated event information when sufficient event information has been accumulated or based on instructions from the user. The event sequence model creation unit 115 extracts from the event information storage unit 200 a sequence of event information between events of a type specified by the user or set in advance. For example, the processing program execution start event may be defined as the starting event, and the processing program execution end event as the ending event, and the sequence of event information relating to the operation events between these events may be extracted. Similarly, for example, the sequence of event information relating to the operation events between a predetermined alarm occurrence event and the event in which the alarm was correctly cleared may be extracted. Generally, one event type should be defined as the starting event for one type of event information sequence, but there may be multiple types for the ending event. The event sequence model creation unit 115 may also accept or set in advance information from the user regarding the order of event occurrences and the allowable time between events, and narrow down the sequence of event information to be extracted based on that information. In this way, cases where the time interval between events is abnormally long can be excluded from being used to create the model without being treated as consecutive events.

[0028] The event sequence model creation unit 115 extracts event information sequences to create an event sequence model, but these sequences are limited to those that have correctly transitioned from a predetermined starting event to a predetermined ending event. The event sequence model creation unit 115 may also exclude from extraction event information sequences that include predetermined exclusion events between the predetermined starting event and the predetermined ending event. For example, when extracting an information sequence where the starting event is an event in which a predetermined alarm occurs and the ending event is an event in which the alarm is correctly cleared, events in which other alerts occur may be excluded, and sequences containing events in which other alerts occur within the event information sequence from the starting event to the ending event may be excluded from extraction. With this configuration, only event sequences that occur when the industrial machine 3 operates correctly as expected, and event sequences that occur when the user performs correct operations as expected, can be used as correct event information sequences to create the event sequence model.

[0029] If a sufficient amount of event data has been collected in advance, multiple sequences of the same type of event information can be extracted. The event sequence model creation unit 115 creates an event sequence model based on the multiple sequences of the same type of event information extracted in this way. For example, if the event sequence model is created as a hidden semi-Markov model, the model can be created by statistically determining the transition probabilities of other events occurring after a certain event has occurred, and the distribution of the time taken from the occurrence of one event to the occurrence of other events. The event sequence model creation unit 115 can create multiple event sequence models with different starting events. The event sequence model creation unit 115 stores the event sequence models created in this way in the event sequence model storage unit 210.

[0030] Figure 4 shows an example of an event sequence model. The event sequence model illustrated in Figure 4 models a sequence of four operational events in industrial machine 3—execution of machining program A, termination of machining program A, execution of machining program B, and termination of machining program B—using a hidden semi-Markov model. Note that Figure 4 shows an example of an event sequence model where the occurrence of events does not branch, for the sake of simplicity. In the example in Figure 4, the distribution of time taken from the start of execution of machining program A to the termination of machining program A is denoted as time interval distribution a1. The time taken from the start to the end of execution of machining program A is not necessarily the same. Therefore, in the event sequence model illustrated in Figure 4, this time interval is represented by a probability distribution. Similarly, the distribution of time taken from the termination of machining program A to the start of execution of machining program B is denoted as time interval distribution a2, and the distribution of time taken from the start of execution of machining program B to the termination of machining program B is denoted as time interval distribution a3.

[0031] The output unit 195 in this embodiment outputs the event sequence model created by the event sequence model creation unit 115. The destination to which the output unit 195 outputs the event sequence model may be, for example, a display device 70, or a display unit of the industrial machine 3 (not shown). Alternatively, it may be configured to output to other computers such as a fog computer 6 or a cloud server 7. Furthermore, the output unit 195 may be configured to narrow down the event information to be displayed and output based on context information specified by the user.

[0032] The event information collection device 1 according to this embodiment, which has the above configuration, collects data related to events that occur in the industrial machine 3 and creates an event sequence model based on the collected event data. By referring to this event sequence model, the user can understand the content of tasks that are generally performed in the industrial machine 3 and standard time intervals, and can consider the content of tasks to be performed in the future.

[0033] [Third Embodiment] The following describes an event information collection device according to the third embodiment. The event information collection device 1 according to this embodiment has the same hardware configuration as the event information collection device 1 according to the first embodiment.

