Concentrator fault pre-judgment method

Abstract

The invention discloses a concentrator fault pre-judgment method, and belongs to the technical field of power system information, and the method comprises the steps: obtaining historical operation data, historical business execution data and historical fault maintenance data of a concentrator; calculating the isolation degree of the historical operation data, associating the isolation degree of the historical operation data with the historical fault maintenance data, and determining an isolation degree fault threshold value; determining a concentrator availability threshold according to the historical service execution data; determining a fault point and a critical point of the concentrator according to the isolation degree fault threshold value and the availability threshold value of the concentrator; and predicting the remaining available time of the concentrator according to the isolation degree of the current operation data of the concentrator based on the fault point and the critical point of the concentrator. According to the method, fault pre-judgment can be carried out in the operation process of the concentrator, and the problem that fault maintenance of the concentrator is lagged in the prior art is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power system information technology, and particularly relates to a method for predicting concentrator faults. Background Art

[0002] With the rapid development of the smart grid, its structure is becoming increasingly complex and its scale is expanding day by day.

[0003] Currently, for the fault management of concentrators, it is limited to replacing or repairing the concentrators after the faults actually occur, and there are many factors restricting the replacement timeliness in some relatively remote areas.

[0004] In the field of power consumption information collection, there are quite high requirements for the collection timeliness and stability of lower-layer devices. In order to ensure stability, it is necessary to predict the operating state of the concentrator before an actual concentrator fault occurs, so as to formulate a maintenance strategy. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for predicting concentrator faults, which can perform fault prediction during the operation of the concentrator and solve the problem of lagging concentrator fault maintenance in the prior art.

[0006] To achieve the above object, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for predicting concentrator faults, including: Obtaining the historical operation data, historical service execution data, and historical fault repair data of the concentrator; Calculating the isolation degree of the historical operation data, and correlating the isolation degree of the historical operation data with the historical fault repair data to determine the isolation degree fault threshold; Determining the concentrator availability threshold according to the historical service execution data; Determining the fault point and critical point of the concentrator according to the isolation degree fault threshold and the concentrator availability threshold; Based on the fault point and critical point of the concentrator, predicting the remaining available time of the concentrator according to the isolation degree of the current operation data of the concentrator.

[0007] In combination with the first aspect, further, obtaining the historical operation data, historical service execution data, and historical fault repair data of the concentrator includes: Collecting the historical operation data, historical service execution data, and historical fault repair data of various models of concentrators according to different concentrator models; Performing data cleaning on the historical operation data, historical service execution data, and historical fault repair data of various models of concentrators to remove outliers, missing values, and duplicate data; Store the historical operation data, historical business execution data, and historical fault repair data of concentrators of various models after data cleaning in the database.

[0008] Combined with the first aspect, further, the historical operation data of the concentrator includes the CPU temperature of the concentrator, the ambient temperature, the one-time success rate of the first N task communications, and the time consumption of the first N task communications.

[0009] Combined with the first aspect, further, the isolation degree of the historical operation data is calculated based on the isolation forest algorithm, and the calculation formula is: ; Among them, represents the historical operation data in the historical operation data sample the isolation degree in, represents the normalized value of the historical operation data of the concentrator, represents the path length in the isolation tree, represents the historical operation data sample the average path length of.

[0010] Combined with the first aspect, further, correlating the isolation degree of the historical operation data with the historical fault repair data to obtain the isolation degree fault threshold includes: Correlate the isolation degree of the historical operation data with the historical fault repair data, analyze the change trend of the isolation degree of the historical operation data before the occurrence of historical faults, and establish an isolation degree curve; Normalize the isolation degree curve to determine the isolation degree fault threshold.

