Abnormal inspection parameter adjustment display device

Abstract

The present invention relates to an abnormality inspection parameter adjustment and display device, which adjusts parameters for inspecting and determining operating abnormalities of mechanical equipment based on input from a user, collects operating information of the mechanical equipment, generates a data set based on the operating information, determines whether the data set represents a normal operating state based on the parameters, displays the data set through a graph, and adjusts the display to a position where the relationship between the data and the parameters of the graph can be easily visually grasped.

Description

Technical Field

[0001] The invention relates to an abnormality inspection parameter adjustment display device. Background Art

[0002] Mechanical equipment such as robots and machine tools operating in a factory are connected to a network and managed. Among the devices connected to the network laid in the factory, there are, for example, a device that monitors the overall operating status of the factory based on information about mechanical equipment, including the operating status of mechanical equipment connected to the network, and a device that determines the operating schedule of mechanical equipment based on a production plan or information about personnel and equipment.

[0003] In operation management software for managing the operation status of mechanical equipment, it is possible to detect mechanical equipment failures and issue alarms by setting arbitrary thresholds for signal data obtained by the user. However, there are many abnormal patterns in mechanical equipment failures that are difficult to detect with simple thresholds.

[0004] Therefore, as a method that can be used to diagnose whether newly acquired data is normal or abnormal, the k-nearest neighbor method is considered to be introduced. The k-nearest neighbor method is a method for determining whether a certain data point is close or far from other data points. It cannot be checked by a simple threshold, but can check data changes, etc. (for example, refer to Japanese Patent Publication No. 2011-070635).

[0005] For example, Figure 6 As exemplified above, it is considered that an operational abnormality of a mechanical device may be detected based on time series data obtained by acquiring data of pressure detected by a pressure sensor at a predetermined sampling period.

[0006] When the pressure value detected by the pressure sensor deviates from a certain track range, it is determined that the operation of the mechanical equipment is abnormal. Figure 6 The upper and lower thresholds of the pressure value are determined as shown, and when a pressure value exceeding the upper threshold is detected, or when a pressure value below the lower threshold is detected, it can be determined that the mechanical device is abnormal in operation. However, in the abnormal operation detection method using such thresholds, when the pressure detected by the pressure sensor is within the range of the upper and lower thresholds, it is not possible to detect the situation where the pressure deviates from the normal change as abnormal operation.

[0007] In such a case, the k-nearest neighbor method can be used to determine whether the action is abnormal based on whether the distance from the data point representing the change in the value of the normal pressure is close or far. In this way, the abnormality detection method of time series data based on the k-nearest neighbor method has the characteristic that it is easy for users to intuitively understand compared to other abnormality detection methods of time series data.

[0008] When checking for anomalies based on time series data using the k-nearest neighbor method, the user can determine the data window width w, the threshold value ε of the anomaly degree, and the number k of data for anomaly degree calculation, respectively, to adjust the accuracy of the anomaly check. However, it is not easy to reflect the user's feeling of whether to detect anomalies when the expected value of the time series data deviates from the expected value in the setting of these parameters. When the time series data is repeatedly monitored based on the determined parameters, when there is a judgment that deviates from the user's feeling (when the value that the user considers to be an abnormal value cannot be detected as an abnormal value, when the value that the user considers to be a normal value is detected as an abnormal value), it is necessary to repeat the operation such as adjusting the parameters and monitoring the time series data again, which increases the labor of the user. Summary of the invention

[0009] Therefore, an object of the present invention is to provide an abnormality inspection parameter adjustment display device which can easily determine parameters used for abnormality inspection.

[0010] The abnormal inspection parameter adjustment display device of the present invention solves the above-mentioned problem through the following structure: after the sample designated as normal data or abnormal data by the user from the signal data collected by the operation management device, etc. is displayed on the screen, the user can adjust each parameter while confirming the normal data and abnormal data on the screen.