[0034] Figure 5 is a schematic block diagram showing the functions of the event information collection device 1 according to the third embodiment of this disclosure. Each function of the event information collection device 1 according to this embodiment is realized by the CPU 11 of the event information collection device 1 shown in Figure 1 executing a system program and controlling the operation of each part of the event information collection device 1.

[0035] The event information collection device 1 of this embodiment includes an event data acquisition unit 100, a context information acquisition unit 105, an event information creation unit 110, an event sequence model creation unit 115, and an output unit 195, as well as an abnormal event sequence detection unit 120. Furthermore, the RAM 13 to non-volatile memory 14 of the event information collection device 1 is pre-configured with an event information storage unit 200, which is an area for storing event information created by the event information creation unit 110, and an event sequence model storage unit 210, which is an area for storing event sequence models created based on the event information.

[0036] The event data acquisition unit 100, context information acquisition unit 105, event information creation unit 110, and event sequence model creation unit 115 in this embodiment each have the same functions as the event data acquisition unit 100, context information acquisition unit 105, event information creation unit 110, and event sequence model creation unit 115 in the second embodiment.

[0037] The abnormal event sequence detection unit 120 refers to the event sequence model stored in the event sequence model storage unit 210 and detects abnormal event sequences from the event information sequences created by the event information creation unit 110. The abnormal event sequence detection unit 120 determines whether the event information created by the event information creation unit is related to the starting event of the event sequence model stored in the event sequence model storage unit 210. If it is related to the starting event of the event sequence model, it obtains the sequence of event information following that event information and determines whether it is included in the transition sequence in the event sequence model. If it is not included in the transition sequence in the event sequence model, it determines that the sequence of event information is an abnormal event sequence. Alternatively, the abnormal event sequence detection unit 120 may determine whether it is an abnormal event sequence by calculating an abnormality score based on the probability of data occurrence in a general hidden Markov model or hidden semi-Markov model. In this case, an abnormality score formula is prepared in which the abnormality score increases as the probability of a predetermined event sequence occurring in the event sequence model decreases. Then, for event sequence models where the starting event matches, if the abnormality score calculated based on the event information sequence created by the event information creation unit 110 exceeds a predetermined threshold, the event information sequence is determined to be an abnormal event sequence. When the abnormal event sequence detection unit 120 determines that an abnormal event sequence has occurred, it outputs a message to that effect to the output unit 195.

[0038] The operation of the abnormal event sequence detection unit 120 will be explained using Figure 6. Figure 6 shows an example of an event sequence model. The event sequence model illustrated in Figure 6 is a hidden semi-Markov model that models the sequence of operation events in the industrial machine 3 from the time alarm α is generated until it is released. Note that in Figure 6, the distribution of time intervals between events is omitted from the notation. Also, for the sake of simplicity, context information included in the event information is also omitted. As illustrated in Figure 6, when alarm α is generated, a sequence of event information is modeled in which alarm α is released by performing operations A and B in that order, and a sequence of event information is modeled in which alarm α is released by performing operations C, D and E in that order. Furthermore, in the past data used when creating this event sequence model, among the procedures in which alarm α was released using the correct procedure, p1 represents the proportion in which the alarm was released using the operation procedure A and B, and (1-p1) represents the proportion in which the alarm was released using the operation procedure C, D and E. In such a case, suppose that event information indicating that alarm α has been generated is created in the event information creation unit 110. In this case, when the event alarm α occurs, the abnormal event sequence detection unit 120 reads out the event sequence model exemplified in Figure 6, which uses the alarm α as the starting event. Next, when the event information creation unit 110 creates event information related to operation A, the abnormal event sequence detection unit 120 determines whether the sequence of events alarm α occurs and operation A is performed is included in the event sequence model exemplified in Figure 6. In this case, it is included, so the abnormal event sequence detection unit 120 determines that the event that occurred in the industrial machine 3 is not abnormal at this point. Next, when the event information creation unit 110 creates event information related to operation D, the abnormal event sequence detection unit 120 determines whether the sequence of events alarm α occurs, operation A is performed and operation D is performed is included in the event sequence model exemplified in Figure 6.In this case, since the sequence of events is not included in the event sequence model illustrated in Figure 6, the abnormal event sequence detection unit 120 determines that this sequence of events is abnormal and outputs that fact to the output unit 195.

