Data processing system, data processing method, and program

Abstract

The present invention makes it difficult to bring about a misleading deterioration diagnosis for a servo system (7). This data processing system comprises an acquisition unit (2), a first generation unit (3A), a second generation unit (3B), and a presentation unit (47). The acquisition unit (2) acquires at least one signal among a control signal of the servo system (7) and detection signals from a sensor (61). The first generation unit (3A) generates first data pertaining to three or more feature amounts from a prescribed area in the waveform of the signal. The second generation unit (3B) generates three or more pieces of second data respectively indicating the tendencies of change, over time, of the first data pertaining to three or more feature amounts. The presentation unit (47) compares the three or more pieces of second data, and when there is a specific piece of the second data showing a different tendency of change, executes specific processing pertaining to target data that is first data which is a cause of the different tendency of change, and presents the specific piece of the second data. The specific processing includes processing pertaining to at least one among non-presentation of the target data, correction of the target data, or notification pertaining to the target data.

Description

Technical Field

[0001] The present disclosure generally relates to a data processing system, a data processing method, and a program. More specifically, the present disclosure relates to a data processing system, a data processing method, and a program used for deterioration diagnosis of a servo system. Background Art

[0002] The deterioration estimation device described in Patent Document 1 includes an inspection result acquisition unit, an attribute data acquisition unit, and an estimation unit. The inspection result acquisition unit acquires an inspection result of equipment to be maintained. The attribute data acquisition unit acquires attribute data of the equipment to be maintained. The estimation unit estimates a deterioration rank of the equipment to be maintained by inputting the inspection result and the attribute data of the equipment to be maintained. Prior Art Literature Patent Documents

[0003] Patent Document 1 Japanese Unexamined Patent Publication No. 2020-160528 Summary of the Invention

[0004] By the way, information related to deterioration diagnosis may exhibit temporary irregular tendencies due to the influence of noise and the like. Presentation of such information may lead to misunderstanding (misleading) for a user.

[0005] The present disclosure has been made in view of the above circumstances, and an object thereof is to provide a data processing system, a data processing method, and a program that make it less likely to cause misleading regarding deterioration diagnosis.

[0006] A data processing system according to one aspect of the present disclosure is used for diagnosing the deterioration of at least one of a load and a servo motor in a servo system including a load, a servo amplifier, and a servo motor that powers the load in accordance with the control of the servo amplifier. The data processing system comprises an acquisition unit, a first generation unit, a second generation unit, and a presentation unit. The acquisition unit acquires at least one signal from among control signals used to control the servo system and detection signals output from a sensor that detects the state of the servo system. The first generation unit generates first data relating to three or more feature quantities from a predetermined region of the signal waveform of the at least one signal. The second generation unit generates three or more second data, which are time-series data showing the progression of deterioration of the target, and each shows the change trend of the first data relating to the three or more feature quantities over time. The presentation unit compares the three or more second data with each other. If there is a specific second data that exhibits a different change trend from the other second data, the presentation unit performs specific processing on the target data, which is the first data that caused the different change trend, and presents the specific second data. The specified processing includes processing relating to at least one of the following: not presenting the target data, correcting the target data, and providing notification regarding the target data.

[0007] Another aspect of the present disclosure relates to a data processing method for diagnosing the deterioration of at least one of a load and a servo motor in a servo system including a load, a servo amplifier, and a servo motor that powers the load in accordance with the control of the servo amplifier. The data processing method includes an acquisition step, a first generation step, a second generation step, and a presentation step. In the acquisition step, at least one signal is acquired from among control signals used to control the servo system and detection signals output from a sensor that detects the state of the servo system. In the first generation step, first data relating to three or more feature quantities is generated from a predetermined region of the signal waveform of the at least one signal. In the second generation step, three or more second data are generated, which are time-series data showing the progression of deterioration of the object, and each shows the change trend of the first data relating to the three or more feature quantities over time. In the presentation step, the three or more second data are compared with each other. In the presentation step, if there is a specific second data that exhibits a different change trend from the other second data, the specific second data is presented after performing a specific processing on the target data which is the first data that caused the different change trend. The specified processing includes processing relating to at least one of the following: not presenting the target data, correcting the target data, and providing notification regarding the target data.

[0008] Another aspect of the present disclosure relates to a program that causes one or more processors to execute the data processing method described above.

[0009] According to this disclosure, there is an advantage in that it makes it less likely to mislead people regarding deterioration diagnosis. [Brief explanation of the drawing]

[0010] [Figure 1] Figure 1 is a schematic block diagram of the entire system, including a data processing system according to one embodiment. [Figure 2] Figure 2 shows an example of a screen display in the data processing system described above. [Figure 3] Figure 3 is a graph showing the waveform of the current used as a diagnostic parameter in the data processing system described above. [Figure 4A] Figure 4A is a graph used to explain the first data point related to feature A in the same data processing system. [Figure 4B] Figure 4B is a graph used to explain the first data point related to feature B in the same data processing system. [Figure 4C] Figure 4C is a graph used to explain the first data point related to feature C in the same data processing system. [Figure 5] Figure 5 is a diagram illustrating the second data displayed from the display device and its overall confidence level in the data processing system described above. [Figure 6] Figure 6 is a flowchart illustrating the operation of the data processing system described above. [Modes for carrying out the invention]

[0011] (overview) The following describes a data processing system, data processing method, and program according to the embodiments, with reference to the drawings. However, the embodiments described below are only one of many embodiments of this disclosure. The embodiments described below can be modified in various ways depending on the design, etc., as long as the objectives of this disclosure are achieved.

[0012] (Feature 1) Figure 1 is a schematic block diagram of the entire system including a data processing system 1 according to one embodiment. As shown in Figure 1, the data processing system 1 according to one embodiment comprises an acquisition unit 2 and a generation unit 3. The acquisition unit 2 acquires at least one of a control value and a detection value as an input value. The control value is used to control a servo system 7 including a load 73, a servo amplifier 71, and a servo motor 72 that provides power to the load 73 in accordance with the control of the servo amplifier 71. The detection value is output from a sensor 61 that detects the state of the servo system 7. The generation unit 3 performs at least an extraction process on the input value acquired by the acquisition unit 2 to generate diagnostic parameters. The diagnostic parameters are used to diagnose the deterioration of at least one of the target OB1 of the load 73 and the servo motor 72 in the servo system 7. The input value is a value that changes over time. In the extraction process, the generation unit 3 extracts values ​​for a portion of the input value over a certain period as diagnostic parameters.

[0013] According to the data processing system 1 described above, the degradation diagnosis can use the values ​​for the period extracted by the generation unit 3 from the input values. In this way, by removing features from the input values ​​that may hinder the degradation diagnosis, the accuracy of the degradation diagnosis can be improved compared to when the values ​​for the entire period of the input are used for the degradation diagnosis.

[0014] A data processing method according to one embodiment includes an acquisition step and a generation step. In the acquisition step, at least one of a control value and a detection value is acquired as an input value. The control value is used to control a servo system 7 including a load 73, a servo amplifier 71, and a servo motor 72 that provides power to the load 73 in accordance with the control of the servo amplifier 71. The detection value is output from a sensor 61 that detects the state of the servo system 7. In the generation step, at least an extraction process is performed on the input value acquired in the acquisition step to generate diagnostic parameters. The diagnostic parameters are used to diagnose the deterioration of at least one of the target OB1 of the load 73 and the servo motor 72 in the servo system 7. The input value is a value that changes over time. In the generation step, in the extraction process, values ​​for a portion of the input value over a certain period are extracted as diagnostic parameters.

[0015] This data processing method is used on a computer system (data processing system 1). That is, this data processing method can also be embodied as a program. A program according to one aspect is a program for causing one or more processors to execute the above data processing method. The program may be recorded on a computer-readable non-transitory recording medium.

[0016] (Characteristic 2) Further, as shown in FIG. 1, the data processing system 1 according to one aspect is used for deterioration diagnosis of at least one target object OB1 of a load 73 and a servo motor 72 in a servo system 7. The servo system 7 includes the load 73, a servo amplifier 71, and the servo motor 72 that supplies power to the load 73 in accordance with control by the servo amplifier 71. Hereinafter, as an example, it is assumed that the target object OB1 is the load 73, and the data processing system 1 is used for deterioration diagnosis of the load 73. However, the target object OB1 may be the servo motor 72, or may be both the load 73 and the servo motor 72. The load 73 mentioned herein is not particularly limited, and is, for example, a ball screw, a gear, a belt, or the like. The servo system 7 is used, for example, to perform a predetermined operation in a manufacturing process of a product (or semi-finished product).

[0017] As shown in FIG. 1, the data processing system 1 includes an acquisition unit 2, a first generation unit 3A, a second generation unit 3B, and a presentation unit 47.

[0018] The acquisition unit 2 acquires at least one signal among a control signal used for controlling the servo system 7 and a detection signal output from a sensor 61 that detects a state of the servo system 7.

[0019] 4A, 4B and 4C are graphs for explaining first data D1 and second data D2 relating to feature amount A, feature amount B, and feature amount C, respectively, in the data processing system 1. The first generating unit 3A generates first data D1 (see FIGS. 4A to 4C) relating to three or more feature amounts from a predetermined region (all or part of a feature period in which a predetermined feature appears) in a signal waveform of at least one signal. The "feature amount" as used herein is assumed to be, for example, a statistical feature amount, and specifically, it is an average (value), a standard deviation (value), a maximum (value), a minimum (value), a histogram feature, or the like in the predetermined region. Alternatively, the "feature amount" may be a first principal component obtained by so-called "Principal Component Analysis". The "three or more feature amounts" may be feature amounts of the same type as each other, or may be feature amounts of different types. For example, when the first generating unit 3A generates first data D1 relating to three feature amounts from a predetermined region in one signal waveform (current signal waveform), the "three feature amounts" may be an average, a standard deviation, and a maximum, respectively. On the other hand, for example, when the first generating unit 3A generates first data D1 relating to three feature amounts from predetermined regions in three signal waveforms (e.g., current, torque, and speed signal waveforms), the "three feature amounts" may all be averages, for example. Hereinafter, as an example, the first generating unit 3A generates first data D1 relating to three feature amounts from predetermined regions in three signal waveforms, respectively. In FIGS. 4A to 4C, the first data D1 is illustrated as plots.

