Failure prediction device and program

Abstract

This failure prediction device is provided with a determination unit, a setting unit, and a prediction unit. The determination unit executes, upon input of the failure prediction instruction, a determination process for an operation stop date of the industrial machine on the basis of first operation data indicating an operation of the industrial machine or second operation data indicating an operation of a peripheral device that operates in coordination with the industrial machine. The setting unit sets an evaluation period on the basis of the determined number of operation stop days. The prediction unit predicts the occurrence of a malfunction in the industrial machine on the basis of the first operation data generated during the set evaluation period.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a failure prediction device having a function of predicting a failure of an industrial machine, and a program. BACKGROUND

[0002] In recent years, industrial machines such as robots are used in various places. In particular, in a production plant that produces goods, a robot that can repeatedly perform a prescribed motion with high precision has become indispensable. A sudden failure of a robot can cause the production of goods to stop for a long time, and thus such a situation is one of the situations that is most required to be avoided in terms of utilizing a robot. In order to prevent a sudden failure of a robot, a technique of predicting a failure of a robot is known (Patent Literature 1).

[0003] PRIOR ART DOCUMENTS

[0004] PATENT LITERATURE

[0005] Patent Literature 1: Japanese Patent Application Publication No. 2021-022074 SUMMARY

[0006] PROBLEMS TO BE SOLVED BY THE INVENTION

[0007] In recent years, there has been a situation in which the operation of a plant is stopped for a long time due to infection disease prevention measures and the like. Such a situation can lead to a decrease in the accuracy of failure prediction of a robot. Therefore, it is desirable to present a technique in which the accuracy of prediction of a failure of a robot does not decrease even if a situation in which the operation of a robot is stopped for a long time occurs.

[0008] MEANS FOR SOLVING THE PROBLEMS

[0009] The failure prediction device of the present disclosure has a determination section that, on the occasion of input of a failure prediction instruction, performs determination processing of a day of operation stop of an industrial machine on the basis of first motion data indicating a motion of the industrial machine or second motion data indicating a motion of a peripheral device that cooperates with the industrial machine; a setting section that sets an evaluation period on the basis of the number of days of the determined day of operation stop; and a prediction section that predicts the occurrence of a failure of the industrial machine on the basis of the first motion data that occurs within the set evaluation period. BRIEF DESCRIPTION OF DRAWINGS

[0010] Figure 1 A diagram is shown for a failure prediction system including the failure prediction device of the present embodiment.

[0011] Figure 2 A hardware configuration diagram is shown for the failure prediction device of the present embodiment.

[0012] Figure 3 A functional block diagram is shown for the failure prediction device of the present embodiment.

[0013] Figure 4 FIG. 1 is a diagram showing one example of torque data stored in the storage section of the torque data storage section 102. Figure 3

[0014] Figure 5 FIG. 6 is a supplementary diagram for supplementing the explanation of the process of the determination section of the torque data storage section 102. Figure 3

[0015] Figure 6 FIG. 7 is a flowchart showing one example of the sequence of the process of the failure prediction device performed by the failure prediction device of the present embodiment.

[0016] Figure 7 FIG. 8 is a flowchart showing one example of the sequence of the setting process of the evaluation period of the process S1 of the failure prediction device of the present embodiment. Figure 6

[0017] Figure 8 FIG. 9 is a flowchart showing another first example of the sequence of the setting process of the evaluation period of the process S1 of the failure prediction device of the present embodiment. Figure 6

[0018] Figure 9 FIG. 10 is a diagram showing one example of the acceptance screen displayed by the process S23 of the failure prediction device of the present embodiment. Figure 8

[0019] Figure 10 FIG. 11 is a flowchart showing another second example of the sequence of the setting process of the evaluation period of the process S1 of the failure prediction device of the present embodiment. Figure 6

[0020] Figure 11 FIG. 12 is a flowchart showing another third example of the sequence of the setting process of the evaluation period of the process S1 of the failure prediction device of the present embodiment. Figure 6

[0021] Figure 12 FIG. 13 is a flowchart showing another fourth example of the sequence of the setting process of the evaluation period of the process S1 of the failure prediction device of the present embodiment. Figure 6 DETAILED DESCRIPTION