[0039] The output unit 195 in this embodiment outputs a message indicating that the abnormal event sequence detection unit 120 has detected an abnormal event sequence. The output unit 195 may output a message indicating that an abnormal event has been detected to, for example, a display device 70, or to a display unit (not shown) of the industrial machine 3 where the abnormal event sequence occurred. It may also output to other computers such as a fog computer 6 or a cloud server 7. When an abnormal event sequence is detected, the output unit 195 may also output the correct event sequence following the starting event based on the event sequence model stored in the event sequence model storage unit 210. In this case, all correct event sequences may be output, or only event sequences with a high rate of matching the abnormal event sequence may be output. For example, in the above example, the abnormal event sequence of Alarm α occurring, Operation A performed, Operation D performed has a high rate of matching the correct event sequence of Alarm α occurring, Operation A performed, Operation B performed, Alarm α canceled, so this event sequence may be output.

[0040] The event information collection device 1 according to this embodiment, having the above configuration, collects data related to events that occur in the industrial machine 3 and creates an event sequence model based on the collected event data. Then, using this event sequence model, it is possible to detect abnormalities in the sequence of events that occur in the industrial machine 3 and notify the operator, thereby supporting the work being performed in the industrial machine 3. [Fourth Embodiment] The following describes an event information collection device according to the fourth embodiment. The event information collection device 1 according to this embodiment has the same hardware configuration as the event information collection device 1 according to the first embodiment.

[0041] FIG. 7 shows, as a schematic block diagram, the functions of the event information collection device 1 according to the fourth embodiment of the present disclosure. Each function of the event information collection device 1 according to the present embodiment is realized by the CPU 11 of the event information collection device 1 shown in FIG. 1 executing a system program and controlling the operations of each part of the event information collection device 1.

[0042] In addition to the event data acquisition unit 100, the context information acquisition unit 105, the event information creation unit 110, and the output unit 195, the event information collection device 1 of the present embodiment further includes a sensor data acquisition unit 125 and a sensor information creation unit 130. Also, on the RAM 13 to the non-volatile memory 14 of the event information collection device 1, there are provided in advance an event information storage unit 200 which is an area for storing the event information created by the event information creation unit 110, and a sensor information storage unit 220 which is an area for storing the sensor information created by the sensor information creation unit 130.

[0043] The event data acquisition unit 100, the context information acquisition unit 105, and the event information creation unit 110 according to the present embodiment each have the same functions as the event data acquisition unit 100, the context information acquisition unit 105, and the event information creation unit 110 according to the first embodiment.

[0044] The sensor data acquisition unit 125 acquires sensor data detected by sensors related to the industrial machine 3. Examples of the sensor data acquired by the sensor data acquisition unit 125 include the current, voltage, temperature, vibration, sound, etc. of a motor. Each sensor data includes at least information related to the type of the sensor and information related to the timing when the sensor value is detected. The sensor data acquisition unit 125 outputs the acquired sensor data to the sensor information creation unit 130.

[0045] The sensor information creation unit 130 creates sensor information associating the sensor data acquired by the sensor data acquisition unit 125 and the context information acquired by the context information acquisition unit 105. The sensor information includes at least information associating the sensor data with the context information at the time when the sensor value related to the sensor data was detected. The series obtained by arranging this sensor information in chronological order is the series of sensor values detected by the industrial machine 3. The sensor information creation unit 130 stores the created sensor information in the sensor information storage unit 220.

[0046] The output unit 195 according to the present embodiment outputs both the event information created by the event information creation unit 110 and the sensor information created by the sensor information creation unit 130 in a format that allows the chronological order before and after to be grasped. The destination of the output by the output unit 195 may be, for example, the display device 70, or a display unit (not shown) of the industrial machine 3 where the series of the abnormal events occurred. Also, it may be output to other computers such as the fog computer 6 and the cloud server 7. Further, the output unit 195 may narrow down and output the event information and sensor information to be displayed according to the context information specified by the user.

[0047] The event information collection device 1 according to the present embodiment having the above configuration can display the series of the event information generated in the industrial machine 3 and the sensor information detected in the industrial machine 3 so that the chronological order can be grasped, and can be used to assist the work of analyzing the abnormality of the industrial machine 3.

[0048] [Fifth Embodiment] Hereinafter, the event information collection device according to the fifth embodiment will be described. The event information collection device 1 according to the present embodiment has the same hardware configuration as the event information collection device 1 according to the first embodiment.

