A method and system for automatic fault detection of distribution network

Abstract

The present application provides a method and system for automatic fault detection of a distribution network frame, and relates to the technical field of distribution network fault detection. The method comprises: establishing a frame communication with a distribution network frame, and obtaining construction data of the distribution network frame; configuring a linkage monitoring node mapped with a working mode; when the distribution network frame is working, locating the linkage monitoring node according to the working mode; reading the timing input data of the distribution network frame, and performing timing compensation on the located linkage monitoring node; generating a timing data set; performing anomaly determination, wherein the anomaly decision channel is constructed by the construction data; and reporting an equipment anomaly of the distribution network frame according to the anomaly determination result, thereby solving the technical problem in the prior art that anomaly detection is inaccurate due to a relatively simple monitoring data analysis process, and achieving the technical effect of improving the accuracy and precision of anomaly detection, and thus improving the operation safety of the distribution network frame.

Description

Technical Field

[0001] The present application relates to the technical field of distribution network fault detection, and in particular to a distribution network grid automatic fault detection method and system. Background Art

[0002] The distribution network grid refers to the grid structure composed of power lines, transformers, circuit breakers, load switches, lightning arresters and other distribution equipment. It can transmit electricity to the user end to achieve the distribution and transmission of electric energy. As an important part of the power system, once a failure occurs in the distribution network grid, it will have a serious impact on the power supply capacity of the power system, causing power outages, equipment damage and damage to power facilities. Therefore, fault detection is of great significance. Traditional fault detection mostly uses sensors for abnormal detection and early warning, but due to the relatively simple monitoring data analysis process, abnormal detection is inaccurate. Summary of the invention

[0003] The present application provides a method and system for automatic fault detection of a distribution network framework, which is used to solve the technical problem in the prior art that inaccurate abnormality detection occurs due to a relatively simple monitoring data analysis process.

[0004] According to a first aspect of the present application, a method for automatic fault detection of a distribution network frame is provided, comprising: establishing a frame communication with a distribution network frame, completing information interaction based on the frame communication, and obtaining construction data of the distribution network frame, wherein the construction data includes frame structure data, frame attribute data, and frame working data; configuring a linkage monitoring node mapped to a working mode, wherein the linkage monitoring node includes a monitoring node of a single device and a summary node for result supervision; when the distribution network frame is working, obtaining the working mode of the distribution network frame in real time, and locating the linkage monitoring node according to the working mode; reading the timing input data of the distribution network frame, and performing timing compensation on the located linkage monitoring node; performing data monitoring of the single device and output result of the distribution network frame according to the linkage monitoring node, and generating a timing data set; sending the timing compensation and the timing data set to an abnormal decision channel to perform abnormal determination, wherein the abnormal decision channel is constructed by the construction data; and reporting equipment abnormalities of the distribution network frame according to the abnormal determination result.

[0005] According to a second aspect of the present application, there is provided a distribution network grid automatic fault detection system, comprising: a grid communication interaction module, the grid communication interaction module is used to establish grid communication with the distribution network grid, and complete information interaction based on the grid communication, and obtain construction data of the distribution network grid, wherein the construction data includes grid structure data, grid attribute data, and grid working data; a linkage monitoring node configuration module, the linkage monitoring node configuration module is used to configure linkage monitoring nodes mapped to working modes, wherein the linkage monitoring nodes include monitoring nodes of single devices and summary nodes for result supervision; a working mode acquisition module, the working mode acquisition module is used to obtain the distribution network grid in real time when the distribution network grid is working. working mode, and locates the linkage monitoring node according to the working mode; a timing compensation module, the timing compensation module is used to read the timing input data of the distribution network grid, and perform timing compensation on the located linkage monitoring node; a data monitoring module, the data monitoring module is used to perform data monitoring of the single device and output results of the distribution network grid according to the linkage monitoring node, and generate a timing data set; an abnormality determination module, the abnormality determination module is used to send the timing compensation and the timing data set to an abnormal decision channel, and perform abnormality determination, the abnormality decision channel is built by the construction data; an equipment abnormality reporting module, the equipment abnormality reporting module is used to report equipment abnormalities of the distribution network grid according to the abnormality determination result.