[0020] The second generating unit 3B generates three or more pieces of second data D2 (see FIGS. 4A to 4C), which are time-series data indicating the progress of deterioration of the object OB1 and each indicate a change tendency of the first data D1 relating to three or more feature amounts with the passage of time. In FIGS. 4A to 4C, each second data D2 is illustrated as polygonal line data obtained by connecting a plurality of first data D1 (plots) arranged in time series.

[0021] The presentation unit 47 compares three or more second data D2s with each other. If there is a specific second data D20 (see Figure 4C) that exhibits a different change trend from the other second data D2s, the presentation unit 47 performs specific processing on the target data D10, which is the first data D1 that caused the different change trend, and presents the specific second data D20. The specific processing includes processing related to at least one of the following: not presenting the target data D10, correcting the target data D10, and notifying the target data D10. For example, if the change trends of more than half of the three or more second data D2s show a monotonically decreasing trend, but less than half (e.g., one) of the second data D2s exhibit a change trend that does not correspond to a monotonically decreasing trend, then the latter second data D2 corresponds to the specific second data D20. In the example in Figure 4C, a specific second data point D20 has a period (times t2 to t3) where it temporarily shows an increasing trend, and the first data point D1 at time t3 can be determined to be the target data point D10 that caused a different trend in change.

[0022] According to the data processing system 1 described above, if there is a specific second data D20 that exhibits a different change trend, the specific second data D20 is presented after applying specific processing to the target data D10 that caused the different change trend. Therefore, users are less likely to mistake target data D10, which is highly likely to have been generated by the influence of noise, etc., for the true data. As a result, the data processing system 1 has the advantage of making it less likely to mislead regarding degradation diagnosis.

[0023] Furthermore, a data processing method relating to one embodiment relates to the deterioration diagnosis of the target OB1 described above. The data processing method includes an acquisition step, a first generation step, a second generation step, and a presentation step. In the acquisition step, at least one signal is acquired from among the control signals used to control the servo system 7 and the detection signals output from the sensor 61 that detects the state of the servo system 7. In the first generation step, first data D1 relating to three or more feature quantities is generated from a predetermined region of the signal waveform of at least one signal. In the second generation step, three or more second data D2s are generated, which are time-series data showing the progression of deterioration of the target OB1, and each shows the change trend over time of the first data D1 relating to three or more feature quantities. In the presentation step, the three or more second data D2s are compared with each other. Then, in the presentation step, if there is a specific second data D20 that exhibits a different change trend from the other second data D2s, specific processing is performed on the target data D10, which is the first data D1 that caused the different change trend, and the specific second data D20 is presented. The specific processing includes processing related to at least one of the following: not presenting the target data D10, correcting the target data D10, and providing notification regarding the target data D10. The above data processing method has the advantage of making it less likely to mislead regarding deterioration diagnosis. This data processing method is used on a computer system (data processing system 1). In other words, this data processing method can also be implemented as a program. A program according to one embodiment is a program that causes one or more processors to execute the above data processing method. The program may be recorded on a computer-readable non-temporary recording medium.

[0024] (Feature 3) Furthermore, one embodiment of the data processing system 1 is a data processing system 1 that outputs diagnostic parameters. The diagnostic parameters are used for diagnosing the deterioration of at least one of the target OB1 of the load 73 and the servo motor 72 in a servo system 7 which includes a load 73, a servo amplifier 71, and a servo motor 72 that provides power to the load 73 according to the control of the servo amplifier 71. The diagnostic parameters include at least one of the following: a control value used to control the servo system 7, a detection value output from a sensor 61 that detects the state of the servo system 7, and a generated value generated from the control value or the detection value. The data processing system 1 comprises an acquisition unit 2, a reliability determination unit 44, and a reliability output unit (output unit 51). The acquisition unit 2 acquires the diagnostic parameters. The reliability determination unit 44 adds information regarding the reliability of the diagnostic parameters in the deterioration diagnosis of the servo system 7 to the diagnostic parameters. The reliability output unit outputs the reliability information linked to the diagnostic parameters to a presentation device (display device 81).

[0025] According to the data processing system 1 described above, the user can determine whether or not to use a specific diagnostic parameter for deterioration diagnosis by referring to the confidence level presented on the display device. This increases convenience for the user.

[0026] One embodiment of the data processing method is a data processing method that outputs diagnostic parameters. The diagnostic parameters are used for the deterioration diagnosis of at least one of the target OB1 of the load 73 and the servo motor 72 in a servo system 7 which includes a load 73, a servo amplifier 71, and a servo motor 72 that provides power to the load 73 according to the control of the servo amplifier 71. The diagnostic parameters include at least one of the following: a control value used to control the servo system 7, a detection value output from a sensor 61 that detects the state of the servo system 7, and a generated value generated from the control value or the detection value. The data processing method includes an acquisition step, a reliability determination step, and a reliability output step. In the acquisition step, the diagnostic parameters are acquired. In the reliability determination step, information regarding the reliability of the diagnostic parameters in the deterioration diagnosis of the servo system 7 is attached to the diagnostic parameters. In the reliability output step, the reliability information is linked to the diagnostic parameters and output to a display device (display device 81).

[0027] This data processing method is used on a computer system (data processing system 1). In other words, this data processing method can also be implemented as a program. A program according to one embodiment is a program that causes one or more processors to execute the above data processing method. The program may be recorded on a non-temporary recording medium that is readable by a computer.

[0028] (detail) (1) Overall structure As shown in Figure 1, the data processing system 1 is used together with the sensor 61, the higher-level controller 62, the servo system 7, the display device 81, and the operating device 82.

[0029] The data processing system 1 performs a degradation diagnosis of the servo system 7 using information acquired from at least one of the sensor 61 and the higher-level controller 62. The display device 81 displays the diagnosis results. In addition, the user of the data processing system 1 can make settings related to the degradation diagnosis by operating the control device 82. By making appropriate settings, the accuracy of the degradation diagnosis can be improved.

[0030] The data processing system 1 is located in a place away from the facility where the servo system 7 is installed (e.g., a factory). The sensor 61, the higher-level controller 62, the display device 81, and the operating device 82 are installed in the facility where the servo system 7 is installed. The sensor 61, the higher-level controller 62, the display device 81, and the operating device 82 communicate with the data processing system 1 via a wide-area network such as the Internet.

[0031] (2) Servo system The servo system 7 includes a servo amplifier 71, a servo motor 72, and a load 73. The data processing system 1 is used to diagnose the deterioration of the target OB1 of at least one of the load 73 and the servo motor 72 (in this case, the load 73). The servo motor 72 may be a linear motor or a rotary motor. In this embodiment, the case where the servo motor 72 is a rotary motor will be described as an example.

[0032] The servo motor 72 has an output shaft and rotates the output shaft according to the control of the servo amplifier 71. The load 73 is connected to the output shaft of the servo motor 72. The load 73 is powered by the servo motor 72. An example of the load 73 is a ball screw, gear, or belt. The servo system 7 is used, for example, to perform a predetermined task in the manufacturing process of a product (which may be a semi-finished product).

[0033] When the servo system 7 deteriorates, malfunctions such as abnormal noise, oscillation, slippage of the load 73, and a decrease in the operational accuracy of the load 73 may occur. The data processing system 1 performs a deterioration diagnosis of the servo system 7, allowing the user to know whether or not the servo system 7 has deteriorated.

[0034] The causes of malfunction in the servo system 7 include, for example, the adhesion of foreign matter to the servo motor 72 or load 73, wear of the servo motor 72 or load 73, and damage to the servo motor 72 or load 73. Malfunctions in the servo system 7 manifest as detectable characteristics such as fluctuations in the torque of the servo motor 72, changes in resonance characteristics, and changes in frictional force. Based on these characteristics, the data processing system 1 performs a deterioration diagnosis.

[0035] The data processing system 1 may perform a degradation diagnosis while the servo system 7 is performing a predetermined task, or it may perform a degradation diagnosis by having the servo system 7 undergo a test run while the predetermined task is paused.

[0036] (3) Sensor Sensor 61 detects the state of the servo system 7. Sensor 61 outputs a detection signal (electrical signal) including the detected value to the servo amplifier 71. The servo amplifier 71 controls the operation of the servo motor 72 based on the detection signal and a control signal (control value) from the higher-level controller 62, which will be described later. In this embodiment, sensor 61 also outputs the detection signal to the data processing system 1. It is preferable that multiple types of sensors 61 are provided. In this embodiment, it will be explained that multiple sensors 61 include a current sensor, a torque sensor, a speed sensor, and a position sensor (encoder, etc.).

[0037] The current sensor detects the current supplied to the servo motor 72. The torque sensor detects the torque of the servo motor 72. The speed sensor detects the rotational speed of the servo motor 72. The position sensor detects the position of the object to be detected as it moves in accordance with the rotation of the servo motor 72. For example, the position sensor detects the rotation angle of the servo motor 72. Also, for example, if the load 73 is a ball screw, the position sensor detects the position of the ball screw or a member connected to the ball screw in the axial direction of the ball screw. Alternatively, the position of the object to be detected may be detected by a camera instead of the position sensor.

[0038] (4) Higher-level controller The higher-level controller 62 outputs a control signal to the servo amplifier 71. This allows the higher-level controller 62 to control the operation of the servo system 7. The control signal includes control values, such as at least one of the command values ​​for the rotational speed, rotational angle, and torque of the servo motor 72. The servo amplifier 71 adjusts the power supplied to the servo motor 72 according to this control signal and the detection signal from the sensor 61, thereby controlling the operation of the servo motor 72.

