Information processing method, information processing device, display method, display device, program, recording medium, production method of article, and acquisition method of learning data

Abstract

To desirably obtain an information processing method and an information processing device that allow an operator to extract a plurality of pieces of arbitrary partial data from time-series data collected over a long period of time at a high sampling rate, and to easily perform work when performing data analysis work of checking and comparing the data.SOLUTION: Provided is an information processing method in which an information processing device: obtains time-series data of a physical amount related to a status of a machinery and event data related to an event occurring in the machinery; extracts a plurality of pieces of partial time-series data from the time-series data for a predetermined event included in the event data; and forms an image in which information related to each of the plurality of pieces of extracted partial time-series data is arranged so that a distance between the information is smaller than the case when the information are arranged on a linear scale with time used as an index.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an information processing method, an information processing apparatus, and the like.

Background Art

[0002] The operating state of a mechanical device can change moment by moment due to changes in the state of its components. If the operating state is within the allowable range in light of the purpose of use of the mechanical device, it is called a normal state, and if it is outside the allowable range, it is called a failure state. For example, in the case of a production machine, when it is in a failure state, problems such as manufacturing defective products or stopping the production line will occur.

[0003] In production machines and the like, in order to prevent the occurrence of failure states as much as possible, even when the same operation is repeatedly performed, it is common to perform maintenance work regularly or irregularly. To increase preventive safety, it is effective to shorten the implementation interval of maintenance work. However, since the production machine etc. is stopped during maintenance work, if the frequency of maintenance work is increased excessively, the operating rate of the production machine etc. will decrease. Therefore, it is desirable to be able to detect when a machine or the like is still in a normal state but the occurrence of a failure state is approaching. If it is possible to detect that the occurrence of a failure state is approaching (predict the occurrence of a failure), then maintenance work on the machine or the like can be performed at that time, so that it is possible to suppress the operating rate from decreasing more than necessary.

[0004] As a method for predicting the occurrence of a failure, a method is known in which a learned model obtained by machine learning the state of a mechanical device is created in advance, and the state of the mechanical device at the time of evaluation is evaluated using the learned model. To improve the prediction accuracy, it is important to construct a learned model suitable for failure prediction. For this purpose, it is important to prepare learning data (teacher data) used when generating a failure prediction model of a mechanical device by machine learning. To determine whether the extracted data is suitable as learning data, detailed data analysis such as waveform confirmation and comparison is required.

[0005] For example, in the data analysis method described in Patent Document 1, a plurality of partial time-series data are extracted from time-series data in which physical quantities of production equipment are associated with measurement times, and are plotted on a single graph with the elapsed time from a predetermined reference time as the axis for each partial time-series data. Then, each of the plotted partial time-series data is shifted in the direction of the elapsed time axis by a user operation, and the time-series data are compared by aligning the reference points.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0007] Generally, in a mechanical device, in order to manage its operating state, measurements are made on various parameters (physical quantities), and a huge amount of time-series data is acquired. In order to construct a learned model suitable for predicting the failure of a mechanical device, it is necessary to appropriately extract data from the acquired huge amount of time-series data and perform detailed data analysis operations such as waveform confirmation and comparison to determine whether the data is suitable for learning data.

[0008] However, in the case of a mechanical device such as an industrial robot installed on a production line, generally the frequency of occurrence of failures is not large, so it is necessary to collect time-series data over a long period of time. Furthermore, since the time-series data to be collected is data for managing the operating state of the mechanical device, the number of measurement parameters is large, and in order to analyze the waveform and the like in detail, it is necessary to increase the sampling rate, resulting in a huge amount of data to be collected. Thus, when extracting data related to an irregularly occurring failure from data collected at a high sampling rate over a long period of time and performing operations such as comparison, the conventional data display method places a heavy burden on the operator, causing problems in work efficiency and accuracy.

[0009] As the most basic method, when time-series data is displayed on a single graph with the horizontal axis being the measurement time, partial data related to a failure will be scattered irregularly within a long time axis direction, and it is not always the case that a plurality of partial data related to the failure will be displayed within the same screen. Also, when compressed in the time axis direction for display within the same screen, the graph waveform will be distorted despite the high sampling rate during measurement, making it difficult to confirm and compare the waveforms. For detailed examination, operations such as partial enlargement of the data by the operator himself / herself are required, which will take a great deal of time for the data analysis work.

[0010] Also, in Patent Document 1, from a vast amount of acquired time-series data, first, the time-series data to be added to the comparison graph is selected, and partial time-series data is extracted. Then, an elapsed time is assigned to each of the extracted partial time-series data, and alignment is performed so that the phases of the elapsed times of the plurality of partial time-series data are in-phase on the comparison graph, and they are displayed so as to overlap. By aligning the phases of the elapsed times of the plurality of partial time-series data to match on the comparison graph, comparison between the partial time-series data becomes possible, but the operation by the operator was complicated.