[0011] Furthermore, the abnormality inspection parameter adjustment display device of the present invention visually supports the user to adjust the parameters for judging the abnormal operation of the mechanical equipment according to the operation information of the mechanical equipment, wherein the abnormality inspection parameter adjustment display device has: a parameter adjustment unit that adjusts the parameters according to the input from the user; a data collection unit that collects the operation information of the mechanical equipment; a data set generation unit that generates at least one data set according to the operation information; a judgment unit that judges whether the data set is a data set indicating a normal operation state or a data set indicating an abnormal operation state according to the parameters; and a display unit that generates display data for displaying the data set in a graph. Furthermore, the parameter adjustment unit has: an adjustment unit that is displayed at a position where the relationship between the data of the graph and the parameter can be visually grasped.

[0012] According to the present invention, when checking for anomalies in signal data that cannot be checked by a simple threshold, even a user lacking mathematical knowledge can adjust the parameters for determining the normality / abnormality of time series data through intuitive operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a schematic hardware configuration diagram of an abnormality check parameter adjustment display device according to one embodiment.

[0014] Figure 2 This is a schematic functional block diagram of an abnormality inspection parameter adjustment display device according to one embodiment.

[0015] Figure 3 is a graph representing an example of a dataset.

[0016] Figure 4 It is a figure which shows an example of the display screen displayed by a display part.

[0017] Figure 5 It is a diagram showing an example of an adjustment method provided by the parameter adjustment unit.

[0018] Figure 6 FIG. 1 is a diagram showing an example of abnormality determination of data in the related art. DETAILED DESCRIPTION

[0019] Figure 1 This is a schematic hardware configuration diagram showing main components of an abnormality check parameter adjustment display device according to an embodiment of the present invention.

[0020] The abnormality check parameter adjustment display device 1 can be implemented as a unit computer, a host computer, a cloud server or other computer connected to a control device for controlling a mechanical device via a network. In addition, the abnormality check parameter adjustment display device 1 can also be implemented as a control device for controlling a mechanical device itself, or a personal computer installed in conjunction with the control device involved.

[0021] Figure 1 An example is shown in which the abnormality check parameter adjustment display device 1 is implemented as a computer connected to a control device that controls a mechanical device via a network.

[0022] The CPU 11 included in the abnormality check parameter adjustment display device 1 of the present embodiment is a processor that controls the entire abnormality check parameter adjustment display device 1. The CPU 11 reads a system program stored in the ROM 12 via the bus 20, and controls the entire abnormality check parameter adjustment display device 1 according to the system program. Temporary calculation data or display data, various data input by the operator via an input unit (not shown), etc. are temporarily stored in the RAM 13.

[0023] The nonvolatile memory 14 is configured as a memory that can maintain the storage state even if the power supply of the abnormality check parameter adjustment display device 1 is disconnected, for example, by backing up with a battery (not shown). The nonvolatile memory 14 stores programs or data read from an external storage device (not shown), data input from an input device 71 such as a keyboard or a mouse and input via the interface 16, and various data obtained from the mechanical device 2 via the network 3 (and, via the interface 18). The program stored in the nonvolatile memory 14 can be expanded in the RAM 13 when used. In addition, various system programs required for the operation of the abnormality check parameter adjustment display device 1 are pre-written in the ROM 12. In addition, the abnormality check parameter adjustment display device 1 displays the information to be provided to the user on the display device 70 via the interface 15.

[0024] The abnormality check parameter adjustment display device 1 sets various parameters for the machine 2 via the interface 18. Furthermore, the abnormality check parameter adjustment display device 1 acquires various data necessary for managing the machine 2 from each part of the machine 2 via the interface 18.

[0025] Figure 2 This is a schematic functional block diagram of the abnormality check parameter adjustment display device 1 according to one embodiment of the present invention.

[0026] pass Figure 1 The CPU 11 of the abnormality check parameter adjustment display device 1 shown in the figure executes the system program to control the operation of each part of the abnormality check parameter adjustment display device 1, thereby realizing Figure 2 The functional blocks shown.

[0027] The abnormality inspection parameter adjustment display device 1 of this embodiment includes: a data collection unit 30, a data set generation unit 31, a determination unit 32, a display unit 33, and a parameter adjustment unit 34. In addition, a data set storage unit 40 and a parameter storage unit 41 are ensured on the non-volatile memory 14. The data set storage unit 40 stores the data set of the chart displayed on the display device 70, and the parameter storage unit 41 stores the parameters used to inspect the abnormality of the data.