[0022] The failure prediction device of the present embodiment will be described below with reference to the accompanying drawings. In the following description, for structural elements having substantially the same function and structure, the same reference numerals are annotated, and repeated description is made only when necessary. Figure 1

[0023] Figure 1 ​​​​​​​​​A fault prediction system 1 including a fault prediction device 2 according to this embodiment is shown. The fault prediction system 1 includes multiple robot systems 100 (100a, 100b) and a fault prediction device 2 connected to the multiple robot systems 100 via a network 90, such as a LAN (Local Area Network). The robot systems 100 (100a, 100b) are composed of robot devices 110 (110a, 110b) and conveyor devices 130 (130a, 130b) that operate in conjunction with the robot devices 110. The fault prediction device 2 is connected to the robot control devices of the robot devices 110 via the network 90.

[0024] The fault prediction device 2 of this embodiment has a fault prediction function for predicting the occurrence of a fault in the robot device 110 using data representing the motion of the robot device 110 (motion data). The motion data of the robot device 110 generated during the evaluation period is used in calculating the fault prediction of the robot device 110. One of the features of the fault prediction device 2 of this embodiment is that the evaluation period for the motion data used in calculating the fault prediction of the robot device 110 is set based on the fact that the operation of the robot device 110 has stopped. This function allows the prediction of the occurrence of a fault in the robot device 110 using sufficient motion data even when the robot device 110 is stopped for an extended period of time, thereby suppressing the reduction in the accuracy of the fault prediction caused by the extended stoppage of the robot device 110.

[0025] Typically, the failure prediction device 2 of this embodiment is configured as follows.

[0026] like Figure 2 As shown, the failure prediction device 2 is a computer device in which hardware such as an operating device 4, a display device 5, a communication device 6, and a storage device 7 are connected to a processor 3 such as a CPU (Central Processing Unit).

[0027] The operating device 4 is implemented by a keyboard, a mouse, a knob, etc. The operating device 4 can also be implemented by a touch panel that also serves as a display device 5. The user can input various information to the fault prediction device 2 through the operating device 4. The display device 5 is implemented by an LCD (Liquid Crystal Display) and the like. Various screens are displayed on the display device 5 according to the control of the processor 3. The communication device 6 is implemented by a communication module that meets any communication standards. The communication device 6 sends and receives various data to and from external devices such as the robot device 110 according to the control of the processor 3. The storage device 7 is implemented by an HDD (Hard Disk Drive), an SSD (Solid State Drive), etc. A fault prediction program is stored in the storage device 7.

[0028] like Figure 3 As shown, the processor 3 executes the fault prediction program stored in the storage device 7, thereby the fault prediction device 2 of this embodiment functions as an input unit 21, a display unit 22, a receiving unit 23, a storage unit 24, a judgment unit 25, a setting unit 26 and a fault prediction unit 27.

[0029] The input unit 21 inputs user operations via the operating device 4 to the failure prediction device 2. Specifically, the input unit 21 inputs failure prediction instructions and the number of operation stop days used for setting the evaluation period to the failure prediction device 2.

[0030] The display unit 22 is Figure 2 The display unit 22 displays various screens related to failure prediction. For example, the display unit 22 displays a screen including a software button for accepting failure prediction instructions, a screen for accepting the number of downtime days used to calculate the evaluation period from the user, and a screen displaying the predicted failure occurrence date as a result of the failure prediction.

[0031] The receiving unit 23 receives Figure 2 The functions of the communication device 6 shown in FIG. are implemented. The receiving unit 23 receives torque data from the robot 110 as motion data for the robot 110. For example, the torque data is repeatedly measured at a predetermined interval by torque sensors attached to the motors driving the joints of the robot 110. The torque data received from the robot 110 is stored in the storage unit 24 in association with the date and time information generated by the robot 110 (measured by the torque sensors) or received by the robot 110.

[0032] The storage unit 24 Figure 4The function of the storage device 7 shown is realized by the storage section 24. The storage section 24 stores the torque data received by the robot device 110. For example, as shown in Figure 5 As shown, the torque data is aggregated in a text file for each date on which the torque data was generated. Date-time information for determining the date-time on which the torque data was generated is described in the text file together with the torque data (Trl, Tr2,..., Tr2784). The torque data is described in order received from the robot device 110. The date-time information includes the start date-time of measurement of the torque data, the end date-time of measurement of the torque data, the measurement period, and the like.