[0011] Therefore, when an operator extracts a plurality of arbitrary partial data from time-series data collected at a high sampling rate over a long period and performs a data analysis operation such as confirmation and comparison, an information processing method and an information processing apparatus that facilitate the operation have been demanded.

Means for Solving the Problem

[0012] A first aspect of the present invention is to acquire a plurality of time series data of a plurality of types of physical quantities related to the state of a mechanical device, extract a plurality of partial time series data corresponding to the physical quantities from the time series data, and display an image in which the partial time series data are arranged so as to be smaller than the distance between display regions corresponding to the partial time series data when displayed based on the time axis of the time series data, extract first partial time series data corresponding to an event that has occurred in the mechanical device from among the plurality of partial time series data, extract second partial time series data corresponding to the extracted first partial time series data, and display the image in which the first partial time series data and the second partial time series data are arranged in correspondence with each other. This is an information processing method characterized by the above.

[0013] Also, a second aspect of the present invention is to acquire a plurality of time series data of a plurality of types of physical quantities related to the state of a mechanical device, extract a plurality of partial time series data corresponding to the physical quantities from the time series data, and display an image in which the partial time series data are arranged so as to be smaller than the distance between display regions corresponding to the partial time series data when displayed based on the time axis of the time series data, extract first partial time series data corresponding to an event that has occurred in the mechanical device from among the plurality of partial time series data, extract second partial time series data corresponding to the extracted first partial time series data, and display the image in which the first partial time series data and the second partial time series data are arranged in correspondence with each other. This is an information processing apparatus characterized by the above. Further, a third aspect of the present invention is a display method for displaying a plurality of types of physical quantities related to the state of a mechanical device, wherein information related to partial time-series data extracted in plurality corresponding to the physical quantities from a plurality of time-series data of the physical quantities is displayed based on the time axis of the time-series data, and an image in which the partial time-series data is arranged so as to be smaller than the distance between display regions corresponding to the partial time-series data is displayed, time-series data related to a plurality of types of the physical quantities is acquired as the time-series data, a plurality of the partial time-series data are extracted for the plurality of types of the physical quantities from the time-series data, and an image in which information related to the plurality of the partial time-series data for the plurality of types of the physical quantities is arranged according to the number of samples or the number of operation cycles is displayed. This is a display method characterized by the above. Further, a fourth aspect of the present invention is a display device for displaying a plurality of types of physical quantities related to the state of a mechanical device, wherein information related to partial time-series data extracted in plurality corresponding to the physical quantities from a plurality of time-series data of the physical quantities is displayed based on the time axis of the time-series data, and an image in which the partial time-series data is arranged so as to be smaller than the distance between display regions corresponding to the partial time-series data is displayed, time-series data related to a plurality of types of the physical quantities is acquired as the time-series data, a plurality of the partial time-series data are extracted for the plurality of types of the physical quantities from the time-series data, and an image in which information related to the plurality of the partial time-series data for the plurality of types of the physical quantities is arranged according to the number of samples or the number of operation cycles is displayed. This is a display device characterized by the above.

Advantages of the Invention

[0014] The present invention can provide an information processing method and an information processing device that facilitate the work of an operator when extracting arbitrary partial data from the collected time-series data for confirmation, comparison, etc.

Brief Description of the Drawings

[0015]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Embodiments for Carrying Out the Invention

[0016] With reference to the drawings, an information processing method, an information processing apparatus, etc., which are embodiments of the present invention, will be described. In the drawings referred to in the following description of the embodiments, unless otherwise specified, elements denoted by the same reference numerals have the same functions.

[0017] FIG. 1 is a schematic diagram for explaining the configuration of functional blocks included in the information processing apparatus according to the embodiment. In FIG. 1, functional elements necessary for explaining the features of the present embodiment are represented by functional blocks, but descriptions of general functional elements not directly related to the problem-solving principle of the present invention are omitted. Further, each functional element illustrated in FIG. 1 is conceptually functional, and does not necessarily have to be physically configured as illustrated. For example, the specific form of dispersion or integration of each functional block is not limited to the illustrated example, and all or part of them can be functionally or physically dispersed and integrated in any unit according to the usage situation or the like.