[0039] Furthermore, the higher-level controller 62 also outputs a control signal including the (first) control value to the data processing system 1. The data processing system 1 performs a degradation diagnosis based on the control signal (first control value). The data processing system 1 may also perform a degradation diagnosis based on the detection signal from the sensor 61.

[0040] Incidentally, the servo amplifier 71 has a power conversion unit and generates a control signal (second control value) for adjusting the power supplied from the power conversion unit to the servo motor 72. The servo amplifier 71 may output a control signal including the second control value to the data processing system 1. In this case, the data processing system 1 may acquire the control signal generated by the servo amplifier 71 and perform a degradation diagnosis based on the control signal. In other words, the "control signal used to control the servo system 7" in this disclosure may be a control signal including a first control value from the higher-level controller 62, or a control signal including a second control value generated by the servo amplifier 71. The "control value used to control the servo system 7" may be the first control value or the second control value.

[0041] The various detection values ​​output from sensor 61, the various (first) control values ​​output from higher-level controller 62, and the (second) control value output from servo amplifier 71 each correspond to input values ​​acquired by acquisition unit 2. That is, the input values ​​(control values ​​and detection values) as diagnostic parameters include at least one of the following: the value of the current supplied to servo motor 72, the torque of servo motor 72, the speed of servo motor 72, and the position value (rotation angle) of servo motor 72.

[0042] (5) Data processing system (5.1) Components The data processing system 1 includes a computer system having one or more processors and memory. At least some of the functions of the data processing system 1 are realized by the execution of a program recorded in the memory of the computer system by the processor of the computer system. The program may be recorded in memory, provided via a telecommunication line such as the Internet, or provided on a non-temporary recording medium such as a memory card.

[0043] The data processing system 1 comprises an acquisition unit 2, a generation unit 3, a processing unit 4, an output unit 51, a receiving unit 52, and a storage unit 53. Note that the acquisition unit 2, generation unit 3, processing unit 4, output unit 51, and receiving unit 52 merely represent functions implemented by one or more processors and do not necessarily represent an actual physical configuration.

[0044] (5.2) Acquisition part The acquisition unit 2 acquires diagnostic parameters. For example, the data processing system 1 further includes a communication interface device, and the acquisition unit 2 acquires diagnostic parameters via the communication interface device.

[0045] As shown in Figure 1, diagnostic parameters include input values ​​and generated values. An example of an input value is the detected value output from the sensor 61. Another example of an input value is the control value output from at least one of the higher-level controller 62 and the servo amplifier 71. An example of a generated value is an extracted value, a statistic, or a principal component. The generated value is generated by the generation unit 3 based on the input value. In this embodiment, the acquisition unit 2 acquires the input value as a diagnostic parameter. In other words, the acquisition unit 2 acquires at least one signal from at least one of the higher-level controller 62 and the servo amplifier 71 (input value), and the detected signal (input value) from the sensor 61 (acquisition step). The acquisition unit 2 outputs the input value to the generation unit 3 and the processing unit 4.

[0046] The generation unit 3 generates one or more diagnostic parameters from a single input value. Furthermore, the input value itself is also a diagnostic parameter and can be used for degradation diagnosis. In this embodiment, the diagnostic parameters used for degradation diagnosis are the generated values ​​produced by the generation unit 3.

[0047] (5.3) Generation part The generation unit 3 generates generated values ​​and outputs them to the processing unit 4. The generation unit 3 includes an extraction unit 31, a statistical unit 32, a statistical analysis unit 33, and a trend generation unit 34. In this embodiment, the extraction unit 31, the statistical unit 32, and the statistical analysis unit 33 correspond to the first generation unit 3A, and the trend generation unit 34 corresponds to the second generation unit 3B (see Figure 1).

[0048] The extraction unit 31 performs an extraction process on the input values ​​to generate extracted values. The extraction process is the process of extracting values ​​from the input values ​​for a specific period as extracted values.

[0049] The statistics unit 32 generates statistics based on the input values. Examples of statistics include the mean, standard deviation, maximum value, minimum value, or features obtained from a histogram.

[0050] The statistical analysis unit 33 performs statistical analysis on the input values ​​and generates diagnostic parameters separate from the input values. More specifically, the statistical analysis unit 33 performs principal component analysis on multiple input values ​​and generates principal components.

[0051] Furthermore, the generation unit 3 can generate diagnostic parameters by combining two or more processes from among the extraction process in the extraction unit 31, the processing in the statistical quantity unit 32, and the processing in the statistical analysis unit 33. In this embodiment, the generation unit 3 can generate diagnostic parameters by combining at least one of the processes from among the extraction process in the extraction unit 31, the processing in the statistical quantity unit 32, and the processing in the statistical analysis unit 33. In other words, the generation unit 3 generates diagnostic parameters (generated values) by performing at least one of the processes of converting input values ​​into statistics, statistical analysis, and extraction.

[0052] As an example, the generation unit 3 first generates extracted values ​​through the extraction process in the extraction unit 31. Furthermore, the generation unit 3 generates diagnostic parameters by performing processes such as taking the mean of the extracted values ​​or calculating the standard deviation of the extracted values ​​in the statistical unit 32.

[0053] As another example, the generation unit 3 first generates multiple extracted values ​​through the extraction process in the extraction unit 31. Furthermore, the generation unit 3 generates multiple principal components by performing principal component analysis on the multiple extracted values ​​in the statistical analysis unit 33. The generation unit 3 outputs the multiple principal components as diagnostic parameters.

[0054] In this embodiment, we will explain using as an example the case in which the diagnostic parameters directly used for deterioration diagnosis among a plurality of diagnostic parameters are generated values ​​produced by performing at least the following process. That is, the diagnostic parameters directly used for deterioration diagnosis are generated values ​​produced by taking the average value of the extracted values ​​produced by the extraction unit 31 in the statistics unit 32. On the other hand, the input values ​​as diagnostic parameters are used indirectly in deterioration diagnosis by being converted into generated values ​​and then used in deterioration diagnosis.

[0055] In this disclosure, diagnostic parameters (generated values) may also be referred to as first data D1 (see Figures 4A to 4C). The first generation unit 3A (extraction unit 31, statistical unit 32, and statistical analysis unit 33) generates first data D1 relating to three feature quantities (e.g., mean values) from predetermined regions (all or part of the feature period described later) in the signal waveforms of, for example, three signals (first generation step).

[0056] The trend generation unit 34 (second generation unit 3B) generates second data D2, which is time-series data showing the progression of deterioration of the target OB1 (second generation step). It is preferable for the trend generation unit 34 to generate three or more sets of second data D2 (in this case, three as shown in Figures 4A to 4C). Each second data D2 shows the trend of change over time of the first data D1 (diagnostic parameter) relating to the corresponding feature quantity. Figures 4A to 4C show the second data D2 relating to "feature quantity A," "feature quantity B," and "feature quantity C" that change over time, with the horizontal axis representing time. Feature quantities A to C are all average values, for example, but the underlying signals are of different types. Specifically, "feature quantity A" is, for example, the average value (average current value) of a predetermined region (all or part of the feature period) in the signal waveform of the current supplied to the servo motor 72. "Feature B" is, for example, the average value (average torque) of a predetermined region (all or part of the feature period) in the torque signal waveform of the servo motor 72. "Feature C" is, for example, the average value of a predetermined region (all or part of the feature period) in the speed signal waveform of the servo motor 72.

[0057] The generation unit 3 stores the first data D1 related to features A to C, generated by the first generation unit 3A, in the storage unit 53 as historical information. The trend generation unit 34 generates second data D2 corresponding to each feature based on the first data D1 related to features A to C generated at a certain point in time (for example, the present time) and the past first data D1 related to features A to C stored in the storage unit 53. In Figures 4A to 4C, each second data D2 is illustrated as a broken linear data formed by connecting multiple first data D1 (plots) arranged in time series for each generation interval T1.

[0058] The generation interval T1 between two adjacent first data points D1 (plots), i.e., the interval between time points t1-t2, t2-t3, t3-t4, etc., is constant in the illustrated example. The generation interval T1 is not particularly limited and could be, for example, several hours, one day, or one month. The generation interval T1 can be changed by the user, for example, via the operating device 82.

[0059] In Figures 4A to 4C, time t7 is used as the current time as an example. For each of the feature quantities A to C, the trend generation unit 34 generates the second data D2 using the latest first data D1 generated at time t7 (the current time) and the first data D1 stored in the memory unit 53 that was generated at time t1 to t6, which are earlier than time t7.

[0060] The trend generation unit 34 outputs the three generated second data D2 to the processing unit 4. The second data D2 is processed by the presentation unit 47 of the processing unit 4, which will be described later.

[0061] Features A to C may generally show a monotonically decreasing trend as the aging deterioration of the target OB1 progresses. However, at least one of features A to C may also show a monotonically increasing trend as the aging deterioration of the target OB1 progresses.

[0062] (5.4) Output section The output unit 51 functions as a display output unit that outputs input values ​​to the display device 81. The output unit 51 outputs input values, for example, via a communication interface device. The output unit 51 also outputs generated values ​​to the display device 81. In other words, the output unit 51 outputs diagnostic parameters (input values ​​or generated values) to the display device 81 (display device).

[0063] Furthermore, the output unit 51 functions as a reliability output unit that outputs reliability information linked to diagnostic parameters to the display device 81 (presentation device).

[0064] (5.5) Receiver The receiving unit 52 receives signals from the operating device 82. The receiving unit 52 receives signals, for example, via a communication interface device.

[0065] (5.6) Processing Unit The processing unit 4 has the functions of a classification unit 41, a period determination unit 42, a correction unit 43, a reliability determination unit 44, a learning unit 45, a diagnosis unit 46, a presentation unit 47, and an estimation unit 48.

[0066] (5.6.1) Division The division unit 41 divides the input value into multiple periods according to the waveform characteristics of the input value in order to display it on the display device 81. The information regarding the multiple periods determined by the division unit 41 is output to the display device 81 via the output unit 51. The display device 81 displays the input value divided into the multiple periods determined by the division unit 41.