[0033] The determination section 25 determines the operation stop date of the robot device 110 on the basis of the torque data. Typically, the determination section 25 determines the day on which no torque data was generated as the operation stop date of the robot device 110. For example, as shown in Figure 6 As shown, the determination section 25 assigns an operation flag "0" to the day determined as the operation stop date, and assigns an operation flag "1" to the day not determined as the operation stop date, i.e., the operation date. By the determination processing of the operation stop date by the determination section 25, each day in the past is classified as the operation stop date or the operation date.

[0034] The setting section 26 sets an evaluation period corresponding to the number of days of the operation stop date of the robot device 110. The evaluation period indicates the period from the start date of the evaluation period to the end date of the evaluation period, and the evaluation days indicate the length of the evaluation period. In order to perform the setting processing of the evaluation period by the setting section 26, an initial evaluation period (hereinafter referred to as an initial period) is set in advance. Typically, the initial period is a period in which the day before the day on which the failure prediction instruction was input is set as the end date, and the day from the end date back by an initial evaluation day number (hereinafter referred to as an initial day number) set in advance is set as the start date. The end date can also be a day specified by the user. The initial evaluation day number can also be changed by the user.

[0035] When there is no operation stop day in the initial period, the setting unit 26 sets the initial period as the evaluation period. On the other hand, when there is an operation stop day in the initial period, the setting unit 26 sets the number of days obtained by adding the initial number of days to the number of days of the operation stop day as the length of the evaluation period. Specifically, the setting unit 26 sets the evaluation period to a period with the day before the day on which the fault prediction indication is input as the end date and the day obtained by adding the initial number of days to the number of days of the operation stop day from the end date as the start date. For example, when the initial period is 90 days from April 5, 2023 (start date) to June 6, 2023 (end date), and there are 3 operation stop days during the period, the evaluation period is set to 93 days from April 2, 2023 (start date) to June 6, 2023 (end date). Since the evaluation period includes three downtime days, the number of operating days included in the initial period when there are no downtime days in the initial period is the same as the number of operating days included in the evaluation period set when there are downtime days in the initial period.

[0036] The failure prediction unit 27 predicts the occurrence of a failure in the robot apparatus 110 based on the motion data generated during the evaluation period set by the setting unit 26. For example, based on the torque data generated during the evaluation period, the failure prediction unit 27 generates an evaluation expression representing the temporal variation of the torque data and calculates the time until the torque value represented by the torque data reaches a preset threshold. Based on the time until the torque value reaches the threshold and the current time, the failure prediction unit 27 predicts the date on which the failure of the robot apparatus 110 will occur.

[0037] The following reference Figure 6 The following describes the outline of the fault prediction process performed by the fault prediction device 2 of this embodiment. The fault prediction process is executed when a fault prediction instruction is input by the user. Figure 7 As shown, the failure prediction device 2 sets an evaluation period ( S1 ), executes a failure prediction process of the robot device 110 based on torque data generated during the evaluation period ( S2 ), and displays a prediction result ( S3 ).

[0038] The following reference Figure 6 ,illustrate Figure 7 An example of setting processing of the evaluation period of step S1. Figure 6As shown, when there is no operation stop day in the initial period of M days (S11; No), the failure prediction device 2 sets the initial period as the evaluation period (S12). By the process of S12, for example, the evaluation period is set to a period in which the day before the day on which the failure prediction instruction is input is the end day, and the day that is M days before the end day is the start day. On the other hand, when there is an operation stop day in the initial period of M days (S11; Yes), the failure prediction device 2 determines the number of days N of the operation stop day in the initial period (S13), and sets the evaluation period to an evaluation period of (M+N) days (S14). By the process of S14, for example, the evaluation period is set to a period in which the day before the day on which the failure prediction instruction is input is the end day, and the day that is M days before the end day plus N days of the operation stop day is the start day.