[0018] As shown in FIG. 1, a time-series data display device 100 as an information processing apparatus according to the embodiment is communicably connected to a mechanical device 10 as a mechanical device to be measured. The mechanical device 10 is various industrial devices such as, for example, an industrial robot or a production device installed on a production line. Various sensors 11 for measuring physical quantities related to the state of the mechanical device are installed in the mechanical device 10. For example, when the mechanical device 10 is a multi-joint robot, sensors for measuring the current value of the motor that drives the joint, joint angle sensors, sensors for measuring speed, vibration, and sound, etc. can be installed. However, this is merely an example, and appropriate types and numbers of sensors can be installed as sensors 11 at appropriate positions depending on the type of the mechanical device 10, the working purpose, etc. Various sensors such as force sensors, torque sensors, vibration sensors, sound sensors, imaging sensors, distance sensors, temperature sensors, humidity sensors, flow sensors, pH sensors, pressure sensors, viscosity sensors, gas sensors, etc. can be used as the sensors 11. In FIG. 1, the sensor 11 is shown in a single number for the sake of illustration, but usually a plurality of sensors are installed communicably with the time-series data display device 100.

[0019] The mechanical device 10 is connected to the time-series data display device 100 as an information processing device in a wired or wireless manner so as to be communicable. The time-series data display device 100 can acquire the data measured by the sensor 11 through communication. Hereinafter, the functional blocks of the time-series data display device 100 will be described in order. The time-series data display device 100 includes a control unit 110, a storage unit 120, a display unit 130, and an input unit 140.

[0020] The control unit 110 includes a plurality of functional blocks. These functional blocks are configured by, for example, the CPU of the time-series data display device 100 reading and executing a control program stored in a storage device or a non-temporary recording medium. Alternatively, a part or all of the functional blocks may be configured by hardware such as an ASIC provided in the time-series data display device 100.

[0021] The storage unit 120 includes a time-series data storage means 121, an event data storage means 122, an extraction data storage means 123, and a combined data storage means 124. These means included in the storage unit 120 are configured by being appropriately allocated to the storage areas of storage devices such as hard disk drives, RAMs, and ROMs. The storage unit 120 is a data storage unit that stores various necessary data in order to create an image for easily displaying time-series data.

[0022] The display unit 130 and the input unit 140 are user interfaces provided in the time-series data display device 100. For the display unit 130, display devices such as liquid crystal displays and organic EL displays are used, and for the input unit 140, input devices such as keyboards, jog dials, mice, pointing devices, and voice input devices are used.

[0023] The data collection means 111 included in the control unit 110 acquires time-series data and event data related to the mechanical device 10 from the mechanical device 10, and stores them in the time-series data storage means 121 and the event data storage means 122, respectively. The data collection means 111 can also be called a data acquisition unit.

[0024] The data collection means 111 collects time-series data of physical quantities related to the state of the mechanical device, such as current, speed, pressure, vibration, sound, and temperature of each part, measured by the sensor 11 of the mechanical device 10, and stores it in the time-series data storage means 121. Alternatively, the data collection means 111 may calculate the maximum value, minimum value, average value, integral value, integral conversion value in the frequency domain, differential value, second differential value, etc. for each predetermined period of the measured values obtained from the sensor 11, and store them in the time-series data storage means.

[0025] In addition, the data collection means 111 collects event data related to events occurring in the mechanical device and stores it in the event data storage means 122. An event is set when the mechanical device reaches a predetermined state, and for example, time information when the event occurs is collected as event data and stored in the event data storage means 122. For example, when it is determined that an event occurs when a mechanical device that normally operates continuously (cycle operation) stops, the date and time when the stop state occurs are stored in the event data storage means 122. Events such as failures and maintenance that are the causes of the stop state generally occur irregularly or at long intervals, but the information processing device of the embodiment is suitable for handling events that occur discretely or irregularly in time in this way.

[0026] The data extraction means 112 extracts partial time-series data related to the event from the time-series data stored in the time-series data storage means 121 based on the event data stored in the event data storage means 122, and stores it in the extraction data storage means 123. The data extraction means 112 can also be called a data extraction unit.

[0027] For example, when the extraction condition is the stop of the mechanical device, the date and time data of the stop of the mechanical device is read from the event data storage means 122 as event data. Then, based on this event data, for example, the measured values of the sensors collected during the operation cycle immediately before the stop of the mechanical device are extracted and stored in the extraction data storage means 123 as partial time series data. Alternatively, the maximum value, minimum value, average value, integral value, integral conversion value in the frequency domain, differential value, second differential value, etc. for each predetermined period of the measured values before one operation of the stop of the mechanical device are extracted from the time series data storage means 121. And they are stored in the extraction data storage means 123 as partial time series data.

[0028] In addition, although the processing in the case where there is one type of event data stored in the event data storage means 122 has been described, there may also be a case where event data related to a plurality of types of events is stored in the event data storage means 122. In that case, the operator selects a predetermined event from among the plurality of types of events via the input unit 140, and the data extraction means 112 may extract the partial time series data related to the selected predetermined event and store it in the extraction data storage means 123. Alternatively, a predetermined event selected from among the plurality of types of events may be registered in advance, and the partial time series data related to the registered predetermined event may be automatically extracted and stored in the extraction data storage means 123.