[0067] Figure 2 shows an example of a configuration in which the control value of the speed of the servo motor 72 is used as the input value, and the input value is divided into multiple periods p1 to p15 and displayed on the display device 81. The servo motor 72 is controlled to repeat the same operation with periods p1 to p15 as one cycle. In other words, the servo motor 72 performs an operation with periods p1 to p15 as one cycle for one workpiece. Each of the periods p1 to p15 has the characteristics of monotonically decreasing, monotonically increasing, or having a constant value. The division unit 41 divides the input value into multiple periods p1 to p15 by analyzing the rate of change of the input value.

[0068] The user specifies a desired period from among several periods p1 to p15 to be used for deterioration diagnosis by operating the operating device 82 (see Figure 1). The receiving unit 52 receives a specification signal output from the operating device 82 in response to the user's operation of the operating device 82. The specification signal includes information that identifies the period specified by the user. When the receiving unit 52 receives the specification signal, the generating unit 3 extracts the value of the period specified by the user's operation of the operating device 82 from the input values ​​during the extraction process. More specifically, when the receiving unit 52 receives the specification signal, the generating unit 3 extracts the value of the period specified by the user's operation of the operating device 82 from among the input values ​​during the extraction process.

[0069] For example, the user specifies a period p7. The extraction unit 31 of the generation unit 3 generates the value for period p7 from the input values ​​as the extracted value.

[0070] Here, the extraction unit 31 may generate an extracted value from among the input values ​​displayed on the display device 81 for the period p7 specified by the user. Alternatively, the extraction unit 31 may generate an extracted value from among input values ​​different from the input values ​​displayed on the display device 81 for the period p7 specified by the user. In the example in Figure 2, the input value displayed on the display device 81 is speed. Figure 3 is an example of the current value for the period p7 specified by the user. That is, the period from the starting point ti to the ending point tf in Figure 3 coincides with the period p7. The extraction unit 31 may use such a current value as the extracted value.

[0071] In other words, the acquisition unit 2 acquires multiple input values. These multiple input values ​​include a first input value and a second input value. In the example above, the first input value is speed, and the second input value is the current value. In the extraction process, the generation unit 3 performs two steps: a first step to identify a characteristic period in the first input value in which a predetermined characteristic is present, and a second step to extract the value included in the characteristic period from the second input value as a diagnostic parameter (extracted value). Specifically, in the first step, the generation unit 3 uses the period p7 specified by the user's operation on the operating device 82 as the characteristic period. The predetermined characteristic in period p7 is that the speed (first input value) is decreasing monotonically. In addition, the input values ​​are divided into multiple periods p1 to p15 so that periods other than period p7 also have their own characteristics. Therefore, the extraction process is the process of extracting values ​​from the input values ​​for periods determined based on predetermined characteristics. In the second process, the generation unit 3 extracts values ​​from the current values ​​(second input values) that are included in the period p7 (characteristic period) as diagnostic parameters.

[0072] The generation unit 3 may use the entire set of values ​​included in the characteristic period as diagnostic parameters.

[0073] Alternatively, the generation unit 3 may use values ​​from a portion of the feature period as diagnostic parameters. In short, in the process of extracting values ​​from the second input value that are included in the feature period as diagnostic parameters, the generation unit 3 may extract values ​​from a portion of the feature period as diagnostic parameters. For example, the generation unit 3 may extract values ​​from a portion of the feature period p101 (see Figure 3) as diagnostic parameters. The generation unit 3 may set the period p101 to a period specified, for example, by user operation on the operating device 82. Alternatively, the period p101 may be determined, for example, by the period determination unit 42 based on the waveform characteristics of the second input value. The waveform characteristics of the second input value include, for example, the amplitude, frequency, or average value per unit time of the second input value. Alternatively, the period p101 may be determined, for example, by the period determination unit 42 to a period corresponding to one cycle of the second input value.

[0074] In other words, the aforementioned "feature A (see Figure 4A)" may be the average value of the current during the feature period (period p7), or it may be a feature (in this case, the average value) of the current during a portion of the feature period (period p101).

[0075] (5.6.2) Period Determination Section The period determination unit 42 has a function to determine the period to be used for deterioration diagnosis from among the input values. This function can be switched on and off, for example, by user operation on the operating device 82. When this function is disabled, the period to be used for deterioration diagnosis is determined by user operation on the operating device 82. The extraction unit 31 extracts the values ​​of the period determined by the period determination unit 42 or by user operation from among the input values ​​as extracted values.

[0076] The period determination unit 42, for example, selects a characteristic period from the input values ​​in which a predetermined characteristic is exhibited as the period to be used for degradation diagnosis. Information on which period to use for degradation diagnosis is stored in the storage unit 53. The period determination unit 42 determines the period to use for degradation diagnosis by referring to this information. This information can also be updated by user operation on the operating device 82.

[0077] Allowing the user to determine the timeframe for degradation diagnosis has the advantage of incorporating the user's expertise into the diagnosis. The appropriate timeframe for degradation diagnosis is expected to vary depending on the installation environment and usage conditions of the servo system 7. By allowing the user to determine the timeframe, the accuracy of the degradation diagnosis can be improved.

[0078] Furthermore, if the period determination unit 42 determines the period to be used for deterioration diagnosis, there is the advantage that the user does not have to go through the trouble of determining that period.

[0079] (5.6.3) Presentation section The presentation unit 47 has a function (presentation step) to compare three or more second data points D2 (in this case, the three shown in Figures 4A to 4C) generated by the trend generation unit 34 (second generation unit 3B). For each second data point D2, the presentation unit 47 identifies the trend of change by analyzing the direction of change (increase, decrease, or constant) and the rate of change between adjacent first data points D1 (plots). Here, "trend of change" is assumed to be a trend that correlates with the progression of deterioration. In other words, it means a monotonic change trend in one direction from time point t1 to time point t7 in Figures 4A to 4C. "One direction" is assumed to be, for example, a monotonic decrease, but depending on the type of diagnostic parameter, it may also be a monotonic increase. Feature quantities A and B in Figures 4A and 4B both decrease monotonically from time point t1 to time point t7, and can be said to show the same trend of change. Furthermore, even if one of feature quantities A and B monotonically decreases from time t1 to time t7, and the other monotonically increases from time t1 to time t7, the same trend of change (a trend correlated with the progression of degradation) is considered to be observed. However, feature quantity C in Figure 4C does not change in one direction from time t1 to time t7. In other words, although the second data D2 in Figure 4C shows a decreasing trend, unlike the second data D2 in Figures 4A and 4B, there is an interval (time t2 to t3) where it temporarily shows an increasing trend due to the influence of noise, etc. In other words, the second data D2 in Figure 4C contains an interval that is not correlated with the progression of degradation. If the second data D2, which includes such an interval that is not correlated with the progression of degradation due to the influence of noise, etc., is presented to the user, it may lead to misleading judgments about degradation.

[0080] Therefore, the presentation unit 47 determines whether there is a specific second data D20 (see Figure 4C) that exhibits a different trend of change from the other second data D2s. For example, if the majority of the second data D2s out of three or more second data D2s show a monotonically decreasing trend of change, and there are less than half (e.g., one) of the second data D2s that exhibit a different trend of change, the presentation unit 47 determines that second data D2 to be the specific second data D20.

[0081] If the presentation unit 47 determines that all three second data D2s have the same change trend and that no specific second data D20 exists, it presents the three second data D2s without performing any "specific processing". The presentation unit 47 outputs the three second data D2s to the display device 81 via the output unit 51, and the display device 81 presents (displays) the three second data D2s.

[0082] On the other hand, if there is a specific second data D20 that exhibits a different trend of change from other second data D2s, the presentation unit 47 performs specific processing on the target data D10, which is the first data D1 that caused the different trend of change, and presents the specific second data D20 (presentation step). The specific second data D20 illustrated in Figure 4C has an interval (times t2 to t3) that shows a temporary increasing trend. The presentation unit 47 determines that the first data D1 at time t3 is the target data D10 that caused the different trend of change. The presentation unit 47 may also determine that the first data D1 before and after the first data D1 at time t3 is also the target data D10.

[0083] The specific processing includes processing related to at least one of the following: not presenting the target data D10, correcting the target data D10, and providing notification regarding the target data D10.

[0084] The display unit 47 outputs the specific second data D20, which has undergone specific processing, to the display device 81 via the output unit 51, and presents the specific second data D20 on the screen of the display device 81. The display device 81 displays the specific second data D20 as a graph of piecewise linear data, for example, as shown in Figure 4C. The user can visually confirm the progress of deterioration through the specific second data D20 displayed on the screen.

[0085] Furthermore, the presentation unit 47 presents a specific second data D20 in a manner that includes the first data D1 generated by the first generation unit 3A at a certain point in time (in this case, time t7), and the first data D1 generated by the first generation unit 3A before a certain point in time (time t7). Specifically, as shown in Figure 4C, the presentation unit 47 presents a specific second data D20 in a manner that includes seven first data D1 from time t1 to t7. Therefore, the user can check the progress of deterioration up to time t7 through the presented specific second data D20.

[0086] The presentation unit 47 preferably presents not only the specific second data D20 but also other second data D2s. That is, the presentation unit 47 preferably further presents at least one of the three or more second data D2s other than the specific second data D20 (in this case, two second data D2s, feature A and feature B). The display device 81 displays the two second data D2s, feature A and feature B, as a graph of piecewise linear data, for example, as shown in Figures 4A and 4B. The "presentation" of data as used here is not limited to screen display and may include audio output, etc. For example, notification regarding the target data D10 may be made by audio output. The second data D2s other than the specific second data D20 are preferably displayed together with the specific second data D20 on the same screen, but are not particularly limited and may be displayed on separate screens.

[0087] The non-display processing, correction processing, and notification processing will be explained below. The display unit 47 may perform specific processing only if the temporary increase in the monotonic decrease of the feature quantity is greater than or equal to a threshold. Alternatively, the display unit 47 may perform specific processing only if the temporary decrease in the monotonic increase of the feature quantity is greater than or equal to a threshold.