[0006] According to one or more technical solutions adopted in this application, the beneficial effects that can be achieved are as follows:

[0007] Establish grid communication with the distribution network grid, and complete information interaction based on grid communication to obtain the construction data of the distribution network grid, wherein the construction data includes grid structure data, grid attribute data, and grid working data, configure linkage monitoring nodes mapped with working modes, wherein the linkage monitoring nodes include monitoring nodes of single equipment and summary nodes of result supervision, and when the distribution network grid is working, obtain the working mode of the distribution network grid in real time, locate the linkage monitoring nodes according to the working mode, read the timing input data of the distribution network grid, and perform timing compensation on the located linkage monitoring nodes; perform data monitoring of single equipment and output results of the distribution network grid according to the linkage monitoring nodes, generate a timing data set, send the timing compensation and timing data set to the abnormal decision channel, and perform abnormal judgment. The abnormal decision channel is built by the construction data, and the equipment abnormality of the distribution network grid is reported according to the abnormal judgment result. In this way, the linkage monitoring nodes are obtained based on the working mode for abnormal analysis and judgment, so as to improve the accuracy and precision of abnormal detection, and thus improve the technical effect of the operation safety of the distribution network grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in this application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. The drawings constituting a part of this application are used to provide a further understanding of this application. The illustrative embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation on this application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without creative work.

[0009] Figure 1 A schematic diagram of a flow chart of a method for automatic fault detection of a distribution network provided in an embodiment of the present application;

[0010] Figure 2 A schematic diagram of the structure of an automatic fault detection system for a distribution network framework provided in an embodiment of the present application.

[0011] Explanation of the reference numerals: grid communication interaction module 11 , linkage monitoring node configuration module 12 , working mode acquisition module 13 , timing compensation module 14 , data monitoring module 15 , abnormality determination module 16 , equipment abnormality reporting module 17 . DETAILED DESCRIPTION

[0012] In order to make the purpose, technical solution and advantages of the present application more obvious, the exemplary embodiments of the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0013] The terms used in the specification are used to describe the embodiments, rather than to limit the present application. As used in the specification, the singular terms "a", "an" and "the" are intended to also include plural forms, unless the context clearly indicates otherwise. When used in the specification, the terms "comprise" and / or "include" specify the presence of steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other steps, operations, elements, components and / or groups thereof.

[0014] Unless otherwise defined, all terms (including technical and scientific terms) used in this specification shall have the same meaning as commonly understood by those skilled in the art to which this application belongs. Terms, such as those defined in commonly used dictionaries, should not be interpreted in an idealized or overly formal sense unless explicitly defined herein. Throughout the specification, the same reference numerals represent the same elements.

[0015] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0016] Embodiment 1

[0017] Figure 1 A method for automatic fault detection of a distribution network frame provided in an embodiment of the present application is shown in FIG. The method comprises:

[0018] Establishing grid communication with the distribution network grid, and completing information exchange based on the grid communication to obtain construction data of the distribution network grid, wherein the construction data includes grid structure data, grid attribute data, and grid working data;

[0019] The distribution network grid refers to a grid structure composed of power distribution equipment such as power lines, transformers, circuit breakers, load switches, and lightning arresters, which can deliver power to the user end to achieve the distribution and transmission of electric energy. The distribution network grid in the embodiment of the present application refers to any distribution network grid to be automatically fault detected. To establish communication with the distribution network grid, wireless communication technology or wired communication technology in the prior art can be used to establish communication through an existing communication interface or communication module. Based on the established grid communication, information interaction is performed by sending and receiving data packets to obtain construction data from the distribution network grid, wherein the construction data includes grid structure data, grid attribute data, and grid working data. The grid structure data includes data characterizing the composition of the distribution network grid structure, such as the layout of the power line, the type of equipment, and the connection method of the equipment. The grid attribute data includes data such as the type and specification of the equipment. The grid working data includes the equipment operation status of the power lines, transformers, circuit breakers and other equipment in the distribution network grid.