[0039] According to the failure prediction device 2 of the present embodiment, when there is no operation stop day in the initial period set in advance, the occurrence of a failure of the robot device 110 can be predicted based on the operation data generated in the initial period. On the other hand, when there is an operation stop day in the initial period set in advance, the occurrence of a failure of the robot device 110 can be predicted based on the operation data generated in the evaluation period obtained by adding the number of days of the operation stop day to the initial period.

[0040] For example, in a case where the number of days of the operation stop day in the initial period of 90 days is 0, the failure prediction device 2 performs failure prediction using the operation data generated in the initial period. On the other hand, in a case where the number of days of the operation stop day in the initial period of 90 days is 3, the failure prediction device 2 performs failure prediction based on the operation data generated in the evaluation period of 93 days obtained by adding the number of days of the operation stop day 3 to the initial number of days 90. That is, in any case, failure prediction can be performed using operation data of at least the number of days of operation 90. Therefore, regardless of the presence or absence of the operation stop day in the initial period, the occurrence of a failure of the robot device 110 can be predicted using operation data of the number of days of operation 90, and thus, variation in prediction accuracy due to the presence or absence of the operation stop day is suppressed, and high-accuracy prediction can be achieved.

[0041] According to Figure 8 As shown in the setting process of the evaluation period, when there is an operation stop day in the initial period, the evaluation period is automatically set to a period longer than the initial period. However, in a case where there is an unexpected continuous operation stop day that is not scheduled, the number of days of the operation stop day used for the setting of the evaluation period can be accepted from the user. The unexpected continuous operation stop day is, for example, a day on which a factory is stopped for a long time in order to prevent the spread of infectious diseases.

[0042] Hereinafter, other examples of the setting process of the evaluation period of the process S1 will be described with reference to Figure 9 , Figure 6 . Figure 8 As shown in Figure 9 , when there is no continuous operation stop day more than the number of days set in advance in the initial period of the initial number of days M days (S21; No), the failure prediction device 2 of the first modified example of the present embodiment sets the initial period as the evaluation period (S22). By the process of the process S22, for example, the evaluation period is set to a period in which the day before the day on which the failure prediction instruction is input is the end day, and the day that is M days before the end day is the start day. On the other hand, when there is a continuous operation stop day more than the number of days set in advance in the initial period of the initial number of days M days (S21; Yes), the failure prediction device 2 displays an acceptance screen for accepting the number of days of the operation stop day used for setting the evaluation period from the user (S23). Figure 9 One example of the acceptance screen is shown. As shown in Figure 6 , the number of days of the continuous operation stop day, the day of the week of the continuous operation stop day are displayed on the acceptance screen 200 as information related to the continuous operation stop day included in the initial period. In addition, the acceptance screen 200 is configured to be able to accept the number of days of the operation stop day used for the calculation of the evaluation period. Based on the decision button being clicked, the operation stop day specified by the user is input to the failure prediction device 2. The failure prediction device 2 sets the evaluation period of (M+P) days if the number of days of the operation stop day is accepted (S24). By the process of the process S25, for example, the evaluation period is set to a period in which the day before the day on which the failure prediction instruction is input is the end day, and the day that is M days before the end day plus the number of days P of the operation stop day input by the user (M+P) days is the start day.

[0043] According to the first variant of this embodiment, when a factory shuts down for an extended period to prevent the spread of an infectious disease, the fault prediction device 2 can set, based on user instructions, an evaluation period for the motion data used for fault prediction of the robot device 110 installed in the factory. For example, if the number of consecutive shutdown days is 30, by the user inputting a number of days greater than 30, fault prediction for the robot device 110 can be performed based on a larger amount of motion data than used in normal prediction processing, thereby improving the accuracy of fault prediction. Furthermore, if the number of consecutive shutdown days is six, including two weekend days that are public holidays, the user can also input four days as shutdown days for the evaluation period, taking public holidays into account. This allows comparison of fault predictions for a period that includes extended shutdown days with those that exclude these days under the same conditions.

[0044] according to Figure 10 The evaluation period setting process shown is to set the evaluation period so that, regardless of whether or not there are any downtime days within the initial period, the motion data for the initial number of days (M operating days) can be used to perform failure prediction of the robot apparatus 110. However, the evaluation period may be set so that, if there are downtime days within the initial period, the motion data for a number of days greater than the initial number of days (M) can be used to perform failure prediction of the robot apparatus 110.