[0011] Combined with the first aspect, further, determining the fault point and critical point of the concentrator according to the isolation degree fault threshold and the concentrator availability threshold includes: When the isolation degree of the operation data of the concentrator is greater than the isolation degree fault threshold, it is determined that the concentrator enters the fault state, and the time point when the concentrator enters the fault state is used as the fault point of the concentrator; When the average time consumption of the business execution of the concentrator is greater than the concentrator availability threshold, it is determined that the concentrator enters the unavailable state, and the isolation degree and timestamp of the operation data corresponding to when the concentrator enters the unavailable state are used as the critical point of the concentrator.

[0012] Combined with the first aspect, further, the calculation formula for the average time consumption of the business execution of the concentrator is: ; Among them, represents the average time consumption of the business execution of the concentrator, represents the The time taken for one task represents the total number of tasks in the service window. When , it is determined that the concentrator enters an unavailable state, where represents the concentrator availability threshold, that is, the total time allowed for task execution by the service.

[0013] Combined with the first aspect, further, based on the fault point and critical point of the concentrator, predicting the remaining available time of the concentrator according to the isolation degree of the current operation data of the concentrator includes: Predicting the remaining time when the isolation degree of the current operation data of the concentrator reaches the isolation degree of the operation data corresponding to the fault point or critical point as the remaining available time of the concentrator according to the isolation degree of the current operation data of the concentrator.

[0014] Combined with the first aspect, further, it further includes: Formulating a maintenance plan for the concentrator according to the remaining available time of the concentrator.

[0015] Combined with the first aspect, further, giving priority to maintaining the concentrator with the remaining available time less than the preset time threshold.

[0016] In a second aspect, the present invention provides a concentrator fault prediction system, including: A data acquisition module, configured to acquire the historical operation data, historical service execution data, and historical fault repair data of the concentrator; A threshold determination module, configured to calculate the isolation degree of the historical operation data, correlate the isolation degree of the historical operation data with the historical fault repair data, and determine the isolation degree fault threshold; and configured to determine the concentrator availability threshold according to the historical service execution data; A prediction module, configured to determine the fault point and critical point of the concentrator according to the isolation degree fault threshold and the concentrator availability threshold; and configured to predict the remaining available time of the concentrator according to the isolation degree of the current operation data of the concentrator based on the fault point and critical point of the concentrator.

[0017] In a third aspect, the present invention provides a computer device, including: A storage medium: used to store computer programs; A processor: used to execute the computer program to implement the concentrator fault prediction method described in any item of the first aspect.

[0018] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the concentrator fault prediction method described in any item of the first aspect.

[0019] In a fifth aspect, the present invention provides a computer program product, including a computer program which, when executed by a processor, implements the concentrator fault prediction method according to any one of the first aspect.

[0020] Compared with the prior art, the beneficial effects of the present invention are as follows: The concentrator fault prediction method provided by the present invention collects the operation data of the concentrator in real time, combines the historical operation data and relevant fault characteristics, and can perform real-time evaluation and fault prediction on the working state of the concentrator during the operation of the concentrator, which can effectively alleviate the problem that only after-treatment can be carried out on the concentrator after a fault at present. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 is a flowchart of the concentrator fault prediction method provided by an embodiment of the present invention; Figure 2 is a schematic diagram of the temperature fault isolation degree curve of the operation data of the type-I concentrator provided by an embodiment of the present invention; Figure 3 is a schematic diagram of the communication fault isolation degree curve of the operation data of the type-I concentrator provided by an embodiment of the present invention; Figure 4 is a schematic diagram of the power-off fault isolation degree curve of the operation data of the type-I concentrator provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] The technical solutions of the present application will be further described in detail below in conjunction with the specific embodiments.

[0023] The embodiments of the present application will be described in detail below. The examples of the embodiments are shown in the drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the drawings are exemplary only for explaining the present application and should not be construed as limiting the present application. Without conflict, the embodiments of the present application and the technical features in the embodiments may be combined with each other.

[0024] An embodiment of the present application provides a concentrator fault prediction method. The concentrator fault prediction method provided by the embodiment of the present application can be applied to a terminal and can be executed by a concentrator fault prediction system. The concentrator fault prediction system can be implemented in a software and / or hardware manner, and the concentrator fault prediction system can be integrated in the terminal, for example: any tablet computer or computer device with a communication function.