[0088] [Not displayed] For example, if a specific process includes a non-display process related to not displaying the target data D10, the display unit 47 will suppress the screen display of the target data D10 by the display device 81. In the example in Figure 4C, the display unit 47 will suppress the screen display of the first data D1 (target data D10) at time t3 by the display device 81 for a specific second data D20. Specifically, for the specific second data D20, the display unit 47 will hide (blank) the piecewise linear data from time t2 to time t4, including the first data D1 (plot) at time t3, and will display the piecewise linear data from time t1 to t2 and from time t4 onwards. In this way, the screen display of the target data D10, which is highly likely to have been generated due to the influence of noise, etc., is suppressed, making it less likely to lead to misinterpretations in degradation diagnosis.

[0089] [Correction process] For example, if a specific process includes a correction process for correcting the target data D10, the presentation unit 47 corrects the target data D10 in the correction unit 43 based on at least one change trend of the second data D2 other than the specific second data D20 among three or more second data D2s. The presentation unit 47 then presents the corrected target data D10. Here, the presentation unit 47 presents (displays) the specific second data D20, which includes the corrected target data D10, from the display device 81.

[0090] The corrected target data D10 is data to be presented by the presentation unit 47. For the deterioration diagnosis by the diagnostic unit 46, either the uncorrected target data D10 or the corrected target data D10 may be used.

[0091] The correction unit 43 performs correction processing on the target data D10 in response to a command from the presentation unit 47. In the example shown in Figure 4C, the correction unit 43 corrects the first data D1 (target data D10) at time t3 for a specific second data D20.

[0092] Here, the presentation unit 47, in the correction unit 43, corrects the target data D10 based on the past change trend of the target data D10 in a specific second data D20, and presents the corrected target data D10. In the example in Figure 4C, the correction unit 43 corrects the target data D10 at time t3 based on two first data D1 at time t1 to t2, which are prior to time t3, and their monotonically decreasing rate of change (slope). Therefore, the reliability of the correction of the target data D10 is improved. In this embodiment, the correction unit 43 estimates the true first data D1X at time t3 (see Figure 4C) based on the two first data D1 at time t1 to t2 and their rate of change, as well as multiple first data D1 from time t4 onward and their rates of change. The correction unit 43 interpolates the true piecewise linear data K1 (see dashed line in Figure 4C) between time t2 to t4, which includes the estimated first data D1X.

[0093] The correction unit 43 may interpolate the piecewise linear data K1 based on correlation data. The correlation data is information relating to the correlation between the type of first data D1 (diagnostic parameter), the value of first data D1, and the progression of deterioration of the servo system 7, and is stored in the storage unit 53. The correlation data is information prepared based on verification results such as simulation results, and may be stored as a data table or a mathematical formula.

[0094] The presentation unit 47 preferably corrects the target data D10 using an approximation curve. In other words, it is preferable that the true first data D1X at time t3 is determined by the approximation curve. In this case, the reliability of the correction of the target data D10 is improved.

[0095] In this way, the target data D10, which is highly likely to have been generated by noise or other influences, is corrected and presented, making it less likely to mislead in degradation diagnosis.

[0096] When the display unit 47 displays a specific second data D20 from the display device 81, it is preferable to display the corrected target data D10 on the screen in a display format that distinguishes it from other first data D1s. A distinguishable display format may be, for example, color or line type. For example, as shown in Figure 4C, the display unit 47 may display the corrected target data D10 (first data D1X) as a white plot and the other first data D1s as black plots. Alternatively, as shown in Figure 4C, the display unit 47 may display the true polylinear data K1 between time points t2 to t4, which includes the first data D1X, as a red dashed line and the others as black solid lines. By displaying the corrected target data D10 in a display format that distinguishes it from other first data D1s, the user can easily visually confirm that the target data D10 is corrected data.

[0097] [Notification Processing] For example, if a specific process includes notification processing regarding target data D10, the display unit 47 displays warning information on the screen indicating that the specific second data D20 exhibits a different change trend from other second data D2 when displaying the specific second data D20 on the screen from the display device 81. The warning information includes data such as the string "There is data that exhibits a different change trend from others." Preferably, the warning information is displayed on the same screen as the specific second data D20, but it is not limited to that and may be displayed on a separate screen. Furthermore, notification of the warning information is not limited to screen display and may also be done by audio output from a speaker. Because the warning information is notified, the user can easily learn through the warning information that there is target data D10 that is highly likely to have been generated by the influence of noise or the like. Therefore, it is less likely to lead to misinterpretation in degradation diagnosis.

[0098] If a specific process includes both notification and non-display processing, the warning information may include string data such as, for example, "Data with a different change trend exists, so it has been hidden." Also, if a specific process includes both notification and correction processing, the warning information may include string data such as, for example, "Data has been corrected because data with a different change trend exists."

[0099] In this embodiment, the user may appropriately change the settings via the operating device 82 to determine which of the following processes is included in the specific processing: non-display processing, correction processing, and notification processing.

[0100] Thus, according to this embodiment, if there is a specific second data D20 exhibiting a different change trend, the specific second data D20 is presented after applying specific processing to the target data D10 that caused the different change trend. Therefore, users are less likely to mistake the target data D10, which is highly likely to have been generated by the influence of noise, for the true data. As a result, the data processing system 1 has the advantage of making it less likely to mislead regarding degradation diagnosis.

[0101] (5.6.4) Confidence Determination Unit and Estimation Unit The reliability determination unit 44 adds information regarding the reliability of the diagnostic parameter in the degradation diagnosis of the servo system 7 to the diagnostic parameter (first data D1). The diagnostic parameter here may be an input value acquired by the acquisition unit 2, or a generated value generated by the generation unit 3 based on the input value. The reliability should be set higher as the correlation between the diagnostic parameter and the degradation degree of the servo system 7 becomes stronger.

[0102] The reliability determination unit 44 has a function to determine the reliability according to the user's operation on the operating device 82, and a function to determine the reliability automatically. First, the former will be explained.

[0103] As shown in Figure 2, the display device 81 displays diagnostic parameters. The user specifies the confidence level of the diagnostic parameters by operating the operating device 82. The receiving unit 52 receives a specified signal output from the operating device 82 in response to the user's operation of the operating device 82. The specified signal includes information on the confidence level specified by the user. When the receiving unit 52 receives the specified signal, the confidence level determination unit 44 determines the confidence level to the value specified by the user's operation of the operating device 82.

[0104] Furthermore, if the display device 81 displays multiple diagnostic parameters (second data D2 including multiple first data D1), the reliability of each of the diagnostic parameters can be individually specified by the user through operation on the operating device 82.

[0105] Next, the function of the reliability determination unit 44 to automatically determine reliability will be explained. Here, the reliability determination unit 44 has the function of automatically determining not only the reliability of a single diagnostic parameter (first data D1), but also the reliability of second data D2 which includes multiple diagnostic parameters (hereinafter sometimes referred to as "overall reliability"). In particular, the reliability determination unit 44 has the function of automatically determining the overall reliability of a specific second data D20 which exhibits a different change trend from other second data D2s. The reliability determination unit 44 determines the reliability (overall reliability) of the change trend of the specific second data D20 based on the change trend of reference data which has the same type of feature as the specific second data D20. The "reference data" here refers to diagnostic parameters generated in the past, or diagnostic parameters related to other equipment. Specifically, the "reference data" is one or more first data D1 or second data D2 generated in the past for a servo motor other than the servo motor 72 of interest. However, if the feature of the specific second data D20 is the average value of the current, then the feature of the reference data is also the average value of the current. The reference data is stored in the storage unit 53. The reference data may be stored as a data table or as a mathematical formula.

[0106] In this embodiment, the confidence determination unit 44 determines the confidence level of the diagnostic parameters and the overall confidence level using a trained model generated by machine learning. The trained model may be generated by the learning unit 45, or it may be generated by an external device and provided to the data processing system 1.

[0107] A pre-trained model includes, for example, a classifier using a pre-trained neural network. A pre-trained neural network may include, for example, a CNN (Convolutional Neural Network) or a BNN (Bayesian Neural Network). A pre-trained model is realized by implementing the pre-trained neural network on an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field-Programmable Gate Array).

[0108] When the user determines the reliability of the diagnostic parameters, there is the advantage that the user's knowledge can be reflected in the reliability. It is expected that the appropriate value for the reliability of the diagnostic parameters will vary depending on the installation environment and usage conditions of the servo system 7. By allowing the user to determine the reliability of the diagnostic parameters, it is possible to set an appropriate reliability value.

[0109] Furthermore, when the reliability is determined by the period determination unit 42, there is the advantage that the user does not have to go through the trouble of determining the reliability of the diagnostic parameters.

[0110] Once the reliability of the diagnostic parameters is determined, the reliability determination unit 44 outputs the diagnostic parameters and information regarding the reliability of the diagnostic parameters to the presentation device via the output unit 51. The presentation device presents the diagnostic parameters and their reliability. In this embodiment, the display device 81 functions as the presentation device. The display device 81 displays the diagnostic parameters and their reliability.

[0111] The reliability determination unit 44 also determines the overall reliability of a specific second data D20. The display unit 47 displays information regarding the reliability (overall reliability) on the screen when the display device 81 displays a specific second data D20. The reliability determination unit 44 automatically determines the overall reliability of not only a specific second data D20 but also other second data D2s. The display unit 47 displays information regarding the overall reliability on the screen when the display device 81 displays the relevant second data D2. Through the overall reliability information displayed on the screen, the user can easily find out how reliable the specific second data D20 is. This makes it less likely to be misled when diagnosing degradation.

[0112] Figure 5 schematically illustrates the display of the display device 81, including the display of the overall confidence level. As an example, Figure 5 shows the second data D2 of feature A, explained in Figure 4A, and its overall confidence level of "0.8". In Figure 5, time t12 is considered the present time. That is, the first data D1 at time t12 shows the latest actual data, and the first data D1 at time points prior to time t12 shows past actual data. The display device 81 displays the overall confidence level in the range of "0" to "1.0", but is not particularly limited and may also display it as a percentage in the range of "0%" to "100%".