[0029] The data combining means 113 creates a graph arranging the partial time series data related to the event based on the partial time series data stored in the extraction data storage means 123. The data combining means 113 can also be called an image creation unit. The data combining means 113 creates, for example, a graph in which the partial time series data related to the event is combined or a graph in which they are arranged in proximity along the horizontal axis indicating the number of data, and stores it in the combined data storage means 124. The created graph can be displayed on the display unit 130 or printed using a printing device (not shown) according to the needs of the operator (worker).

[0030] Next, FIG. 2 schematically shows an example of the hardware configuration of the time-series data display device according to the embodiment. As shown in FIG. 2, the time-series data display device can include PC hardware having a CPU 1601 as main control means, a ROM 1602 as a storage device, and a RAM 1603. Information such as a processing program for realizing the information processing method described later can be stored in the ROM 1602. Further, the RAM 1603 is used as a work area of the CPU 1601 when executing the information processing method. An external storage device 1606 is connected to the PC hardware. The external storage device 1606 is composed of an HDD, an SSD, an external storage device of another system mounted on a network, or the like.

[0031] The processing program of the CPU 1601 for realizing the information processing apparatus or the information processing method according to the embodiment can be stored in a storage unit such as the external storage device 1606 composed of an HDD, an SSD, etc., or the ROM 1602 (for example, EEPROM area). In that case, the processing program of the CPU 1601 for realizing the information processing method (for example, the time-series data display method) is supplied to each of the above storage units via the network interface 1607 and can be updated to a new different program. Alternatively, the processing program of the CPU 1601 for realizing the information processing method is supplied to each of the above storage units via storage means such as various magnetic disks, optical disks, flash memories, and a drive device therefor, and the content thereof can be updated. Various storage means, storage units, or storage devices in a state of storing a program capable of executing the processing of the CPU 1601 for realizing the information processing method are computer-readable recording media according to the information processing method or the information processing apparatus of the present invention.

[0032] The sensor 11 shown in FIG. 1 is connected to the CPU 1601. In FIG. 2, for the sake of simplicity of illustration, the sensor 11 is shown as being directly connected to the CPU 1601, but it may be connected via, for example, IEEE488 (so-called GPIB). Further, the sensor 11 may be configured to be communicably connected to the CPU 1601 via a network interface 1607 and a network 1608.

[0033] The network interface 1607 can be configured using communication standards such as wired communication such as IEEE 802.3, or wireless communication such as IEEE 802.11 and 802.15. The CPU 1601 can communicate with the external devices 1104 and 1121 via the network interface 1607. For example, if the object of time-series data display is an industrial robot, the external devices 1104 and 1121 may be a supervisory control device such as a PLC or a sequencer arranged for controlling and managing the industrial robot, or a management server.

[0034] In the example shown in FIG. 2, as a UI device (user interface device), an operation unit 1604 corresponding to the input unit 140 shown in FIG. 1 and a display device 1605 corresponding to the display unit 130 are connected. The operation unit 1604 can be constituted by a terminal such as a handy terminal, or a device such as a keyboard, jog dial, mouse, pointing device, voice input device (or a control terminal having them). The display device 1605 may be any device capable of displaying information related to the processing executed by the data extraction means 112, data combining means 113, etc. on the display screen, and for example, a liquid crystal display device can be used.

[0035] Next, with reference to the flowchart of FIG. 3, an information processing method (time-series data display method) executed by the time-series data display device 100 will be described. FIG. 3 shows an example of the processing procedure executed by the time-series data display device. First, in step S101, the time-series data display device 100 collects time-series data and event data from the mechanical device 10. FIG. 4(a) shows an example of the time-series data collected by the time-series data display device 100. This is a series of data obtained by periodically sampling and measuring the drive current of an industrial robot included in the mechanical device 10. The data collection means 111 of the time-series data display device 100 collects such time-series data from the sensor 11 of the mechanical device 10 and stores it in the time-series data storage means 121.

[0036] Here, the collected time-series data will be described in more detail. FIG. 5 illustrates, as an image in the form of a current waveform graph, the time-series data for one cycle operation when the industrial robot included in the mechanical device 10 is operating normally. Further, FIG. 6(a) illustrates, as an image in the form of a current waveform graph, the time-series data collected when the industrial robot is continuously performing cycle operations. FIG. 6(a) exemplifies the case where a waveform SPW with a specific amplitude is included in the graph. Furthermore, FIG. 6(b) illustrates a graph showing the time-series data collected over a long period with the time axis direction compressed more than in FIG. 6(a). In FIG. 6(b), although it can be seen that two waveforms SPW with specific amplitudes are included, since the waveforms of each cycle operation are compressed in the time axis direction, it can be understood that detailed confirmation and comparison between waveforms cannot be performed with such an image.