[0025] Figure 1 is a flowchart of the concentrator fault prediction method provided by this embodiment. This flowchart only shows the logical order of the method of this embodiment. On the premise of not conflicting with each other, the steps shown or described can be completed in a different order from Figure 1 the order shown.

[0026] As shown Figure 1 below, the concentrator fault prediction method provided in this embodiment includes: Obtain the historical operation data, historical service execution data, and historical fault repair data of the concentrator; Calculate the isolation degree of the historical operation data, associate the isolation degree of the historical operation data with the historical fault repair data, and determine the isolation degree fault threshold; Determine the concentrator availability threshold according to the historical service execution data; Determine the fault point and critical point of the concentrator according to the isolation degree fault threshold and the concentrator availability threshold; Based on the fault point and critical point of the concentrator, predict the remaining available time of the concentrator according to the isolation degree of the current operation data of the concentrator.

[0027] The concentrator fault prediction method provided in this embodiment collects the operation data of the concentrator in real time, combines the historical operation data and relevant fault characteristics, and can perform real-time evaluation and fault prediction on the working state of the concentrator during the operation of the concentrator, which can effectively alleviate the problem that the concentrator can only be processed after the fault at present. Combining with the maintenance plan can improve the utilization rate of human resources and help the concentrator maintenance unit increase economic benefits.

[0028] In a possible embodiment, the concentrator fault prediction method specifically includes the following steps: Step 1: Obtain the historical operation data, historical service execution data, and historical fault repair data of the concentrator; In this embodiment, obtaining the historical operation data, historical service execution data, and historical fault repair data of the concentrator specifically includes the following steps: Step ①: According to the models of different concentrators, respectively collect the historical operation data, historical service execution data, and historical fault repair data of various models of concentrators; Specifically, according to the models of different type-I concentrators, respectively collect the historical operation data, historical service execution data, and historical fault repair data of M type-I concentrators in the past 3 years.

[0029] In this embodiment, the historical operation data of the concentrator includes the CPU temperature, ambient temperature, one-time success rate of the first N tasks, and communication time-consuming of the first N tasks of the concentrator.

[0030] Step ②: Clean the historical operation data, historical service execution data, and historical fault repair data of various models of concentrators, and remove outliers, missing values, and duplicate data; Step ③: Store the historical operation data, historical service execution data, and historical fault repair data of various models of concentrators after data cleaning in the database.

[0031] Step 2: Calculate the isolation degree of the historical operation data, associate the isolation degree of the historical operation data with the historical fault repair data, and determine the isolation degree fault threshold. In this embodiment, the isolation degree of the historical operation data is calculated based on the isolation forest algorithm, and the calculation formula is: ; where represents the historical operation data in the historical operation data sample the isolation degree in, represents the normalized value of the historical operation data of the concentrator, represents the path length in the isolation tree, represents the historical operation data sample the average path length of.

[0032] Associating the isolation degree of the historical operation data with the historical fault repair data to obtain the isolation degree fault threshold specifically includes: Step 1: Associate the isolation degree of the historical operation data with the historical fault repair data, analyze the change trend of the isolation degree of the historical operation data before the historical fault occurs, and establish an isolation degree curve. In this embodiment, the calculated isolation degree of the historical operation data is associated with the historical fault repair data, and the change trend of the isolation degree of the historical operation data in the N times before the fault occurs is analyzed.

[0033] According to the correlation analysis result, establish the isolation degree curve of the historical operation data of each model of concentrator, reflecting the change of the isolation degree of the historical operation data of the concentrator before the fault occurs.

[0034] Step 2: Normalize the isolation degree curve to determine the isolation degree fault threshold.

[0035] In this embodiment, the isolation degree curve is normalized so that the isolation degrees of the historical operation data of different models of concentrators are comparable.