[0045] Below, refer to Figure 6 ,illustrate Figure 10 Another example of setting processing of the evaluation period of step S1. Figure 6 As shown, when there is no operation stop day within the initial period of the initial number of days M (S31; No), the fault prediction device 2 of the second variant of the present embodiment sets the initial period as the evaluation period (S32). Through the processing of step S32, for example, the evaluation period is set to a period with the day before the day on which the fault prediction instruction is input as the end date and the day M days after the end date as the start date. On the other hand, when there is an operation stop day within the initial period of the initial number of days M (S31; Yes), the fault prediction device 2 determines the number of operation stop days N in the initial period (S33), and sets the added number of days Q corresponding to the determined number of operation stop days N (S34). Next, the fault prediction device 2 sets the evaluation period with the evaluation number of days (M+N+Q) days (S35). Through the processing of step S35, for example, the evaluation period is set to a period with the day before the day on which the fault prediction indication is input as the end date and the total number of days (M+N+Q) days, which is the initial number of days M, the number of days of operation suspension N, and the number of days added Q, traced back from the end date as the start date.

[0046] The failure prediction device 2 according to the second modification example of the present embodiment is able to predict the occurrence of a failure of the robot device 110 using action data of a larger number of days than when there is no operation stop day during the initial period, when there is an operation stop day during the initial period. For example, in a case where the initial period is 90 days and there are 3 operation stop days during the initial period, according to the setting process of the evaluation period shown in Figure 10 by tracing back the start day of the initial period by 3 days, it is able to predict the occurrence of a failure of the robot device 110 using action data of a total of 90 days. On the other hand, according to the setting process of the evaluation period shown in Figure 7 by tracing back the start day of the initial period by 6 days, it is able to predict the occurrence of a failure of the robot device 110 using action data of a total of 93 days even when considering the 3 operation stop days during the initial period. In this way, the failure prediction device 2 according to the second modification example is able to perform failure prediction using action data of a larger number of days than the failure prediction device 2 of the present embodiment when there is an operation stop day during the initial period, and thus it is possible to further suppress a decrease in the accuracy of failure prediction.

[0047] In the setting process of the evaluation period shown in Figure 8 , Figure 10 , Figure 11 the presence or absence of an operation stop day of the robot device 110 during the initial period is determined on the basis of action data of the robot device 110, i.e., torque data. However, the presence or absence of an operation stop day of the robot device 110 during the initial period can also be determined on the basis of action data indicating the action of the conveyor device 130 that cooperates with the robot device 110.

[0048] The reception unit 23 receives torque data from the robot device 110 as action data indicating the action of the robot device 110. In addition, the reception unit 23 receives speed data from the conveyor device 130 as action data indicating the action of the conveyor device 130. The speed data indicates the carrying speed of the conveyor device 130 and is repeatedly measured by a speed sensor mounted in the conveyor device 130. The action data of the conveyor device 130 is stored in the storage unit 24 in association with date and time information generated by the conveyor device 130 or received from the conveyor device 130 or the like. Furthermore, the action data indicating the action of the robot device 110 or the action data indicating the action of the conveyor device 130 can also be received from a comprehensive control device that comprehensively controls the robot device 110 and the conveyor device 130.

[0049] The storage unit 24 stores the torque data received from the robot device 110 and the action data of the conveyor device 130 received from the conveyor device 130.

[0050] The determination unit 25 determines the operation stop day of the robot device 110 based on the operation data of the conveyor device 130. Specifically, the determination unit 25 calculates the operation stop time of the conveyor device 130 based on the operation data of the conveyor device 130 per day, and converts the calculated operation stop time of the conveyor device 130 into the operation stop day of the robot device 110. For example, the operation stop time of 24 hours of the conveyor device 130 is converted into the operation stop day of 1 day of the robot device 110, and the operation stop time of 12 hours of the conveyor device 130 is converted into the operation stop day of 0.5 day of the robot device 110. Of course, the ratio of the operation stop time of the conveyor device 130 to the operation stop day of the robot device 110 can not be 1 to 1.