[0037] Next, FIG. 4(b) shows an example of the event data collected by the time-series data display device 100. This is data that records the time when an event occurred when the industrial robot included in the mechanical device 10 stopping is set as an event. In this example, stops due to maintenance work performed regularly or irregularly and stops due to faults occurring irregularly are treated as events. The data collection means 111 collects event data by receiving control information from, for example, a control unit that manages the operation of the mechanical device 10 in parallel with the collection of time-series data, and stores it in the event data storage means 122.

[0038] Returning to FIG. 3, in step S102, the data extraction means 112 extracts partial time-series data related to a predetermined event from the time-series data stored in the time-series data storage means. Here, the predetermined event refers to an event arbitrarily selected by an operator (worker) from the event data stored in the event data storage means 122, but it may be configured such that the control unit 110 automatically selects it.

[0039] For example, partial time-series data related to a predetermined event is extracted from the time-series data shown in FIG. 4(a) based on the event data shown in FIG. 4(b). Specifically, the partial time-series data related to the operation one cycle before the point in time when the selected predetermined event occurred (the point in time when the industrial robot stopped) is extracted as the partial time-series data. Note that this is just an example. For example, time-series data separated by a predetermined number of operation cycles from the occurrence of the predetermined event may be extracted, or time-series data for a plurality of consecutive operation cycles may be collectively extracted as the partial time-series data. Alternatively, it is also possible to extract the time-series data of the operation cycle itself in which the predetermined event occurred as the partial time-series data. The extracted partial time-series data is stored in the extraction data storage means 123 together with the time information related to the partial time-series data.

[0040] Here, first assume a case where the extracted partial time-series data is arranged and displayed on a linear scale (i.e., the absolute time axis) with time as an index. Fig. 7 schematically shows the display screen W. However, most of the time-series data during continuous operation is not drawn in the graph because it was not extracted, and only the waveform of the partial time-series data related to the event is drawn. Therefore, it can be said that the redundancy is significantly reduced compared to the graph in Fig. 6(b). However, when time-series data has been collected over a long period of time, on the display screen W, the waveform of the partial time-series data is compressed in the time-axis direction and flattened, so the details of the waveform cannot be confirmed. Also, in order to facilitate the observation of the waveform shape, when trying to expand the time-axis direction, since the partial time-series data are spaced non-uniformly from each other, when trying to compare and observe multiple waveforms, they may go out of the screen.

[0041] Therefore, in the embodiment, in step S103, the data combining means 113 as the processing unit performs a process of combining the partial time-series data stored in the extraction data storage means 123 and stores it in the combined data storage means 124. That is, an image (combined data) is created in which the information (for example, a graph) related to each of the extracted multiple partial time-series data is arranged so that the distance between them is smaller than when arranged on a linear scale with time as an index. The data combining means 113 arranges the information (for example, a graph) related to the partial time-series data so that the information (for example, a graph) related to adjacent partial time-series data is connected to each other or arranged with a predetermined short distance between them. For example, image processing is performed so that the distance in the horizontal axis direction separating the waveforms of each partial time-series data in Fig. 7 becomes zero or a small predetermined distance, thereby reducing the interval between the waveforms.

[0042] Then, in step S104, a graph is image-displayed on the display unit 130 by using the combined data stored in the combined data storage means 124. At this time, the horizontal axis direction can be enlarged as necessary so that waveform observation, comparison, etc. become easier. Also, it is preferable that the index (scale) on the horizontal axis of the graph be the sampling number of the original measurement data, the number of operation cycles, etc., rather than the absolute time. This is because, since the partial time series data that were separated at unequal intervals are arranged side by side, if the index (scale) on the horizontal axis is the absolute time, the values of the index will jump discontinuously at the boundaries between the partial time series data, making it difficult for the operator to intuitively understand.

[0043] Note that in step S104, instead of image-displaying the created image using the display unit 130, the image may be transmitted to a display device separate from the time series data display device 100 for display, or transmitted to a printing device for printing. That is, the created image may be output according to the convenience of the operator.

[0044] Fig. 8 exemplifies the image displayed on the display screen W of the display unit 130 in step S104. The waveforms of the partial time series data related to the events are displayed connected so as to be adjacent in the horizontal axis direction. That is, the fact that the industrial robot has stopped is treated as event data, and a graph in which partial time series data are extracted and combined for each event from the time series data monitoring the current value of the industrial robot is displayed. In this way, since only the partial time series data at the time of event occurrence are combined and displayed, the operator can very easily check and compare the graphs related to event occurrence.