[0036] According to the normalized isolation degree curve, determine the isolation degree fault threshold of each model of concentrator. When the isolation degree of the operation data of the concentrator is greater than the isolation degree fault threshold, it is determined that the concentrator enters the fault state.

[0037] Step 3: Determine the concentrator availability threshold according to the historical service execution data; In this embodiment, in combination with the data related to the task execution status of the concentrator, the availability of the concentrator is judged, and the availability critical point is set.

[0038] The calculation result of the isolation degree is combined with the data related to the task execution status of the concentrator to judge the availability of the concentrator. When the task execution status has affected the execution of the service, it is judged that the concentrator is unavailable.

[0039] According to the availability judgment result, the availability critical point Me of each concentrator model is set. When the isolation degree of the running data of the concentrator approaches or exceeds this critical point, it is judged that the concentrator is about to enter the unavailable state.

[0040] Step Four: Determine the fault point and critical point of the concentrator according to the isolation degree fault threshold and the concentrator availability threshold; In this embodiment, determining the fault point and critical point of the concentrator according to the isolation degree fault threshold and the concentrator availability threshold specifically includes: R1: When the isolation degree of the running data of the concentrator is greater than the isolation degree fault threshold, it is determined that the concentrator enters the fault state, and the time point when the concentrator enters the fault state is used as the fault point of the concentrator; R2: When the average time consumption of the service execution of the concentrator is greater than the concentrator availability threshold, it is determined that the concentrator enters the unavailable state, and the isolation degree and time stamp of the running data corresponding to when the concentrator enters the unavailable state are used as the critical point of the concentrator.

[0041] In this embodiment, the calculation formula for the average time consumption of the service execution of the concentrator is: ; Among them, represents the average time consumption of the service execution of the concentrator, represents the time consumption when the concentrator executes the th task within the service execution window, represents the total number of tasks within the service window. When , it is determined that the concentrator enters the unavailable state. Among them, represents the concentrator availability threshold, that is, the total time consumption allowed for task execution by the service.

[0042] Step Five: Based on the fault point and critical point of the concentrator, predict the remaining available time of the concentrator according to the isolation degree of the current running data of the concentrator; In this embodiment, based on the fault point and critical point of the concentrator, predicting the remaining available time of the concentrator according to the isolation degree of the current running data of the concentrator specifically includes: According to the isolation degree of the current running data of the concentrator, predicting the remaining time when the isolation degree of the current running data of the concentrator reaches the isolation degree of the running data corresponding to the fault point or critical point as the remaining available time of the concentrator.

[0043] Specifically, the remaining time until the isolation degree of the current operation data of the prediction concentrator reaches the isolation degree fault threshold is used as the remaining available time of the concentrator, or the remaining time until the isolation degree of the current operation data of the prediction concentrator reaches the isolation degree of the operation data corresponding to the critical point is used as the remaining available time of the concentrator.

[0044] Step Six: Develop a maintenance plan for the concentrator based on the remaining available time of the concentrator.

[0045] In this embodiment, the concentrators with remaining available time less than the preset time threshold are given priority for maintenance.

[0046] Develop a maintenance plan for the concentrator based on the three key indicators of the fault point, critical point, and remaining available time. When the remaining available time is short, giving priority to maintenance can avoid the concentrator from failing during the business peak period.

[0047] For the concentrator fault prediction method provided in this embodiment, data collection and preprocessing are the foundation, providing high-quality data for subsequent analysis. Isolation degree calculation and correlation analysis are based on the isolation forest algorithm to analyze the data and find the changing trend of the isolation degree before the fault occurs. Normalization and fault threshold determination are to standardize the isolation degree curve and determine the criteria for fault judgment. Availability judgment and critical point setting are to further refine the critical point of fault judgment in combination with the task execution status. Key indicator establishment and maintenance plan planning are based on the results of the previous steps to establish an operable maintenance strategy to ensure that the concentrator is maintained in a timely manner before the fault occurs. Through the above steps, a complete concentrator fault prediction technical solution is formed, which can effectively predict before the actual occurrence of the fault and plan the maintenance strategy to reduce the business impact brought by the fault.