[0051] The conversion method can also be changed according to the maximum operation time of 1 day of the robot device 110. For example, in a case where the maximum operation time of 1 day of the robot device 110 is set to 8 hours, the operation stop time of the conveyor device 130 of less than 16 hours is converted into the operation stop day of 0 day of the robot device 110, and the operation stop time of the conveyor device 130 of 20 hours is converted into the operation stop day of 0.5 day of the robot device 110.

[0052] Of course, the actual operation time of the robot device 110 per unit time, such as one cycle, one hour, and the like, is different from the operation time of the conveyor device 130 (peripheral device), and thus the conversion method can also be changed based on the actual operation time of the robot device 110 and the conveyor device 130 (peripheral device) per unit time.

[0053] Hereinafter, other examples of the setting process of the evaluation period of the process S1 of Figure 6 will be described with reference to Figure 11 . As described above, the evaluation period of the process S1 of Figure 7When there is no operation stop time of the conveyor device 130 in the initial period of M days (S41; No), the failure prediction device 2 of the third modification example sets the initial period as the evaluation period (S42). By the process of S42, for example, the evaluation period is set to a period in which the day before the day on which the failure prediction instruction is input is the end day, and the day that is M days before the end day is the start day. On the other hand, when there is an operation stop time of the conveyor device 130 in the initial period of M days (S41; Yes), the failure prediction device 2 converts the operation stop time of the conveyor device 130 per day into the operation stop days R of the robot device 110 (S43), and sets the evaluation period of (M+R) days (S44). By the process of S44, for example, the evaluation period is set to a period in which the day before the day on which the failure prediction instruction is input is the end day, and the day that is M days plus the number of days R of the robot device 110 that is converted from the operation stop time of the conveyor device 130 from the end day is the start day.

[0054] The failure prediction device 2 according to the third modification example of the present embodiment can achieve the same effects as the failure prediction device 2 of the present embodiment, and can judge the operation stop days of the robot device 110 based on the action data of the peripheral device that cooperates with the robot device 110, so that the degree of freedom of the action data to be processed can be increased.

[0055] In the evaluation period setting process shown in Figure 8 , Figure 10 , Figure 11 , Figure 12 In the evaluation period setting process shown in

[0056] Hereinafter, other examples of the evaluation period setting process of S1 of Figure 6 will be described. As Figure 12 ​ ​As shown, the failure prediction device 2 of the fourth modification example initializes the number of operation days a on the day to the value 0 (S51), and initializes the number of traced days b to the value 1 (S52). The failure prediction device 2 determines whether or not the day b days before the day on which the failure prediction instruction was received is an operation day (S53). When the day b days before the day on which the failure prediction instruction was received is a stop day (S53; No), the value of the number of traced days b is incremented (S54), and the processing returns to the determination processing of step S53. When the day b days before the day on which the failure prediction instruction was received is an operation day (S53; Yes), the failure prediction device 2 increments the value of the number of operation days a (S55), and determines whether or not the number of operation days a has reached the predetermined prescribed number of days (S56). When the number of operation days a has not reached the prescribed number of days (S56; No), the value of the number of traced days b is incremented (S54), and the processing returns to the determination processing of step S53. The processing of steps S53 to S56 is repeatedly executed until the number of operation days a reaches the prescribed number of days. When the number of operation days a has reached the prescribed number of days (S56; Yes), the failure prediction device 2 sets the evaluation period to the period in which the day before the day on which the failure prediction instruction was input is the end day, and the day b days before the end day is the start day (S57).

[0057] One of the features of the failure prediction device 2 of the present embodiment is that the evaluation period of the operation data used for the calculation of the failure prediction of the robot device 110 is set based on the fact that the operation of the robot device 110 has stopped. Therefore, the object of the failure prediction is not limited to the robot device 110. For example, the object of the failure prediction can be an industrial machine such as a machine tool, a lathe machine, a conveyance device, a welding device, or the like.

[0058] In the present embodiment, the conveyer device 130 is employed as the peripheral device that cooperates with the robot device 110, and the stop day of the robot device 110 is determined based on the stop time of the operation of the conveyer device 130, but as long as the stop day of the robot device 110 can be determined, the peripheral device that cooperates with the robot device 110 is not limited to the conveyer device 130. For example, the peripheral device can be another robot device 110 or another various device that cooperates with the robot device 110 to perform a prescribed processing.