[0113] Here, the first data D1 at a time point after time point t12 is not actual data, but rather predicted data relating to the first data D1. The presentation unit 47 of this embodiment presents a specific second data D20 in a manner that includes the first data D1 generated by the first generation unit 3A at a certain time point (time point t12) and predicted data relating to the first data D1 in the future after a certain time point (time point t12).

[0114] Specifically, the estimation unit 48 estimates future first data D1 in response to a command from the presentation unit 47. The estimation unit 48 estimates first data D1 at a time after time t12, for example, based on first data D1 and change trends generated up to a certain time (time t12), as well as the aforementioned reference data and correlation data. It is preferable that the presentation unit 47 also presents other second data D2s besides the specific second data D20 in a manner that includes predicted data regarding future first data D1 after a certain time (time t12).

[0115] Since a specific second data D20 is presented in a manner that includes the future first data D1, the user can check the predicted future rate of deterioration from a certain point in time through the presented specific second data D20.

[0116] (5.6.5) Learning Department The learning unit 45 generates trained models (a first trained model and a second trained model). The first trained model takes diagnostic parameters (first data D1) as input and outputs the confidence level of the diagnostic parameters. The second trained model takes second data D2, which contains multiple diagnostic parameters, as input and outputs the overall confidence level of second data D2.

[0117] The operating device 82 transmits a designated signal containing information on the confidence level (label) of a specific diagnostic parameter and the overall confidence level (label) of the second data D2, in response to user input. The learning unit 45 uses the designated signal as training data to perform machine learning and generate a trained model. Similarly, the learning unit 45 can also retrain the trained model.

[0118] (5.6.6) Diagnostic Department The diagnostic unit 46 performs a deterioration diagnosis of the servo system 7 based on diagnostic parameters. The diagnostic parameters can be either generated values ​​from the generation unit 3 or input values ​​acquired by the acquisition unit 2. In this embodiment, the case where the diagnostic unit 46 uses generated values ​​from the generation unit 3 as diagnostic parameters will be explained as an example.

[0119] The diagnostic parameters used for deterioration diagnosis can be specified (selected) by the user through operation of the control device 82, or they can be automatically determined by the diagnostic unit 46. First, we will explain the case where the user specifies the diagnostic parameters.

[0120] The display device 81 shows multiple diagnostic parameters (see Figures 4A to 4C). The user operates the control device 82 to specify which of the multiple diagnostic parameters to use for degradation diagnosis. The control device 82 then outputs a specification signal containing information about the user's specification.

[0121] The receiving unit 52 receives a specified signal output from the operating device 82 in response to user operations on the operating device 82. When the receiving unit 52 receives the specified signal, the diagnostic unit 46 performs a deterioration diagnosis of the servo system 7 based on the diagnostic parameters specified by the user's operation on the operating device 82.

[0122] The display device 81 shows the confidence level of each of the multiple diagnostic parameters. The user can refer to the confidence levels and select the diagnostic parameters to be used for deterioration diagnosis from among the multiple diagnostic parameters.

[0123] The diagnostic parameters for period p1, period p2, ..., and period p16 in Figure 2 each correspond to the extracted values ​​extracted by the extraction unit 31, and can therefore be candidates for diagnostic parameters used in deterioration diagnosis. Thus, the operation of specifying diagnostic parameters to be used in deterioration diagnosis includes the operation of specifying one or more periods from the multiple periods p1 to p16. The output unit 51 outputs the recommendation information to the display device 81, and the display device 81 displays the recommendation information. In this way, the output unit 51 functions as a display output unit that outputs the recommendation information to the display device 81. The recommendation information is information regarding whether or not a particular period from the multiple periods p1 to p16 is recommended. The recommendation information in this embodiment is a confidence level. That is, each of the multiple periods p1 to p16 is assigned a confidence level. For a given period from the multiple periods p1 to p16, a higher confidence level means that the period is recommended, and a lower confidence level means that the period is not recommended.

[0124] If the diagnostic unit 46 automatically determines the diagnostic parameters to be used for deterioration diagnosis, the diagnostic unit 46 will, for example, use the diagnostic parameter with the highest reliability among multiple diagnostic parameters as the diagnostic parameter to be used for deterioration diagnosis.

[0125] Information regarding the correlation between the type of diagnostic parameter, the value of the diagnostic parameter, and the degree of deterioration of the servo system 7 is stored in the memory unit 53. For example, the above information is stored as a data table or as a mathematical formula. Using the above information, the diagnostic unit 46 performs a deterioration diagnosis of the servo system 7.

[0126] The diagnostic unit 46 is configured to determine the degree of deterioration of the target OB1 with respect to the corresponding feature for each of the three or more second data D2.

[0127] For example, the diagnostic unit 46 uses the average current value, which is the time-averaged value of the amplitude of the current supplied to the servo motor 72, as the diagnostic parameter (first data D1) for each second data D2. The diagnostic unit 46 diagnoses that the smaller the absolute value of the average current value, the greater the degree of deterioration.

[0128] Furthermore, for example, the diagnostic unit 46 uses the average torque, which is the time-averaged value of the torque of the servo motor 72, as a diagnostic parameter for each second data D2. The diagnostic unit 46 diagnoses that the smaller the average torque, the greater the degree of deterioration.

[0129] Furthermore, for example, the diagnostic unit 46 uses principal components, which are values ​​generated by principal component analysis of the detected values ​​of the position sensor, etc., as diagnostic parameters for each second data D2. The diagnostic unit 46 diagnoses that the degree of deterioration is greater the smaller the principal component is, or the greater the principal component is. Of the first principal component and second principal component etc. generated by principal component analysis, the principal component that has a relatively strong correlation with the degree of deterioration of the servo system 7 should be selected by the user or the diagnostic unit 46 and used for deterioration diagnosis.

[0130] To further improve the accuracy of degradation diagnosis, the diagnostic unit 46 may perform degradation diagnosis using a trained model generated by machine learning. The trained model is generated by performing machine learning using pairs of diagnostic parameters and degradation levels (labels) as training data.

[0131] The diagnostic unit 46 outputs the degree of deterioration as a diagnostic result.

[0132] The output unit 51 functions as a presentation output unit that outputs the diagnostic results of the diagnostic unit 46 to the presentation device. The presentation device is a device that presents information, for example, by images, sound, or a combination thereof. In this embodiment, the display device 81 functions as the presentation device.

[0133] Furthermore, the display device 81 may display a judgment corresponding to the range of degradation. In other words, the display unit 47 may display three or more second data D2 on the display device 81 screen in a visually recognizable manner regarding the degradation level determined by the diagnostic unit 46. For example, as shown in Figure 5, a first threshold Th1 and a second threshold Th2 (first threshold Th1 > second threshold Th2) are set, and the diagnostic unit 46 determines the degradation level in three stages: "good," "caution," and "warning," based on the first threshold Th1 and the second threshold Th2. The degradation level is not limited to being determined in three stages; it may be two or four or more stages. When the feature quantity is greater than or equal to the first threshold Th1, the determination that the degradation level is "good" is displayed. In Figure 5, up to time t11, it is determined to be "good," and the second data D2 is shown, for example, by a solid green line. When the feature quantity is less than the first threshold Th1 and greater than or equal to the second threshold Th2, the determination that the degradation level is "caution" is displayed. In Figure 5, the data is judged as "caution" from time t11 to time t13, and the second data D2 is shown, for example, by a yellow dashed line. When the feature quantity is less than the second threshold Th2, the degradation level is judged as "warning". In Figure 5, the data is judged as "warning" from time t13 onwards, and the second data D2 is shown, for example, by a red dashed line. Note that in Figure 5, from time t12 onwards, the future first data D1 estimated by the estimation unit 48 is shown. The diagnostic unit 46 displays the three stages of degradation, including the estimated future first data D1.

[0134] Since three or more second data points D2 are displayed in a manner that allows the degree of deterioration to be visually observed, the user can easily determine the degree of deterioration of the target OB1 through the three or more second data points D2 displayed on the screen.

[0135] (5.7) Storage section The memory unit 53 is, for example, a ROM (Read Only Memory), RAM (Random Access Memory), or EEPROM (Electrically Erasable Programmable Read Only Memory). The memory unit 53 stores information used in the data processing system 1. For example, the memory unit 53 stores information about predetermined features of the input values, which the division unit 41 uses to divide the input values ​​into multiple periods according to their features. The memory unit 53 also stores a trained model used in the confidence determination unit 44. Furthermore, as described above, the memory unit 53 stores reference data and correlation data, etc.

[0136] (6)Display device The display device 81 includes a display. The display device 81 displays information corresponding to the information acquired from the output unit 51. The display device 81 displays the waveform of the diagnostic parameters, the reliability of the diagnostic parameters, and the degree of deterioration as a diagnostic result of the diagnostic unit 46. The display device 81 also displays the setting information of the data processing system 1.

[0137] As described above, the display device 81 functions as a display device that presents information. In other words, the display device 81 is a display device that displays the waveform and reliability of diagnostic parameters.

[0138] (7) Operating device The operating device 82 includes, for example, one or more of a keyboard, touchpad, and buttons. The operating device 82 is used together with the display device 81. The user operates the operating device 82 and inputs information while referring to the information displayed on the display device 81.

[0139] The operating device 82 may be formed integrally with the display device 81. For example, the touchpad of the operating device 82 and the display of the display device 81 may constitute a touch panel.

[0140] (8) Deterioration diagnosis The sequence of steps for diagnosing the degradation of the servo system 7 using the data processing system 1 will be explained with reference to Figure 6. Note that the flowchart shown in Figure 6 is merely one example of the degradation diagnosis flow according to this disclosure, and the order of processing may be changed as appropriate, or processing may be added or omitted as appropriate.

[0141] The acquisition unit 2 acquires multiple input values ​​from the sensor 61 and at least one of the higher-level controller 62 and the servo amplifier 71 (step ST1).