[0049] Figure 8 is a schematic block diagram showing the functions of the event information collection device 1 according to the fifth embodiment of this disclosure. Each function of the event information collection device 1 according to this embodiment is realized by the CPU 11 of the event information collection device 1 shown in Figure 1 executing a system program and controlling the operation of each part of the event information collection device 1.

[0050] The event information collection device 1 of this embodiment includes an event data acquisition unit 100, a context information acquisition unit 105, an event information creation unit 110, an event sequence model creation unit 115, an abnormal event sequence detection unit 120, a sensor data acquisition unit 125, a sensor information creation unit 130, and an output unit 195, as well as an abnormal sensor value detection unit 135. Furthermore, the RAM 13 to non-volatile memory 14 of the event information collection device 1 is pre-configured with an event information storage unit 200, which is an area for storing event information created by the event information creation unit 110, an event sequence model storage unit 210, which is an area for storing event sequence models created based on the event information, and a sensor information storage unit 220, which is an area for storing sensor information created by the sensor information creation unit 130.

[0051] The event data acquisition unit 100, context information acquisition unit 105, event information creation unit 110, event sequence model creation unit 115, and abnormal event sequence detection unit 120 in this embodiment each have the same functions as the event data acquisition unit 100, context information acquisition unit 105, event information creation unit 110, event sequence model creation unit 115, and abnormal event sequence detection unit 120 in the third embodiment. Furthermore, the sensor data acquisition unit 125 and sensor information creation unit 130 in this embodiment each have the same functions as the sensor data acquisition unit 125 and sensor information creation unit 130 in the fourth embodiment.

[0052] The abnormal sensor value detection unit 135 detects abnormalities in the industrial machine 3 using sensor values ​​related to sensor information stored in the sensor information storage unit 220. The abnormal sensor value detection unit 135 may, for example, pre-set a range of normal values ​​for each sensor and determine that an abnormal sensor value has been detected when the sensor value detected by the sensor falls outside the set range of normal values. Alternatively, a mathematical model or machine learning model may be prepared to determine abnormalities in sensor values, and abnormalities in sensor values ​​may be detected using this model. When the abnormal sensor value detection unit 135 detects an abnormality in sensor values, it outputs a message to that effect to the output unit 195.

[0053] The output unit 195 in this embodiment outputs both the detection result from the abnormal sensor value detection unit 135 and the detection result from the abnormal event sequence detection unit 120 in a format that allows the time order of each to be determined. The output destination of the output unit 195 may be, for example, a display device 70, or a display unit (not shown) of the industrial machine 3 where the abnormal event sequence occurred. Alternatively, it may be configured to output to another computer such as a fog computer 6 or a cloud server 7. Furthermore, the output unit 195 may be configured to narrow down the event information and sensor information to be displayed based on context information specified by the user.

[0054] The event information collection device 1 according to this embodiment, which has the above configuration, makes it possible to easily check whether there was a sequence of unusual operation events or motion events prior to the detection of an abnormality in the sensor value of the industrial machine 3, and can be used to support the work of analyzing the abnormality of the industrial machine 3.

[0055] [Sixth Embodiment] The following describes an event information collection device according to the sixth embodiment. The event information collection device 1 according to this embodiment has the same hardware configuration as the event information collection device 1 according to the first embodiment.

[0056] Figure 9 is a schematic block diagram showing the functions of the event information collection device 1 according to the sixth embodiment of this disclosure. Each function of the event information collection device 1 according to this embodiment is realized by the CPU 11 of the event information collection device 1 shown in Figure 1 executing a system program and controlling the operation of each part of the event information collection device 1.

[0057] The event information collection device 1 of this embodiment includes an event data acquisition unit 100, a context information acquisition unit 105, an event information creation unit 110, an event sequence model creation unit 115, and an output unit 195, as well as a work procedure extraction unit 140. Furthermore, the RAM 13 to non-volatile memory 14 of the event information collection device 1 is pre-configured with an event information storage unit 200, which is an area for storing event information created by the event information creation unit 110, and an event sequence model storage unit 210, which is an area for storing event sequence models created based on the event information.