[0028] The data collection unit 30 collects the operation information of the mechanical equipment 2 to be managed. Figure 2As shown, when the abnormality check parameter adjustment display device 1 is connected to the mechanical device 2 via the network 3, the data collection unit 30 collects operation information from each mechanical device 2 via the network 3. In addition, when the abnormality check parameter adjustment display device 1 is installed on the control device, the data collection unit 30 collects the operation information of the mechanical device 2 from each part of the control device. The data collection unit 30 can, for example, collect the operation information of the mechanical device 2 stored in a storage device (not shown). The operation information collected by the data collection unit 30 can be, for example, physical quantities such as the coordinate position or moving speed, acceleration, current value, voltage value of the motor driving the drive unit of the mechanical device 2, or physical quantities such as temperature, humidity, pressure, coordinate position, luminosity, etc. detected by sensors installed on the mechanical device 2. In addition, it can also be values ​​obtained based on these physical quantities. The data collection unit 30 obtains these data as time series data sampled at a predetermined sampling period, and stores the obtained data in the data set storage unit 40.

[0029] The data set generation unit 31 generates a data set displayed on the display device 70 based on the time series data collected by the data collection unit 30. ... Figure 3 As shown in the example, data in a specified range from the time series collected by the data collection unit 30 is generated as a data set. In this case, the numerical designation of the data range can be performed by the user's numerical designation of the data range using the input device 71, or the data range can be designated as follows. Figure 3 After the time series data is displayed in a graph on the display device 70 as shown, the numerical value of the data range is designated by the user through designation of the start point and the end point using the input device 71 (such as a pointing device).

[0030] The dataset generated by the dataset generation unit 31 is also marked to indicate whether the dataset is a normal dataset or an abnormal dataset as observed by the user. The dataset generation unit 31 can mark the dataset according to the user's designation. In addition, the dataset generation unit 31 can, for example, allow the user to generate a dataset based on time series data on the premise of generating an abnormal dataset, and automatically mark the abnormal data for the dataset thus generated. Even if any method is adopted, any method can be adopted as long as the label (normal / abnormal) specified by the user can be given to the dataset selected by the user.

[0031] When marking the dataset generated by the dataset generation unit 31 as normal / abnormal, the user may specify a data portion (data point, etc.) to be the subject of determination, and the accepted subject data portion may be associated with the dataset and stored in the dataset storage unit 40 .

[0032] For example, when performing the normal / abnormal determination illustrated by the k-nearest neighbor method, it is possible to determine whether a certain data point included in the data set is normal or abnormal when viewed from other data points. In this case, the data set generation unit 31 accepts the user's normal / abnormal marking of a certain point in the data set when performing the normal / abnormal marking. In addition, when performing the normal / abnormal determination on the entire data set (for example, on the pattern of data transition, etc.), it is not necessary to specify such a determination object data part.

[0033] The determination unit 32 determines each data set (and the determination target data portion) stored in the data set storage unit 40 based on the parameters stored in the current parameter storage unit 41, and stores the normal / abnormal determination of the data set as a result in association with each data set in the data set storage unit 40. As an example of the determination performed by the determination unit 32, the k-nearest neighbor method is illustrated, and in this case, the window width w of the data, the threshold value ε of the abnormality, and the number k of data for calculating the abnormality are stored in the parameter storage unit 41 as parameters for determination by the k-nearest neighbor method.

[0034] Furthermore, the determination unit 32 determines whether the determination object data portion is normal or abnormal for each data set stored in the data set storage unit 40 based on the window width w of the data stored in the parameter storage unit 41, the threshold value ε of the abnormality, and the number k of data for which the abnormality is calculated, and the determination result of the data set is associated with the data set and stored in the data set storage unit 40.

[0035] The display unit 33 displays each data set stored in the data set storage unit 40 on the display device 70 . Figure 4 This is an example of display of each data set stored in the data set storage unit 40 by the display unit 33 .