[0142] Next, the output unit 51 outputs multiple input values ​​and multiple confidence levels corresponding to the multiple input values ​​to the display device 81. The display device 81 displays the multiple input values ​​and multiple confidence levels (step ST2). The display device 81 also divides the input values ​​into multiple periods p1 to p15 and displays them.

[0143] The user operates the control device 82 to select one or more input values ​​from among multiple input values ​​to be used for deterioration diagnosis (step ST3).

[0144] Next, the user selects a period to be used for degradation diagnosis for the selected input values. If the user selects at least one period from among multiple periods p1 to p15 (Step ST4: Yes), the extraction unit 31 extracts the values ​​for the selected period from the input values ​​and generates extracted values ​​(Step ST5). The period selected by the user may be applied to all of the multiple input values, or a period may be selected individually for each input value.

[0145] The generation unit 3 performs statistical processing on the extracted values ​​(step ST6). Specifically, the statistical unit 32 of the generation unit 3 generates statistics based on the extracted values, and the statistical analysis unit 33 generates principal components based on the extracted values.

[0146] The diagnostic unit 46 performs a deterioration diagnosis of the servo system 7 based on at least one of the extracted values ​​extracted by the extraction unit 31, the statistics generated by the statistics unit 32, and the principal components generated by the statistical analysis unit 33 (step ST7).

[0147] The correction unit 43 compares multiple second data D2s and determines whether or not a specific second data D20 requiring correction is included (step ST8). If it is determined that the specific second data D20 is included, the correction unit 43 corrects the specific second data D20 (step ST9). If it is determined that the specific second data D20 is not included, step ST9 is omitted.

[0148] The display device 81 displays the diagnostic results of the diagnostic unit 46 (step ST10). As shown in Figure 5, the display device 81 displays each second data D2 along with the reliability (overall reliability) for the degree of deterioration determined by the diagnostic unit 46 in a visually identifiable manner. If a specific second data D20 is corrected in step ST9, the corrected specific second data D20 is displayed.

[0149] (modified version) The following are examples of modifications of the embodiment. These modifications may be implemented by combining them as appropriate.

[0150] The statistical processing in the statistical calculation unit 32 and the statistical analysis unit 33 may be performed before the extraction processing in the extraction unit 31.

[0151] In data processing system 1, statistical processing is not a mandatory process.

[0152] Furthermore, the statistical calculation unit 32 and the statistical analysis unit 33 may be located outside the data processing system 1. The acquisition unit 2 may acquire at least one of the statistical quantities generated by the statistical calculation unit 32 and the principal components generated by the statistical analysis unit 33.

[0153] The extraction unit 31 may be located outside the data processing system 1. The acquisition unit 2 may acquire extracted values ​​from the extraction unit 31.

[0154] In deterioration diagnosis, it is not essential to perform the extraction process by the extraction unit 31. The diagnostic unit 46 may perform deterioration diagnosis based, for example, on input values, statistics, or the values ​​of principal components over the entire period.

[0155] It is not essential that the data processing system 1 includes a diagnostic unit 46. Instead of the diagnostic unit 46, an external component of the data processing system 1 may perform the degradation diagnosis. Alternatively, a person may perform the degradation diagnosis by looking at the diagnostic parameters displayed on the display device 81.

[0156] An acceleration sensor or a temperature sensor may be used as sensor 61. Furthermore, when multiple servo systems 7 are used synchronously, a sensor that detects the state of other servo systems 7 may be used as sensor 61. For example, when multi-axis control is performed by multiple servo systems 7, the operating states of the multiple servo systems 7 may influence each other, so the detection results of the state of other servo systems 7 can be used to diagnose the deterioration of one servo system 7.

[0157] The diagnostic unit 46 may output a signal indicating whether or not deterioration is present, instead of outputting the degree of deterioration as a diagnostic result.

[0158] The display device for presenting the diagnostic results of the diagnostic unit 46 is not limited to the display device 81. The display device may be, for example, an audio output device that presents the diagnostic results by voice.

[0159] It is not essential that the display device is a display device 81 that displays the reliability of the diagnostic parameters along with the waveform of the diagnostic parameters. The display device may also present the reliability of the diagnostic parameters along with the labels (names, etc.) of the diagnostic parameters by voice or other means.

[0160] The division unit 41 may divide the input value into multiple periods according to the waveform characteristics of the input value by comparing the input value with model data representing predetermined characteristics of the input value.

[0161] The correction unit 43 is not limited to correcting only the first data D1 (target data D10) at time t3, but may also correct the first data D1 before and after time t3.

[0162] The warning information presented (displayed on the screen) by the display unit 47 is not limited to containing string data such as "There is data with a change trend that differs from others." The warning information may also include data other than string data, such as shapes (e.g., icons), symbols, and numbers.

[0163] The data processing system 1 in this disclosure includes a computer system. The computer system mainly consists of a processor and memory as hardware. At least part of the functions of the data processing system 1 in this disclosure are realized by the processor executing a program recorded in the memory of the computer system. The program may be pre-recorded in the memory of the computer system, provided via a telecommunications line, or provided on a non-temporary recording medium such as a memory card, optical disk, or hard disk drive that can be read by the computer system. The processor of the computer system consists of one or more electronic circuits including semiconductor integrated circuits (ICs) or large-scale integrated circuits (LSIs). The integrated circuits such as ICs and LSIs referred to here are named differently depending on the degree of integration, and include integrated circuits called system LSIs, VLSIs (Very Large Scale Integration), or ULSIs (Ultra Large Scale Integration). Furthermore, FPGAs (Field-Programmable Gate Arrays) that are programmed after the manufacture of the LSI, or logic devices that allow for the reconfiguration of junction relationships or circuit compartments within the LSI, can also be used as processors. Multiple electronic circuits may be integrated onto a single chip or distributed across multiple chips. Multiple chips may be integrated onto a single device or distributed across multiple devices. The computer system referred to herein includes a microcontroller having one or more processors and one or more memories. Therefore, the microcontroller also consists of one or more electronic circuits, including semiconductor integrated circuits or large-scale integrated circuits.

[0164] Furthermore, it is not essential for the data processing system 1 to have multiple functions integrated into a single device; the components of the data processing system 1 may be distributed across multiple devices. In addition, at least some of the functions of the data processing system 1, for example, some of the functions of the processing unit 4, may be implemented by the cloud (cloud computing), etc.

[0165] Conversely, in the embodiment, functions that are distributed across multiple devices may be consolidated into a single device. For example, functions that are distributed across the data processing system 1, the display device 81, and the operating device 82 may be consolidated into a single device.

[0166] (summary) Based on the embodiments described above, the following aspects are disclosed.

[0167] The data processing system (1) according to the first embodiment is used for diagnosing the deterioration of at least one of the load (73) and servo motor (72) in a servo system (7). The servo system (7) includes a load (73), a servo amplifier (71), and a servo motor (72) that provides power to the load (73) in accordance with the control of the servo amplifier (71). The data processing system (1) comprises an acquisition unit (2), a first generation unit (3A), a second generation unit (3B), and a presentation unit (47). The acquisition unit (2) acquires at least one signal from among the control signals used to control the servo system (7) and the detection signals output from a sensor (61) that detects the state of the servo system (7). The first generation unit (3A) generates first data (D1) relating to three or more feature quantities from a predetermined region in the signal waveform of at least one signal. The second generation unit (3B) generates three or more second data (D2) which are time-series data showing the progression of deterioration of the target (OB1), and each second data (D2) shows the change trend over time of three or more features of the first data (D1). The presentation unit (47) compares the three or more second data (D2) with each other. If there is a specific second data (D20) that exhibits a different change trend from the other second data (D2), the presentation unit (47) performs specific processing on the target data (D10), which is the first data (D1) that caused the different change trend, and presents the specific second data (D20). The specific processing includes processing related to at least one of the following: not presenting the target data (D10), correcting the target data (D10), and notifying the target data (D10).

[0168] According to the above configuration, if there is a specific second data (D20) that exhibits a different change trend, the specific second data (D20) is presented after applying specific processing to the target data (D10) that caused the different change trend. Therefore, users are less likely to mistake the target data (D10), which is highly likely to have been generated by the influence of noise, etc., for the true data. As a result, the data processing system (1) has the advantage of making it less likely to mislead regarding degradation diagnosis.

[0169] With respect to the data processing system (1) according to the second embodiment, in the first embodiment, the presentation unit (47) presents specific second data (D20) by screen display from the display device (81).

[0170] According to the above embodiment, the user can visually confirm the progress of deterioration through specific second data (D20) displayed on the screen.

[0171] With respect to the data processing system (1) according to the third embodiment, in the first or second embodiment, the presentation unit (47) presents specific second data (D20) in a manner that includes first data (D1) generated by the first generation unit (3A) at a certain point in time and first data (D1) generated by the first generation unit (3A) before a certain point in time.

[0172] According to the above embodiment, the user can check the progress of deterioration up to a certain point in time through the specific second data (D20) presented.

[0173] With respect to the data processing system (1) according to the fourth embodiment, in any one of the first to third embodiments, the presentation unit (47) presents specific second data (D20) in a manner that includes first data (D1) generated by the first generation unit (3A) at a certain point in time and predicted data relating to the first data (D1) in the future after a certain point in time.

[0174] According to the above embodiment, the user can check the predicted future rate of deterioration from a certain point in time through the specific second data (D20) presented.

[0175] With respect to the data processing system (1) according to the fifth embodiment, in any one of the first to fourth embodiments, the presentation unit (47) further presents at least one of the three or more second data (D2) other than a specific second data (D20).

[0176] According to the above embodiment, the user can visually confirm the progress of deterioration through specific second data (D20) and other second data (D2) displayed on the screen.

[0177] With respect to the data processing system (1) according to the sixth embodiment, in any one of the first to fifth embodiments, the specific processing includes processing related to not displaying the target data (D10). The display unit (47) stops the screen display of the target data (D10) by the display device (81).

[0178] According to the above configuration, the screen display of the target data (D10), which is highly likely to have been generated due to noise or other influences, is suppressed, making it less likely to mislead in the degradation diagnosis.