[0048] In a possible embodiment, the isolation degree curve is combined with the time scale, and the isolation forest algorithm is used to speculate on the recent actual operation data of the Type I concentrator to obtain the equipment operation data for the next 7 to 10 days, and based on the historical operation indicators, the occurrence time of the critical point and the occurrence time of the fault point of the Type I concentrator are obtained, and finally the remaining available time length of the Type I concentrator is calculated.

[0049] In a possible embodiment, during the sampling process of the isolation forest algorithm, some data in the input data are randomly selected to establish a subset. The size of the subset is generally configured as 256. For each subset, the isolation forest algorithm separately establishes a binary isolation tree model. After all the isolation trees are established, the algorithm calculates the path length of the sample in each isolation tree. The longer the path, the higher the isolation degree.

[0050] In a possible embodiment, the isolation degree curve of the type-I concentrator's operation data is characterized and fitted to the historical operation critical point through the Pearson correlation coefficient. The closer the fitting result is to 1, the closer the type-I concentrator is to the operation critical point. If the correlation coefficient between the fitting result and the historical operation critical point is closer to 0, it indicates that the type-I concentrator has not failed, or the failure characteristics do not match the previously collected failures, and further judgment needs to be combined with business operation indicators.

[0051] As Figure 2 , Figure 3 , Figure 4 shown, as time goes by, the isolation degree of the type-I concentrator's operation data gets closer and closer to high, and quickly approaches the critical value obtained through training. In practical applications, various operation data of the type-I concentrator are collected in real time, and at the same time, the path distance of the tree is calculated for the collected operation data. As the path distance of the operation data from the normal operation data gets farther and farther, it indicates that the operation state of the type-I concentrator has entered an abnormal state. At the same time, the isolation degree of the operation data and the time scale are characterized and fitted to the historical operation critical point through the Pearson correlation coefficient. The closer the fitting result is to 1, the closer the type-I concentrator is to the operation critical point. If the correlation coefficient between the fitting result and the historical operation critical point is closer to 0, it indicates that the type-I concentrator has not failed, or the failure characteristics do not match the previously collected failures, and further judgment needs to be combined with business operation indicators. When the isolation degree of the operation data reaches the critical point, the time required for business execution further increases over time, but the business as a whole is still in an available state until the isolation degree of the operation data reaches the failure point. After reaching the failure point, the execution time of all services becomes negative, indicating that the service execution has failed.

[0052] The embodiment of the present application provides a concentrator fault prediction system, including: A data acquisition module, configured to acquire the historical operation data, historical business execution data, and historical fault repair data of the concentrator; A threshold determination module, configured to calculate the isolation degree of the historical operation data, associate the isolation degree of the historical operation data with the historical fault repair data, and determine the isolation degree fault threshold; and configured to determine the concentrator availability threshold according to the historical business execution data; A prediction module, configured to determine the fault point and critical point of the concentrator according to the isolation degree fault threshold and the concentrator availability threshold; and configured to predict the remaining available time of the concentrator based on the fault point and critical point of the concentrator and the isolation degree of the current operation data of the concentrator.

[0053] The concentrator fault prediction system provided by this embodiment can execute the concentrator fault prediction method provided by the embodiments of the present application, and has the corresponding functional modules and beneficial effects for executing the method.

[0054] An embodiment of the present application provides a computer device, including: A storage medium for storing a computer program; A processor for executing the computer program to implement the concentrator fault prediction method provided by the embodiments of the present application.

[0055] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the concentrator fault prediction method provided by the embodiments of the present application.

[0056] An embodiment of the present application provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the concentrator fault prediction method provided by the embodiments of the present application.

[0057] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0058] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0059] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0060] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable apparatus to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one process or multiple processes and / or blocks Figure 1 one process or multiple processes and / or blocks Figure 1 or steps of the functions specified in multiple blocks.