[0045] For example, when the event (stop) is a stop for inspection performed under normal conditions of the mechanical device, the waveform of the partial time-series data is similar to the waveform of one operating cycle during normal operation shown in FIG. 5. According to the display image of the embodiment illustrated in FIG. 8, the operator can easily confirm the similarity of the waveforms. Also, when the event (stop) is a stop due to a failure of the mechanical device, the waveform of the partial time-series data becomes an abnormal waveform that is dissimilar to the normal waveform, such as ABN1 and ABN2 shown in FIG. 8. Thus, since an abnormal waveform different from the normal time can be easily found or compared in association with the event, the operator can easily extract the learning data for creating the failure prediction model.

[0046] The example of FIG. 8 is a case where the extraction condition (predetermined event) in step S102 includes both a stop for inspection during normal operation and a stop during a failure, but the operator can change the extraction condition (predetermined event) in step S102 according to the work purpose. For example, when only the waveforms related to the stop due to a failure are to be compared and the correlation between the cause of the failure and the waveform is to be verified, the predetermined event as the extraction condition in step S102 may be set to a failure stop.

[0047] For example, Fig. 9(a) is a display image when the stop due to a failure is set as a predetermined event, and the waveforms of the partial time-series data related to the predetermined event are connected and displayed so as to be adjacent in the horizontal axis direction. In this example, the index on the horizontal axis is the number of operation cycles, and a vertical line is added at the connected position of the graphs so that the boundaries between events are easy to understand. Fig. 9(b) is detailed information related to the event content stored in the extraction data storage means 123. In Fig. 9(a), the detailed information of the event content is displayed corresponding to the waveform of the partial time-series data using this information. An operator can easily grasp from the waveform of the graph displayed on the screen and the detailed information of the event content that in the case of a failure stop resulting from an excessive motor load, the maximum value of the waveform peak tends to become abnormally high as a sign thereof. Also, in the case of a failure stop resulting from a brake failure, it can be easily grasped that the number of peaks observed during one operation cycle increases as a sign thereof. In this way, by grasping in detail the content of the event for extracting the partial time-series data, it becomes easy to grasp the characteristics of the time-series data extracted by each event, and it is possible to easily determine the suitability of adoption as learning data used for machine learning. Therefore, an operator can efficiently and easily extract the learning data when creating a failure prediction model.

[0048] In addition, in order to improve the work efficiency of the operator, an input area where the operator can input information may be provided in the image in addition to the waveforms of the connected partial time-series data and the detailed information related to the event. For example, check boxes, pull-down menus, flags, etc. for the operation of extracting waveforms as learning data by the operator may be displayed in the image. Alternatively, a field for the operator to enter comments or memos may be provided and displayed in the image.

[0049] FIG. 10 shows another example of the display image according to the embodiment. In this example, adjacent partial time-series data (graphs) are connected and arranged so as to be separated by a predetermined short distance so that an operator can easily visually recognize the boundary of the partial time-series data. Further, a mark indicating information related to the event content is added as a label to each graph. In this example, as labels representing sub-classifications related to the device stop which is an event, a mark indicating a stop in a normal state (for example, inspection) and a mark indicating a stop due to an abnormal state (for example, failure) are set, and are label-displayed in the image in correspondence with each graph. Further, above each label, a check box for selecting a waveform to be adopted as learning data for creating a failure prediction model from the waveforms is displayed. The labels and the check boxes may be displayed by an operator instructing via the input unit 140, or the control program may be configured to display them automatically.

[0050] In the examples described above, partial time-series data of a single type of physical quantity such as “current value” is extracted and graphs are arranged and displayed along the horizontal axis. However, the graphs displayed on one screen are not limited to those related to partial time-series data of a single type of physical quantity. If graphs of partial time-series data related to a plurality of types of physical quantities can be displayed in the same screen, for example, an operator can easily determine the correlation between different physical quantities in the event, which is convenient when extracting learning data for creating a failure prediction model.

[0051] FIG. 11 is another example of a display image according to an embodiment. In this example, in step S102 of the flowchart in FIG. 3, for a predetermined event of "device stop", partial time-series data of "current value" and "pressure" were extracted. Then, in step S103, for each of the "current value" and "pressure", the extracted partial time-series data were combined. Further, in step S104, the graph of the "current value" and the graph of the "pressure" were arranged vertically, and the phases of the events in the horizontal axis direction were aligned and displayed. As a result, it can be seen that when an abnormal waveform occurs in which the peak value in the current value becomes excessive, an abnormal waveform occurs in which the peak value in the pressure becomes too small, and the operator can easily understand that the correlation between the current value and the pressure is large in this event. On the other hand, even if an abnormal waveform occurs in which the number of peaks observed during one operation cycle in the current value increases, since the pressure has a normal waveform, it can be seen that the correlation between the current value and the pressure is small in this event. In this way, since only the partial time-series data at the time when a predetermined event occurs are combined and displayed, the operator (operator) can extremely easily check and compare the graphs related to the event occurrence. The operator (operator) can efficiently and easily extract learning data when creating a failure prediction model.