[0020] Configure a linkage monitoring node mapped with the working mode, wherein the linkage monitoring node includes a monitoring node of a single device and a summary node for result supervision;

[0021] The linkage monitoring node mapped with the working mode is configured, and the linkage monitoring node includes the monitoring node of the single device and the summary node of the result supervision. The linkage monitoring node is a node that monitors the operating status of the distribution network. Among them, the monitoring node of the single device refers to the node used to monitor the working status of a single device, such as the temperature of the transformer, the current of the power line, etc., that is, a sensor is set in each device as the monitoring node of the single device. The summary node of the result supervision refers to a data communication node used to summarize and supervise the operating status of all devices in the distribution network. Specifically, the linkage monitoring node has a mapping relationship with the working mode. The working mode refers to the operating mode of the distribution network when responding to different power distribution needs. That is to say, under different power distribution needs, the number of transformers, circuit breakers, load switches and other equipment used in the distribution network may be different. For example, when the power distribution demand is small, only a part of the equipment will be started to perform power distribution work, so it is only necessary to monitor the equipment that performs the power distribution work. Specifically, the historical distribution operation records of the distribution network in the past period of time can be retrieved to extract all possible working modes and the corresponding equipment performing the distribution work. Each device performing the distribution work is used as a monitoring node of a single device, and the temperature, current, voltage, etc. of each device are monitored; at the same time, a summary node is established to summarize and supervise the operating status of all equipment, thereby obtaining the monitoring nodes of the single devices corresponding to the various working modes and the summary nodes for result supervision, and the monitoring nodes of the single devices and the summary nodes for result supervision are used to form linkage monitoring nodes corresponding to the various working modes, and the various working modes and the corresponding linkage monitoring nodes are mapped and associated according to the corresponding relationship.

[0022] When the distribution network is working, the working mode of the distribution network is obtained in real time, and the linkage monitoring node is located according to the working mode;

[0023] When the distribution network is working, the working mode of the distribution network is obtained in real time, that is, the equipment performing the distribution work is obtained as the working mode, and based on the linkage monitoring nodes configured for the working mode, the linkage monitoring nodes corresponding to the real-time working mode are matched and obtained.

[0024] Reading the timing input data of the distribution network grid and performing timing compensation on the located linkage monitoring nodes;

[0025] Read the time series input data of the distribution network grid. The time series input data refers to the data that changes over time in the distribution network grid. In simple terms, when distributing electricity through the distribution network grid, the voltage, current, etc. will be controlled through the control center of the distribution network grid. The time series input data is the control parameters of the voltage, current, etc. input at different continuous times. The time series input data is used to perform time series compensation on the located linkage monitoring nodes. That is, by monitoring the operation of the distribution network grid through the linkage monitoring nodes, monitoring data at different times can be obtained. The monitoring data will have a collection time mark. The collection time should correspond to the time series input data. Compare the collection time and the time series input data. If the collection time is different from the time series input data, compensation correction is performed, thereby realizing the time series compensation of the linkage monitoring nodes, ensuring the consistency of the timing, and improving the monitoring accuracy.

[0026] Performing data monitoring of the single devices and output results of the distribution network framework according to the linkage monitoring node to generate a time series data set;

[0027] The data of the single equipment and output results of the distribution network grid are monitored according to the linkage monitoring node, that is, the real-time collection of the equipment operation status is carried out through the sensors set by the linkage monitoring node, such as temperature sensors, voltage sensors, etc., to generate a time series data set, which is the monitoring data collected by the linkage monitoring node in a continuous time.

[0028] Sending the timing compensation and the timing data set to an abnormality decision channel to perform abnormality determination, wherein the abnormality decision channel is constructed by the construction data;

[0029] The abnormal decision channel is constructed by the construction data, that is, the abnormal decision channel is constructed through the grid structure data, grid attribute data, and grid working data in the construction data, and then the timing compensation and the timing data set are sent to the abnormal decision channel to perform abnormal judgment. The specific process is described in detail as follows.

[0030] In a preferred embodiment, it also includes:

[0031] The basic construction of the abnormal decision channel is completed based on the construction data, wherein the construction process includes calling the grid attribute data to construct the abnormal decision sub-network of the single device; after the basic construction is completed, the abnormal decision sub-network feedback link under each working mode is performed through the linkage monitoring node to complete the construction compensation of the abnormal decision channel; when the timing compensation and the timing data set are sent to the abnormal decision channel, each abnormal decision sub-network is compensated according to the timing compensation, and the output data of the abnormal decision sub-network is linked and transmitted through the feedback link as the abnormal detection compensation data of the linked abnormal decision sub-network; the abnormal detection of all abnormal decision sub-networks is summarized through the summary sub-network, and the node result of the summary node is verified; the abnormal judgment is completed according to the verification and summary results.