[0059] In the present embodiment, the day on which the torque data is not generated is determined as the stop day of the operation. However, the day on which the operation time is short based on the torque data can also be determined as the stop day of the operation. For example, the day on which the operation time is short can be the day on which the operation time is intentionally shortened, the day on which the maintenance of the robot device 110 is performed, the day on which the robot device 110 is test-operated, the day on which the robot device 110 is not normally operated, the day on which the operation is unstable and only a short time operation is possible, or the like.

[0060] For example, assume that the initial period is 90 days, the maximum operating time per day is 12 hours, and the day on which the robot apparatus 110 has operated for only 2 hours during the initial period is 1 day. In this case, the day on which the robot apparatus 110 has operated for only 2 hours is counted as a stop day, and the evaluation period is set to 91 days. The operation data of the day on which the robot apparatus 110 has operated for only 2 hours can also be used for failure prediction. The operation data of the day on which the robot apparatus 110 has operated for only 2 hours can have a large influence on failure prediction. According to the present embodiment, on the basis of the operation data of the days on which the robot apparatus 110 has operated for 90 days, failure prediction can be performed using the operation data of the stop days on which the operating time is less than the prescribed time, and thus it is possible to improve the accuracy of failure prediction compared to failure prediction using the operation data of only the days on which the robot apparatus 110 has operated for more than the prescribed time.

[0061] In the present embodiment, torque data related to torque values is used as data used for calculation of failure prediction of the robot apparatus 110, but the data is not limited to torque data as long as failure prediction of the robot apparatus 110 can be calculated. For example, arbitrary data indicating the operation of the robot apparatus 110, such as data related to current values flowing to the motor, data related to disturbance torque, and the like can be used.

[0062] The various data such as the failure prediction program stored in the storage device can also be recorded in a removable medium and distributed to users, and can also be distributed by being downloaded to the failure prediction apparatus 2 via a network.

[0063] With respect to the present embodiment and the modified example, the following notes are further disclosed.

[0064] (Note 1)

[0065] The failure prediction apparatus 2 has a determination section 25 that executes determination processing of a stop day of the industrial machine on the basis of first operation data indicating the operation of the industrial machine or second operation data indicating the operation of a peripheral apparatus that cooperates with the industrial machine, in conjunction with input of a failure prediction instruction, a setting section 26 that sets an evaluation period on the basis of the number of days of the stop day determined, and a prediction section 27 that predicts the occurrence of failure of the industrial machine on the basis of the first operation data generated in the set evaluation period.

[0066] (Note 2)

[0067] The evaluation period described in Note 1 is set to a length of days obtained by adding the number of days of the stop day to the initially set evaluation days.

[0068] (Note 3)

[0069] The setting unit 26 described in Note 2 sets an evaluation period, wherein the evaluation period ends on the day before the day when the fault prediction indication is input and starts on the day obtained by adding the number of days of the pre-set initial evaluation days plus the number of days of the operation stoppage from the end date.

[0070] (Note 4)

[0071] The setting unit 26 described in Note 1 sets the length of the evaluation period to the number of days obtained by adding the preset initial evaluation days to the number of days specified by the user when the number of consecutive days of operation suspension exceeds a preset threshold value, and sets the length of the evaluation period to the initial evaluation days when the number of consecutive days of operation suspension does not exceed the preset threshold value.

[0072] (Note 5)

[0073] The setting unit 26 described in Supplementary Note 1 sets the evaluation period so that the number of operating days of the industrial machine within the evaluation period becomes a preset number of days.

[0074] (Note 6)

[0075] The determination unit 25 according to any one of Supplementary Notes 1 to 5 converts the daily operation stop time of the peripheral device into the operation stop day of the industrial machine based on the second operation data.

[0076] (Note 7)

[0077] The determination unit 25 according to any one of Supplementary Notes 1 to 5 determines that the day on which the first motion data is not generated or the day on which the time when the first motion data is generated does not reach a predetermined time is the day on which the operation of the industrial machine is stopped.

[0078] (Note 8)

[0079] The industrial machine described in any one of Notes 1 to 7 is a robot device 110, the first motion data is data related to the load torque applied to the drive shaft of the motor that drives the joint of the robot device 110, and the prediction unit 27 predicts the occurrence of a drive shaft failure based on the data related to the load torque generated within a set evaluation period.