[0052] [Example of connecting the time-series data display device to a robot] FIG. 12 shows an example of connecting the time-series data display device 100 of the embodiment to a 6-axis articulated robot as an example of the mechanical device 10. Links 200 to 206 provided in the 6-axis articulated robot are connected in series by six rotational joints J1 to J6. The 6-axis articulated robot is provided with sensors for measuring the rotational speed of the motor of each rotational joint, sensors for measuring the rotational angle of the joint, torque sensors, sensors for measuring the current of the motor, pressure sensors for measuring the pressure of the air for driving the actuator, and the like. For example, a robot hand 210 can be attached to and detached from the tip link as an actuator.

[0053] A six-axis articulated robot is communicably connected to the time-series data display device 100 of the embodiment. The time-series data display device 100 collects time-series data of physical quantities related to the state of the robot and event data related to events occurring in the robot. The six-axis articulated robot repeats, for example, an operation of manufacturing an article by assembling parts. An operator can give a command via the input unit 140 to the time-series data display device 100 to create an image and display or print the image.

[0054] For example, when the six-axis articulated robot performs an operation of manufacturing an article, an image combining graphs of partial time-series data related to a predetermined event (for example, a failure) can be created and displayed on the display unit 130. By performing such a display, the operator can easily confirm the past history of the robot related to the predetermined event, and thus can make a judgment as to whether it is possible to further continue the manufacturing operation of the article by the robot or not. That is, if the time-series data display device of the present invention is connected to a manufacturing device such as a robot and partial time-series data is displayed, it is possible to manufacture an article while preventing a stop due to a failure in advance.

[0055] In addition, the operator can create teacher data (learning data) used to construct a learned model for predicting a failure of the robot using the time-series data display device 100. The operator selects a predetermined event from the event data acquired by the time-series data display device 100, causes the time-series data display device 100 to extract partial time-series data from the time-series data of various physical quantities, and can display an image that allows easy comparison of graphs and the like. For example, by using the check boxes illustrated in FIG. 10, the operator can easily flag the data determined to be suitable as teaching materials for machine learning, and can easily construct teacher data (learning data).

[0056] [Other Embodiments] Note that the present invention is not limited to the embodiments described above, and many modifications are possible within the technical idea of the present invention. For example, the implementation of the present invention is not limited to the graphical display of physical quantities related to a single type of event. For example, in step S102 of the flowchart in FIG. 3, a plurality of types of predetermined events are set as extraction conditions. Then, for each of the plurality of types of predetermined events in step S103, partial time-series data of physical quantities is extracted and a graph combined along the horizontal axis is created, and can be arranged and displayed within one screen in step S104. This is convenient when an operator verifies whether there is a correlation between a plurality of types of events regarding the physical quantity.

[0057] The present invention can also be realized by supplying a program that realizes one or more functions of the embodiment to a system or device via a network or a storage medium, and having one or more processors in the computer of the system or device read and execute the program. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions. A control program capable of executing the information processing method or display method of the embodiment, and a non-temporary recording medium readable by a computer storing the control program are also included in the embodiments of the present invention. Also, in this embodiment, a six-axis articulated robot has been described as an example of the mechanical device 10, but it is not limited thereto. For example, based on the information in the storage device provided in the control device, a machine that can automatically perform operations of expansion and contraction, flexion and extension, vertical movement, horizontal movement, or turning, or a combination of these operations can be applied as the mechanical device 10.

Explanation of Reference Numerals

[0058] 10... Mechanical device / 11... Sensor / 100... Time series data display device / 110... Control unit / 111... Data collection means / 112... Data extraction means / 113... Data combination means / 120... Memory unit / 121... Time series data memory means / 122... Event data memory means / 123... Extracted data memory means / 124... Combined data memory means / 130... Display unit / 140... Input unit / 200 - 206... Link / 1601... CPU / 1602... ROM / 1603... RAM / 1604... Operation unit / 1605... Display device / 1606... External memory device / 1607... Network interface / 1608... Network / 1104, 1121... External device / J1 - J6... Rotary joint / W... Display screen

Claims

1. Acquire multiple time series data of multiple types of physical quantities related to the state of the machine; extracting a plurality of partial time series data corresponding to the physical quantities from the time series data; displaying an image in which the partial time series data is arranged so that the distance between display areas corresponding to the partial time series data is smaller than a distance between display areas corresponding to the partial time series data when the partial time series data is displayed based on a time axis of the partial time series data; extracting a first partial time series data corresponding to an event that has occurred in the machine from among the plurality of partial time series data; extracting second partial time series data corresponding to the extracted first partial time series data; displaying the images arranged in correspondence with the first partial time series data and the second partial time series data; 23. An information processing method comprising:

2. A plurality of the partial time series data are extracted based on event data relating to the event.

2. The information processing method according to claim 1,

3. acquiring, as the event data, event data relating to a plurality of types of events occurring in the machine, extracting a plurality of the partial time series data relating to a plurality of types of predetermined events selected from the plurality of types of events, and displaying the image in which information relating to the plurality of the partial time series data for the plurality of types of predetermined events is arranged.