[0036] like Figure 4 As shown, the display unit 33 displays the normal / abnormal determination result of each data set obtained by the determination unit 32, so that the determination result of the data set can be grasped. Figure 4 As shown, the display unit 33 can display only the data sets assigned with the designated labels on the display device 70, and can display all the data sets on the basis of being able to grasp the labels assigned to each data set. When there are many data sets to be displayed, the display unit 33 can display each data set by a known display method such as scrolling or page switching. The display unit 33 can display so that the parameters stored in the current parameter storage unit 41 together with the data sets can be confirmed on the display device 70.

[0037] The parameter adjustment unit 34 adjusts the value of the parameter stored in the parameter storage unit 41 according to the input from the input device 71 by the user. Figure 5As shown in the example, the parameter adjustment unit 34 displays the adjustment method of each parameter on the graph of the data set displayed on the screen in correspondence with the display of the graph, and accepts the adjustment of the parameter value based on the user by accepting the input of the adjustment method. In this case, it is preferable to display the position where the parameter adjustment method is displayed at a position where the relationship between the data displayed on the graph and each parameter can be visually grasped. Figure 5 In the example shown, the parameter of the k-nearest method, i.e., the data window width w, is a parameter indicating the range of data from the data point to be determined as the determination criterion, and the threshold value ε of the abnormality is a parameter indicating the proportion of abnormal values ​​(data separation status). Figure 5 When the adjustment method of each parameter is displayed at the position shown, the meaning of the parameter value is more easily understood by the user through intuitive visual linkage.

[0038] When accepting the adjustment of the parameter value from the user, the parameter adjustment unit 34 stores the adjusted parameter value in the parameter storage unit 41, and instructs the determination unit 32 to perform a determination (re-determination) on each data set stored in the data set storage unit 40. After the determination (re-determination) process performed by the determination unit 32 is completed, the display unit 33 is instructed to update the display of the data set.

[0039] Through the above structure, the normal / abnormal determination of each data set by the determination unit 32 based on the adjusted parameters is changed in response to the user adjusting the parameters. For example, in a state where only the data set marked as abnormal is displayed on the screen, the user adjusts the parameters so that the determination unit 32 determines all the data sets displayed by adjusting the parameters as abnormal, so that even users with little mathematical knowledge can intuitively adjust the parameters for determining the normal / abnormality of time series data.

[0040] Furthermore, although the embodiments of the present invention have been described, the present invention is not limited to the examples of the above-mentioned embodiments, and can be implemented in various forms by adding appropriate changes.

[0041] For example, in the above-mentioned embodiment, although the k-nearest neighbor method is shown as an example of a determination method based on the determination unit 32, other determination methods may be used if it is a determination method that allows visual determination of normal / abnormality relative to a data set displayed on a screen by a graph, etc., and it is easy to visually grasp the relationship between the normal / abnormal determination and the parameters.

[0042] In addition, in the above-mentioned embodiment, although time series data (predetermined data for time axis changes) is exemplified as the object of determination, data based on the time axis may be used as the object as long as two or more values ​​change in conjunction. For example, it can also be applied to a data set representing the change in pressure for the change in the nozzle position in an injection molding machine.

Claims

1. An abnormality inspection parameter adjustment display device, which visually supports a user in adjusting a parameter for inspecting an abnormality in operation of a mechanical device based on operation information of the mechanical device, characterized in that: The abnormality inspection parameter adjustment display device comprises: a parameter adjustment unit, which adjusts the parameter according to an input from the user; A data collection unit that collects the operation information of the mechanical equipment; a data set generating unit for generating a plurality of data sets so that data in a range specified from the time series of the operation information are generated as one data set, and for marking the plurality of data sets in accordance with a designation from the user to indicate whether the data is normal or abnormal as observed by the user; a determination unit for determining, based on the parameters, whether the plurality of data sets are data sets indicating a normal operating state or data sets indicating an abnormal operating state; as well as a display unit that generates display data for displaying the plurality of data sets together with the determination result of the determination unit in a graph; The parameters used to determine the abnormal operation are the data window width, the threshold value of the abnormality, and the number of data for calculating the abnormality. The parameter adjustment unit includes: an adjustment unit displayed at a position where the relationship between the data of the graph and the parameter can be visually grasped; The determination unit re-determines the plurality of data sets after the parameter adjustment unit has adjusted the parameter, and the display unit displays the plurality of data sets together with the determination result after the re-determination in a graph.

Images