[0179] With respect to the data processing system (1) according to the seventh aspect, in any one of the first to sixth aspects, the specific processing includes processing related to the correction of the target data (D10). The presentation unit (47) corrects the target data (D10) based on the change trend of at least one of the three or more second data (D2) other than the specific second data (D20), and presents the corrected target data (D10).

[0180] According to the above configuration, the target data (D10), which is highly likely to have been generated due to the influence of noise, is corrected and presented, making it less likely to mislead in the degradation diagnosis.

[0181] With respect to the data processing system (1) relating to the eighth aspect, in any one of the first to seventh aspects, the specific processing includes processing relating to the correction of the target data (D10). The presentation unit (47) corrects the target data (D10) based on the past change trend of the specific second data (D20) compared to the target data (D10), and presents the corrected target data (D10).

[0182] According to the above embodiment, the reliability of the correction of the target data (D10) is improved.

[0183] With respect to the data processing system (1) according to the ninth embodiment, in the seventh or eighth embodiment, when the display unit (47) displays specific second data (D20) from the display device (81) on the screen, it displays the corrected target data (D10) on the screen in a display format that can be distinguished from other first data (D1).

[0184] According to the above embodiment, the user can easily visually confirm that the target data (D10) displayed on the screen is corrected data.

[0185] With respect to the data processing system (1) according to the tenth embodiment, in any one of the seventh to ninth embodiments, the presentation unit (47) corrects the target data (D10) using an approximation curve.

[0186] According to the above embodiment, the reliability of the correction of the target data (D10) is improved.

[0187] With respect to the data processing system (1) according to the 11th embodiment, in any one of the 1st to 10th embodiments, the specific processing includes the processing of a notification concerning the target data (D10). When the display unit (47) displays specific second data (D20) on the screen from the display device (81), it displays warning information on the screen indicating that the specific second data (D20) exhibits a different change trend from other second data (D2).

[0188] According to the above configuration, users can easily learn through the warning information displayed on the screen that there is target data (D10) that is highly likely to have been generated by noise or other influences. Therefore, it is less likely to lead to misinterpretations regarding degradation diagnosis.

[0189] The data processing system (1) according to the twelfth embodiment further comprises a confidence determination unit (44) in any one of the first to eleventh embodiments. The confidence determination unit (44) determines the confidence level regarding the change trend of a specific second data (D20) based on the change trend of reference data having the same type of features as the specific second data (D20). The display unit (47) displays information regarding the confidence level on the screen when the specific second data (D20) is displayed on the screen from the display device (81).

[0190] According to the above embodiment, users can easily determine how reliable a particular second data point (D20) is through the reliability information displayed on the screen. This makes it less likely to be misled when diagnosing degradation.

[0191] The data processing system (1) according to the 13th embodiment further comprises a diagnostic unit (46) in any one of the first to 12 embodiments. The diagnostic unit (46) determines the degree of deterioration of the target (OB1) with respect to the corresponding feature quantity for each of three or more second data (D2). The display unit (47) displays the three or more second data (D2) on a display device (81) screen in a manner that allows the degree of deterioration determined by the diagnostic unit (46) to be visually recognized.

[0192] According to the above embodiment, the user can easily determine the degree of deterioration of the object (OB1) through three or more second data (D2) displayed on the screen.

[0193] The 14th aspect of the data processing method relates to the diagnosis of deterioration of at least one of an object (OB1) in a servo system (7), namely a load (73) and a servo motor (72). The servo system (7) includes a load (73), a servo amplifier (71), and a servo motor (72) that provides power to the load (73) in accordance with the control of the servo amplifier (71). The data processing method includes an acquisition step, a first generation step, a second generation step, and a presentation step. In the acquisition step, at least one signal is acquired from among the control signals used to control the servo system (7) and the detection signals output from a sensor (61) that detects the state of the servo system (7). In the first generation step, first data (D1) relating to three or more feature quantities is generated from a predetermined region in the signal waveform of at least one signal. In the second generation step, three or more second data (D2) are generated, which are time-series data showing the progression of deterioration of the object (OB1), and each shows the change trend of the three or more feature quantities over time in the first data (D1). In the presentation step, three or more second data (D2) are compared with each other. Then, in the presentation step, if there is a specific second data (D20) that exhibits a different trend of change from the other second data (D2), a specific processing is performed on the target data (D10), which is the first data (D1) that caused the different trend of change, and the specific second data (D20) is presented. The specific processing includes processing related to at least one of the following: not presenting the target data (D10), correcting the target data (D10), and notifying the target data (D10).

[0194] According to the above embodiment, a data processing method can be provided that makes it less likely to mislead in deterioration diagnosis.

[0195] The program according to the 15th embodiment is a program that causes one or more processors to execute the data processing method according to the 14th embodiment.

[0196] According to the above embodiment, it is possible to provide a function that makes it less likely to mislead in deterioration diagnosis.

[0197] The configurations relating to the second to thirteenth aspects are not essential to the data processing system (1) and can be omitted as appropriate. [Industrial applicability]

[0198] This disclosure is industrially useful because it can improve the accuracy of degradation diagnosis by making it less likely to lead to misinterpretations in servo system degradation diagnosis, thereby improving the efficiency of equipment maintenance and inspection. [Explanation of symbols]

[0199] 1. Data Processing System 2 Acquisition part 3A 1st generation section 3B 2nd generation part 44. Confidence Determination Unit 46. ​​Diagnostic Department 47 Presentation section 61 Sensors 7 Servo System 71 Servo Amplifier 72 Servo motors 73 load 81 Display device D1, D1X 1st data D10 Target Data D2 Second Data D20 Specific second data OB1 Target ti starting point tf terminus

Claims

1. A data processing system for diagnosing deterioration of at least one of a load and a servo motor in a servo system including a load, a servo amplifier, and a servo motor that powers the load according to the control of the servo amplifier, An acquisition unit that acquires at least one signal from among the control signal used to control the servo system and the detection signal output from a sensor that detects the state of the servo system, A first generation unit generates first data relating to three or more feature quantities based on a predetermined region in the signal waveform of at least one of the signals, A second generation unit generates three or more second data sets, each showing the trend of change over time in the first data set for three or more feature quantities, which are time-series data indicating the progression of deterioration of at least one of the load and the servo motor. A presentation unit compares the three or more second data sets with each other, and if there is a specific second data set that exhibits a different change trend from the other second data sets, it performs specific processing on the target data, which is the first data set that caused the different change trend, and presents the specific second data set. Equipped with, The aforementioned specific processing includes processing relating to at least one of the following: not presenting the target data, correcting the target data, and providing notification regarding the target data. Data processing system.

2. The display unit presents the specific second data by displaying it on the screen of the display device. The data processing system according to claim 1.

3. The presentation unit presents the specific second data in a manner that includes the first data generated by the first generation unit at a certain point in time and the first data generated by the first generation unit before that certain point in time. The data processing system according to claim 1 or 2.

4. The presentation unit presents the specific second data in a manner that includes the first data generated by the first generation unit at a certain point in time and predicted data relating to the first data in the future after that point in time. A data processing system according to any one of claims 1 to 3.

5. The display unit further displays at least one of the three or more second data, other than the specific second data. A data processing system according to any one of claims 1 to 4.

6. The aforementioned specific processing includes processing related to not presenting the target data, The display unit prevents the display of the target data on the display device. A data processing system according to any one of claims 1 to 5.

7. The aforementioned specific processing includes processing related to the correction of the target data, The presentation unit corrects the target data based on the change trend of at least one of the three or more second data, excluding the specific second data, and presents the corrected target data. A data processing system according to any one of claims 1 to 6.

8. The aforementioned specific processing includes processing related to the correction of the target data, The presentation unit corrects the target data based on the past trend of change in the specific second data compared to the target data, and presents the corrected target data. A data processing system according to any one of claims 1 to 7.

9. When the display unit displays the specific second data on the screen from the display device, it displays the corrected target data on the screen in a display format that allows it to be distinguished from other first data. The data processing system according to claim 7 or 8.

10. The display unit corrects the target data using an approximation curve. A data processing system according to any one of claims 7 to 9.

11. The aforementioned specific processing includes processing of notifications regarding the target data, The display unit, when displaying the specific second data on the screen from the display device, displays warning information on the screen indicating that the specific second data exhibits a different change trend from the other second data. A data processing system according to any one of claims 1 to 10.

12. The system further includes a confidence determination unit that determines the confidence level regarding the change trend of the specific second data based on the change trend of reference data having the same type of features as the specific second data, The display unit displays information regarding the reliability when displaying the specific second data from the display device. A data processing system according to any one of claims 1 to 11.

13. The diagnostic unit further comprises a unit that determines the degree of deterioration of the target with respect to the corresponding feature quantity for each of the three or more second data sets. The display unit displays the three or more second data points on the screen from the display device in a manner that allows the degree of deterioration determined by the diagnostic unit to be visually confirmed. A data processing system according to any one of claims 1 to 12.

14. A data processing method for diagnosing deterioration of at least one of a load and a servo motor in a servo system including a load, a servo amplifier, and a servo motor that provides power to the load in accordance with the control of the servo amplifier, At least one signal is acquired from among the control signal used to control the servo system and the detection signal output from the sensor that detects the state of the servo system. From a predetermined region in the signal waveform of at least one of the aforementioned signals, first data relating to three or more feature quantities is generated. Time-series data showing the progression of deterioration of at least one of the load and the servo motor, generating three or more second data sets that each show the trend of change over time of the first data with respect to three or more feature quantities. The three or more second data sets are compared with each other, and if there is a specific second data set that exhibits a different trend of change from the other second data sets, a specific processing is performed on the target data, which is the first data set that caused the different trend of change, and the specific second data set is presented. The aforementioned specific processing includes processing relating to at least one of the following: not presenting the target data, correcting the target data, and providing notification regarding the target data. Data processing method.

15. A program for causing one or more processors to execute the data processing method described in claim 14.

Images