3. The information processing method according to claim 2.

4. The image includes information related to the event.

4. The information processing method according to claim 2 or 3.

5. The event is set based on a peak of the physical quantity.

5. The information processing method according to claim 2, wherein the first and second inputs are input to the first and second inputs.

6. In the event, At least one of the following is set: when the value of the peak is equal to or greater than a predetermined threshold value; or when the number of times of the peak is equal to or greater than a predetermined number.

6. The information processing method according to claim 5,

7. The event data is date and time data on when the event occurred.

7. The information processing method according to claim 2, wherein the first and second inputs are input to the first and second inputs.

8. The time series data before the partial time series data is extracted is arranged on a linear scale using time as an index.

8. The information processing method according to claim 1,

9. The image is displayed on a display unit.

9. The information processing method according to claim 1,

10. The image includes an image in which graphs of the physical quantities related to the extracted plurality of partial time series data are linked together.

10. The information processing method according to claim 1,

11. The image includes an image in which graphs of the physical quantities related to the extracted plurality of partial time series data are arranged at a predetermined distance from each other.

11. The information processing method according to claim 1,

12. The image includes an input area into which an operator can input information.

12. The information processing method according to claim 1,

13. The input area allows a label indicating a classification of the partial time series data to be set.

13. The information processing method according to claim 12.

14. the partial time series data is displayed on the image regardless of the time axis in the time series data before the partial time series data is extracted.

14. The information processing method according to claim 1,

15. The partial time series data is displayed on the image according to a sampling number or a number of operation cycles.

15. The information processing method according to claim 1,

16. arranging the partial time series data in the image so that the distance between display areas corresponding to the partial time series data is smaller than the distance between display areas corresponding to the partial time series data when the partial time series data is arranged in the time series data before the partial time series data is extracted; 16. The information processing method according to claim 1,

17. Acquire multiple time series data of multiple types of physical quantities related to the state of the machine; extracting a plurality of partial time series data corresponding to the physical quantities from the time series data; displaying an image in which the partial time series data is arranged so that the distance between display areas corresponding to the partial time series data is smaller than a distance between display areas corresponding to the partial time series data when the partial time series data is displayed based on a time axis of the partial time series data; extracting a first partial time series data corresponding to an event that has occurred in the machine from among the plurality of partial time series data; extracting second partial time series data corresponding to the extracted first partial time series data; displaying the images arranged in correspondence with the first partial time series data and the second partial time series data; 23. An information processing apparatus comprising:

18. A display method for displaying a plurality of types of physical quantities related to a state of a mechanical device, comprising: displaying an image in which the partial time series data is arranged so that the distance between display areas corresponding to the partial time series data, the distance being smaller than a distance between display areas corresponding to the partial time series data when information related to the partial time series data extracted from the plurality of time series data of the physical quantity is displayed based on a time axis of the time series data; acquiring time series data relating to a plurality of types of the physical quantities as the time series data, extracting a plurality of the partial time series data for the plurality of types of the physical quantities from the time series data, and displaying the image in which information relating to the plurality of the partial time series data for the plurality of types of the physical quantities is arranged according to a sampling number or a number of operation cycles; A display method comprising:

19. A display device that displays a plurality of types of physical quantities related to a state of a mechanical device, displaying an image in which the partial time series data is arranged so that the distance between display areas corresponding to the partial time series data, the distance being smaller than a distance between display areas corresponding to the partial time series data when information related to the partial time series data extracted from the plurality of time series data of the physical quantity is displayed based on a time axis of the time series data; acquiring time series data relating to a plurality of types of the physical quantities as the time series data, extracting a plurality of the partial time series data for the plurality of types of the physical quantities from the time series data, and displaying the image in which information relating to the plurality of the partial time series data for the plurality of types of the physical quantities is arranged according to a sampling number or a number of operation cycles; A display device comprising:

20. A program for causing a computer to execute the information processing method according to any one of claims 1 to 16 or the display method according to claim 18.

21. A computer-readable recording medium having the program according to claim 20 recorded thereon.

22. An article is manufactured using the machine whose state information is acquired by the information processing device according to claim 17. A method for producing an article comprising the steps of:

23. The information processing device according to claim 17 creates the image, and displays the image to an operator when acquiring learning data for creating a trained model for predicting a failure of the mechanical device. A method for acquiring learning data comprising the steps of:

Images