[0032] The basic construction of the abnormal decision channel is completed based on the construction data, wherein the construction process includes calling the grid attribute data to construct the abnormal decision sub-network of the single device. The abnormal decision sub-network is a neural network in the existing machine learning. The construction data includes grid structure data, grid attribute data, and grid working data. The grid attribute data includes the type and specification of the equipment, that is, the grid attribute data is used to call the grid working data corresponding to each device to train and construct the abnormal decision sub-network. Specifically, the grid working data is used as the normal working data of the equipment, and it is stored in the abnormal decision sub-network corresponding to each single device. The subsequent abnormal decision sub-network performs abnormal analysis by comparing the difference between the time series data set and the normal working data. After the basic construction is completed, the abnormal decision sub-network feedback link under each working mode is performed through the linkage monitoring node, that is, the abnormal decision sub-network corresponding to each single device is connected for communication, and the construction compensation of the abnormal decision channel is completed, and multiple connected abnormal decision sub-networks are obtained to form an abnormal decision channel.

[0033] After the timing compensation and the timing data set are sent to the abnormal decision channel, compensation is performed on each abnormal decision sub-network according to the timing compensation, that is, the timing of the timing data set is corrected according to the timing compensation, and then the output data of the abnormal decision sub-network is linked and transmitted through the feedback link. The output data of the abnormal decision sub-network is the abnormal detection result, that is, the timing data set is compared with the normal working data. If the difference between the two exceeds the allowed deviation threshold, the corresponding timing data and the single device are output as the abnormal detection result as the abnormal detection compensation data of the linked abnormal decision sub-network. The abnormal detection of all abnormal decision sub-networks is summarized through the summary sub-network, that is, the abnormal detection compensation data of all abnormal decision sub-networks are integrated together as the summary result, and the node result of the summary node is verified, that is, the abnormal detection compensation data in the summary result of the summary node is once again checked for abnormality, that is, the abnormal detection compensation data is compared with the normal working data, and the difference between the two is once again analyzed for abnormality. If the difference between the two still exceeds the deviation threshold, it is determined that the device is abnormal, and this is used as the abnormal determination result to improve the accuracy of abnormal detection.

[0034] In a preferred embodiment, it also includes:

[0035] When the linked abnormality decision subnetwork receives the corresponding output data, the output data is used as input judgment data, and the collected output data is used as result judgment data to execute the processing judgment of the current single device; the processing judgment result is recorded to generate an abnormal data set of the single device; the abnormality judgment of the single device is performed according to the abnormal data set, wherein the abnormality judgment includes point discrete judgment and trend abnormality judgment; and the abnormality detection summary is completed according to the judgment result.

[0036] When the linked abnormal decision sub-network receives the corresponding output data, the output data is used as the input judgment data. The output data is the output data of the abnormal decision sub-network, which is used as the input judgment data. The collected output data is used as the result judgment data, that is, the output data of the abnormal decision sub-network is used as the basis for abnormal judgment, and the processing judgment of the current single device is performed, that is, the output data of the abnormal decision sub-network is determined whether the device is abnormal, and this is recorded as the processing judgment result. If the device is abnormal, the corresponding time series data set is used as the abnormal data set of the single device. Then, the abnormal judgment of the single device is performed according to the abnormal data set, wherein the abnormal judgment includes point discrete judgment and trend abnormal judgment, that is, based on the existing technology, the data in the abnormal data set is discretely analyzed. Exemplarily, the data in the abnormal data set is classified using the existing classification method, and based on the classification result, only the data in the clustering result with the least number of clusters is filtered out as the point discrete judgment result. At the same time, an abnormal trend analysis is performed on the data in the abnormal data set, that is, the deviation direction of each data relative to the normal working data of the equipment is analyzed, and whether the deviation direction in the abnormal data set is consistent is determined. This is used as the trend abnormality judgment result, and the point discrete judgment result and the trend abnormality judgment result are used to form the judgment result. The abnormal detection summary is completed according to the judgment result, that is, the abnormal detection compensation data output by the abnormal decision subnetwork is integrated and summarized according to the judgment result, which is convenient for subsequent abnormal judgment.