[0058] The event data acquisition unit 100, context information acquisition unit 105, event information creation unit 110, and event sequence model creation unit 115 in this embodiment each have the same functions as the event data acquisition unit 100, context information acquisition unit 105, event information creation unit 110, and event sequence model creation unit 115 in the second embodiment.

[0059] The work procedure extraction unit 140 extracts a sequence of operation events to be performed after the starting event of the event sequence model as a work procedure, based on the event sequence model. The work procedure extraction unit 140 accepts, for example, a starting event from the user. It then reads the event sequence model starting from the accepted event from the event sequence model storage unit 210. It then extracts events related to the user's operations from the read event sequence model. The work procedure extraction unit 140 may also allow the user to directly specify the event sequence model. The work procedure extraction unit 140 may extract all sequences of operation events shown in the event sequence model, or it may extract the sequence of operation events with the highest transition probability from the event sequence model. For example, when extracting all sequences of operation events as a work procedure from the event sequence model illustrated in Figure 6, two sequences of operation events are extracted as work procedures: the sequence of operation events "Operation A -> Operation B" and the sequence of operation events "Operation C -> Operation D -> Operation E". Furthermore, assuming that transition probability p1 > transition probability (1-p1), when extracting the most frequently performed sequence of operation events as a work procedure from the event sequence model illustrated in Figure 6, the sequence of operation events "Operation A -> Operation B" is extracted as the work procedure. The work procedure extraction unit 140 outputs the extracted work procedure to the output unit 195.

[0060] In this embodiment, the output unit 195 outputs the work procedure extracted by the work procedure extraction unit 140. The output destination of the output unit 195 may be, for example, a display device 70, or a display unit (not shown) of the industrial machine 3 where the series of abnormal events occurred. Alternatively, the output may be sent to another computer such as a fog computer 6 or a cloud server 7.

[0061] The event information collection device 1 according to this embodiment, having the above configuration, collects data related to events that occur in the industrial machine 3 and creates an event sequence model based on the collected event data. It then extracts the work procedures to be performed when a predetermined event occurs from the event sequence model and notifies the worker. This makes it possible to support the work that should be performed in the industrial machine 3.

[0062] [Other Embodiments] The following describes an event information collection device according to other embodiments. In the embodiments described above, an example is shown in which an event information storage unit 200, an event sequence model storage unit 210, and a sensor information storage unit 220 are provided within the event information collection device. However, the device is not limited to this, and each of these storage units may be provided on a fog computer 6 or a cloud server. With this configuration, it becomes possible to share and use each storage unit among multiple event information collection devices.

[0063] While embodiments of this disclosure have been described in detail above, this disclosure is not limited to the individual embodiments described above. These embodiments can be added, replaced, modified, partially deleted, etc., in any way that does not depart from the spirit of the invention or from the idea and intent of this disclosure derived from the claims and their equivalents. For example, the order of operations and processes in the embodiments described above are shown as examples only and are not limited thereto. The same applies when numerical values ​​or mathematical formulas are used in the description of the embodiments described above.

[0064] The following are annotations relating to embodiments of the present disclosure. (Annotation 1) An event information collection device (1) according to one aspect of the present disclosure includes: an event data acquisition unit (100) that acquires event data relating to an event that includes at least one of operation events and motion events in an industrial machine (3); a context information acquisition unit (105) that acquires context information relating to a context that influences the occurrence of the event; an event information creation unit (110) that creates event information relating the event data and the context; and an output unit (195) that outputs information relating to the event information.

[0065] (Note 2) An event information collection device (1) according to another aspect of the present disclosure further comprises an event sequence model creation unit (115) that creates an event sequence model which is a model representing at least one of the occurrence order of a plurality of events and the time interval between each event, and the output unit (195) outputs the event sequence model. (Note 3) An event information collection device (1) according to another aspect of the present disclosure further comprises an abnormal event sequence detection unit (120) that uses the event sequence model to detect whether or not a sequence of events that occurred in the industrial machine (3) is abnormal, and the output unit (195) outputs information relating to the event sequence in which an abnormality was detected when the abnormal event sequence detection unit (120) detects an abnormality in a sequence of events that occurred in the industrial machine (3).