[0061] The above are only the preferred embodiments of the present application. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present application, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present application.

Claims

1. A concentrator fault prediction method, characterized in that: include: Obtain the historical operation data, historical service execution data and historical fault repair data of the concentrator; Calculate the isolation degree of historical operation data, and associate the isolation degree of historical operation data with historical fault repair data to determine the isolation degree fault threshold; Determine the concentrator availability threshold based on historical business execution data; Determine the failure point and critical point of the concentrator based on the isolation fault threshold and the concentrator availability threshold; Based on the failure points and critical points of the concentrator, the remaining available time of the concentrator is predicted according to the isolation degree of the current operation data of the concentrator.

2. The concentrator fault prediction method according to claim 1, characterized in that: Obtaining the historical operation data, historical service execution data, and historical fault repair data of the concentrator includes: According to the models of different concentrators, historical operation data, historical service execution data and historical fault maintenance data of various models of concentrators are collected respectively; Clean the historical operation data, historical business execution data, and historical fault repair data of various types of concentrators to remove outliers, missing values, and duplicate data; The historical operation data, historical business execution data and historical fault repair data of various types of concentrators that have undergone data cleaning are stored in the database.

3. The fault prediction method according to claim 1, characterized in that: The historical operation data of the concentrator includes the CPU temperature of the concentrator, the ambient temperature, the success rate of the previous N task communications, and the time consumed for the previous N task communications.

4. The concentrator fault prediction method according to claim 1, characterized in that: The isolation degree of historical operation data is calculated based on the isolation forest algorithm, and the calculation formula is: ； in, Indicates historical operation data Running data samples in history The degree of isolation in Represents the normalized value of the concentrator's historical operating data, express The path length in the isolation tree, Indicates historical running data samples The average path length.

5. The concentrator fault prediction method according to claim 1, characterized in that: Correlate the isolation degree of historical operation data with historical fault repair data to obtain the isolation degree fault threshold, including: The isolation degree of historical operation data is associated with the historical fault maintenance data, the isolation degree change trend of historical operation data before the historical fault occurs is analyzed, and the isolation degree curve is established; The isolation curve is normalized to determine the isolation fault threshold.

6. The concentrator fault prediction method according to claim 1, characterized in that: According to the isolation fault threshold and concentrator availability threshold, the failure points and critical points of the concentrator are determined as follows: When the isolation degree of the concentrator's operating data is greater than the isolation degree fault threshold, the concentrator is determined to be in a fault state, and the time point when the concentrator enters the fault state is taken as the fault point of the concentrator; When the average service execution time of the concentrator is greater than the concentrator availability threshold, the concentrator is determined to be in an unavailable state, and the isolation and timestamp of the operating data corresponding to the concentrator entering the unavailable state are used as the critical point of the concentrator.

7. The concentrator fault prediction method according to claim 6, characterized in that: The calculation formula for the average service execution time of the concentrator is: ； in, Indicates the average time taken for the concentrator to execute services. Indicates that the concentrator executes the service window The time taken for each task, Indicates the total number of tasks in the business window. When , the concentrator is determined to be in an unavailable state, where Indicates the concentrator availability threshold, that is, the total time required for task execution allowed by the business.

8. The concentrator fault prediction method according to claim 6, characterized in that: Based on the failure point and critical point of the concentrator and the isolation degree of the current operation data of the concentrator, the remaining available time of the concentrator is predicted to include: According to the isolation degree of the current operation data of the concentrator, the remaining time until the isolation degree of the current operation data of the concentrator reaches the isolation degree of the operation data corresponding to the failure point or the critical point is predicted as the remaining available time of the concentrator.

9. The concentrator fault prediction method according to claim 1, characterized in that: Also includes: Develop a maintenance plan for the concentrator based on the remaining available time of the concentrator.

10. The concentrator fault prediction method according to claim 1, characterized in that: Maintenance is prioritized for concentrators whose remaining available time is less than the preset time threshold.

Images