[0037] Report equipment abnormalities in the distribution network based on the abnormality determination results.

[0038] Specifically, the abnormality of the equipment in the distribution network is reported according to the abnormality determination result, that is, the abnormality of the equipment in the distribution network is reported according to the abnormality determination result through the existing sound, light and electricity display technology, so as to remind the staff to inspect and maintain the equipment with the reported abnormality.

[0039] In a preferred embodiment, it also includes:

[0040] Environmental data of the distribution network is collected through an environmental sensor group to establish an environmental data set; the equipment environmental impact of the distribution network is initialized based on the construction data; the fluctuation tolerance of the environmental impact is expanded through the equipment environmental impact initialization result and the environmental data set; a tolerance judgment threshold for anomaly detection is generated based on the fluctuation tolerance expansion result; and abnormal judgment compensation of the abnormal decision channel is performed through the tolerance judgment threshold.

[0041] Specifically, the environmental data of the distribution network is collected through an environmental sensor group. The environmental sensor group may include sensors such as temperature and humidity sensors for detecting environmental parameters. The sensor group is used to perform environmental detection to establish an environmental data set. The environmental data set includes parameters such as the temperature and humidity of the environment in which the distribution network is located. The equipment environmental impact of the distribution network is initialized based on the construction data. That is, the environment may have an impact on the abnormal detection of the distribution network. For example, if the ambient temperature is too high, the temperature of the equipment of the distribution network will increase, resulting in inaccurate abnormal detection and false alarms. Therefore, the equipment environmental impact initialization of the distribution network based on the construction data is to perform environmental impact analysis based on the construction data. For example, based on the existing technology, the construction data of the distribution network under normal operation in different environments is retrieved for analysis, and the relationship between the construction data and the environment is obtained, which is used as the equipment environmental impact initialization result. The fluctuation tolerance of the environmental impact is expanded through the initialization result of the equipment environment impact and the environmental data set, that is, according to the relationship between the construction data and the environmental changes in the initialization result of the equipment environment impact, the environmental data set is analyzed and calculated to obtain the fluctuation impact of the environmental data set on the construction data, for example, it will cause the construction data to change within a certain range, which is used as the fluctuation tolerance expansion result. The tolerance judgment threshold of anomaly detection is generated according to the fluctuation tolerance expansion result, that is, the threshold when performing anomaly detection, such as the deviation threshold of the time series data set and the normal working data, is compensated with the fluctuation tolerance expansion result to obtain the tolerance judgment threshold, and the abnormal judgment compensation of the abnormal decision channel is performed through the tolerance judgment threshold, thereby improving the accuracy of anomaly detection by analyzing the impact of the environment.

[0042] In a preferred embodiment, it also includes:

[0043] Continuously monitor the single device for abnormal judgment and compensation to generate a tolerance trigger value; perform trigger accumulation verification within a time window on the tolerance trigger value; if the trigger accumulation verification within any time window meets the preset window threshold, generate window warning information; perform maintenance management of the single device based on the window warning information.

[0044] Specifically, the single device is continuously monitored for abnormal judgment compensation, that is, abnormal judgment monitoring is performed at a continuous time to generate a tolerance trigger value, wherein the tolerance trigger value refers to the parameter value when the time series data set reaches the tolerance judgment threshold. The tolerance trigger value is subjected to a trigger cumulative verification within a time window, that is, the number of triggers is accumulated each time the time series data set reaches the tolerance judgment threshold, and the accumulated number is used as the trigger cumulative verification. If the trigger cumulative verification within any time window meets the preset window threshold, a window warning information is generated, wherein the preset window threshold is set by professional and technical personnel in this field based on actual conditions. The maintenance and management of the single device is performed according to the window warning information, that is, the window warning information is generated and sent to the staff to remind the staff to perform equipment inspection and maintenance to ensure the normal operation of the equipment.

[0045] In a preferred embodiment, it also includes:

[0046] Equipment feature extraction is performed according to the grid working data to obtain equipment feature extraction results; sensitive trigger factors of equipment are established based on the equipment feature extraction results; constraint units are constructed with the sensitive trigger factors, and the constraint units are coupled to the abnormal decision channel to complete abnormal determination.