[0080] (Note 9)

[0081] A program that enables a computer to function as:

[0082] a unit for executing a segment determination process, which, upon input of a fault prediction indication, determines a day on which the operation of the industrial machine should be stopped based on first motion data indicating the motion of the industrial machine or second motion data indicating the motion of a peripheral device operating in cooperation with the industrial machine;

[0083] The unit that sets the evaluation period sets the evaluation period based on the number of days of the operation stop day judged.

[0084] The unit that predicts the occurrence of a failure predicts the occurrence of a failure of the industrial machine based on the first action data generated within the set evaluation period.

[0085] Embodiments of the present disclosure are described in detail, but the present disclosure is not limited to each of the above-described embodiments. These embodiments can be variously added, substituted, changed, partially deleted, and the like, within a range not departing from the gist of the invention, or within a range not departing from the scope of the invention and the gist of the invention derived from the content described in the claims and the equivalent content thereof. For example, in the above-described embodiments, the order of each action or the order of each process is shown as one example, but is not limited thereto. The same applies to the case where a numerical value or a formula is used in the description of the above-described embodiments.

[0086] Explanation of Reference Signs

[0087] 1: failure prediction system, 2: failure prediction device, 3: processor, 4: operation device, 5: display device, 6: communication device, 7: storage device, 21: input unit, 22: display unit, 23: reception unit, 24: storage unit, 25: judgment unit, 26: setting unit, 27: failure prediction unit, 90: network, 100: robot system, 110: robot device, 130: conveyor device.

Claims

1. A failure prediction device characterized by comprising: having: a determination section that, in conjunction with input of a failure prediction instruction, performs a determination process of operation stop days of an industrial machine based on first action data representing actions of the industrial machine or second action data representing actions of a peripheral device that cooperates with the industrial machine; a setting section that sets an evaluation period based on the number of operation stop days of the determination process; and a prediction section that predicts occurrence of a failure of the industrial machine based on the first action data generated within the set evaluation period.

2. The failure prediction device according to claim 1, wherein the evaluation period is set to a length of a number of days obtained by adding an initially set evaluation number of days to the number of operation stop days.

3. The failure prediction device according to claim 2, wherein the setting section sets the evaluation period as a period in which a day before the day on which the failure prediction instruction is input is the end day and a day obtained by adding the initially set evaluation number of days to the number of operation stop days from the end day is the start day.

4. The failure prediction device according to claim 1, wherein the setting section sets the length of the evaluation period to a number of days obtained by adding an initially set evaluation number of days to a number of days specified by a user when the number of consecutive operation stop days exceeds a threshold value set in advance, and sets the length of the evaluation period to the initially set evaluation number of days when the number of consecutive operation stop days does not exceed the threshold value set in advance.

5. The failure prediction device according to claim 1, wherein the setting section sets the evaluation period in such a manner that the number of operation days of the industrial machine within the evaluation period becomes a number set in advance.

6. The failure prediction device according to any one of claims 1 to 5, wherein the determination section converts an operation stop time per day with respect to the peripheral device into the number of operation stop days of the industrial machine based on the second action data.

7. The failure prediction device according to any one of claims 1 to 5, wherein the determination section determines a day on which the first action data is not generated or a day on which a time at which the first action data is generated does not reach a prescribed time as the operation stop day of the industrial machine.

8. The failure prediction device according to any one of claims 1 to 7, wherein the industrial machine is a robot device, the first action data is data related to a load torque applied to a drive shaft of a motor that drives a joint section of the robot device, the prediction section predicts occurrence of a failure of the drive shaft based on data related to the load torque generated within the set evaluation period. a computer functions as:

9. A program, characterized by, a section that, in conjunction with input of a failure prediction instruction, performs a determination process of operation stop days of an industrial machine based on first action data representing actions of the industrial machine or second action data representing actions of a peripheral device that cooperates with the industrial machine; ​ The unit sets an evaluation period based on the number of days of operation stop days from the judgment processing; and The unit predicts the occurrence of a failure of the industrial machine based on the first action data generated within the set evaluation period.

Images