[0066] (Note 4) An event information collection device (1) according to another aspect of the present disclosure further comprises a sensor data acquisition unit (125) that acquires sensor data detected by a sensor related to the industrial machine (3), and a sensor information creation unit (130) that creates sensor information relating the sensor data and the context, wherein the output unit (195) outputs both the event information and the sensor information in a format in which their respective time order can be understood. (Note 5) An event information collection device (1) according to another aspect of the present disclosure further comprises: a sensor data acquisition unit (125) that acquires sensor data detected by a sensor related to the industrial machine (3); a sensor information creation unit (130) that creates sensor information relating the sensor data to the context; and an abnormal sensor value detection unit (135) that detects an abnormality in the industrial machine (3) using the sensor values ​​related to the sensor information, wherein the output unit (195) outputs both the detection result by the abnormal sensor value detection unit (135) and the detection result by the abnormal event sequence detection unit (120) in a format in which the time order of each can be understood.

[0067] (Note 6) An event information collection device (1) according to another aspect of the present disclosure further comprises a work procedure extraction unit (140) that extracts a series of operation events to be performed after the starting event of the event sequence model as a work procedure, based on the event sequence model.

[0068] (Note 7) A computer-readable recording medium according to one aspect of the present disclosure records a program that causes a computer to function as an event data acquisition unit (100) that acquires event data relating to an event including at least one of an operation event and an action event in an industrial machine (3), a context information acquisition unit (105) that acquires context information relating to a context that influences the occurrence of the event, an event information creation unit (110) that creates event information relating the event data and the context, and an output unit (195) that outputs information relating to the event information.

[0069] 1 Event information collection device 3 Industrial machinery 4 Sensor 5 Network 6 Fog computer 7 Cloud server 11 CPU 12 ROM 13 RAM 14 Non-volatile memory 15 Interface 17, 18, 20 Interface 22 Bus 70 Display device 71 Input device 72 External device 100 Event data acquisition unit 105 Context information acquisition unit 110 Event information creation unit 115 Event sequence model creation unit 120 Abnormal event sequence detection unit 125 Sensor data acquisition unit 130 Sensor information creation unit 135 Abnormal sensor value detection unit 140 Work procedure extraction unit 200 Event information storage unit 210 Event sequence model storage unit 220 Sensor information storage unit

Claims

1. An event information collection device comprising: an event data acquisition unit that acquires event data relating to an event that includes at least one of an operation event and an action event in an industrial machine; a context information acquisition unit that acquires context information relating to the context that influences the occurrence of the event; an event information creation unit that creates event information relating the event data and the context; and an output unit that outputs information relating to the event information.

2. The event information collection device according to claim 1, further comprising an event sequence model creation unit that creates an event sequence model which is a model representing at least one of the order in which multiple events occur and the time interval between each event, wherein the output unit outputs the event sequence model.

3. The event information collection device according to claim 2, further comprising an abnormal event sequence detection unit that uses the event sequence model to detect whether or not a sequence of events generated in the industrial machine is abnormal, wherein the output unit outputs information relating to the event sequence in which an abnormality was detected when the abnormal event sequence detection unit detects an abnormality in a sequence of events generated in the industrial machine.

4. An event information collection device according to claim 1, further comprising: a sensor data acquisition unit that acquires sensor data detected by a sensor relating to the industrial machine; and a sensor information creation unit that creates sensor information relating the sensor data and the context, wherein the output unit outputs both the event information and the sensor information in a format in which their respective time order can be understood.

5. An event information collection device according to claim 3, further comprising: a sensor data acquisition unit that acquires sensor data detected by a sensor relating to the industrial machine; a sensor information creation unit that creates sensor information relating the sensor data to the context; and an abnormal sensor value detection unit that detects an abnormality in the industrial machine using the sensor value relating to the sensor information, wherein the output unit outputs both the detection result by the abnormal sensor value detection unit and the detection result by the abnormal event sequence detection unit in a format that allows the time order of each to be understood.

6. The event information collection device according to claim 2, further comprising a work procedure extraction unit that extracts a series of operation events to be performed after the starting event of the event sequence model, based on the event sequence model.

7. A computer-readable recording medium that records a program causing a computer to function as: an event data acquisition unit that acquires event data relating to an event including at least one of operation events and motion events in an industrial machine; a context information acquisition unit that acquires context information relating to a context that influences the occurrence of the event; an event information creation unit that creates event information relating the event data and the context; and an output unit that outputs information relating to the event information.

Images