[0047] The device feature extraction is performed based on the grid working data. The device feature extraction can adopt various methods in the prior art, including time domain analysis, frequency domain analysis, nonlinear analysis, etc. Exemplarily, the time domain analysis can extract the parameter features of the device during operation as the device feature extraction result by analyzing the waveform, peak value, mean value and other features of the time domain signals such as current, voltage and temperature. Further, based on the device feature extraction result, a sensitive trigger factor of the device is established. The sensitive trigger factor is a reference parameter for monitoring the operating state of the device. The floating range of the device operating parameter can be obtained based on the device feature extraction result as a sensitive trigger factor, such as the operating range of voltage, temperature, etc. The sensitive trigger factor is constructed into a constraint unit, and the constraint unit is coupled to the abnormal decision channel to complete the abnormal judgment. That is, the constraint unit is constructed with the sensitive trigger factor to assist the abnormal decision channel in making abnormal judgment. When the time series data set of the distribution network exceeds the sensitive trigger factor, it can be considered that the device is abnormal, thereby improving the accuracy of abnormal judgment.

[0048] In a preferred embodiment, it also includes:

[0049] Perform maintenance judgment on the reported equipment abnormality and record the maintenance judgment data; perform time series clustering on the maintenance judgment data and perform common deviation judgment on the time series clustering results; generate compensation parameters through the common deviation judgment results, and perform channel feedback optimization of the abnormal decision channel based on the compensation parameters.

[0050] Perform maintenance judgment on the reported equipment abnormality and record maintenance judgment data. Maintenance judgment data refers to the abnormal data that is corrected and determined by the staff after the maintenance judgment. Perform time series clustering on the maintenance judgment data, that is, cluster the abnormal data belonging to the same time in time order. Specifically, the existing clustering method can be used for clustering, and the common deviation judgment of the time series clustering results is performed. That is, the abnormal judgment of the reported equipment may deviate from the actual abnormality. The actual abnormal data, that is, the maintenance judgment data, is obtained after the staff performs maintenance. The difference data between the maintenance judgment data and the abnormal judgment in the time series clustering results is analyzed as the common deviation judgment result. Compensation parameters are generated through the common deviation judgment result, and the difference degree of abnormal judgment is calculated based on the difference data in the common deviation judgment result. The calculation result is used as the compensation parameter. The channel feedback optimization of the abnormal decision channel is performed based on the compensation parameter, that is, the compensation parameter is multiplied by the abnormal judgment output by the abnormal decision channel to reduce the deviation of the abnormal judgment and improve the accuracy of the abnormal judgment.

[0051] Based on the above analysis, it can be seen that one or more technical solutions provided by this application can achieve the following beneficial effects:

[0052] Establish grid communication with the distribution network grid, and complete information interaction based on grid communication to obtain the construction data of the distribution network grid, wherein the construction data includes grid structure data, grid attribute data, and grid working data, configure linkage monitoring nodes mapped with working modes, wherein the linkage monitoring nodes include monitoring nodes of single equipment and summary nodes of result supervision, and when the distribution network grid is working, obtain the working mode of the distribution network grid in real time, locate the linkage monitoring nodes according to the working mode, read the timing input data of the distribution network grid, and perform timing compensation on the located linkage monitoring nodes; perform data monitoring of single equipment and output results of the distribution network grid according to the linkage monitoring nodes, generate a timing data set, send the timing compensation and timing data set to the abnormal decision channel, and perform abnormal judgment. The abnormal decision channel is built by the construction data, and the equipment abnormality of the distribution network grid is reported according to the abnormal judgment result. In this way, the linkage monitoring nodes are obtained based on the working mode for abnormal analysis and judgment, so as to improve the accuracy and precision of abnormal detection, and thus improve the technical effect of the operation safety of the distribution network grid.

[0053] Embodiment 2

[0054] Based on the same inventive concept as the method for automatic fault detection of a distribution network in the aforementioned embodiment, Figure 2 As shown, the present application also provides a distribution network automatic fault detection system, the system comprising:

[0055] A grid communication interaction module 11, the grid communication interaction module 11 is used to establish grid communication with the distribution network grid, and complete information interaction based on the grid communication to obtain construction data of the distribution network grid, wherein the construction data includes grid structure data, grid attribute data, and grid working data;

[0056] A linkage monitoring node configuration module 12, wherein the linkage monitoring node configuration module 12 is used to configure linkage monitoring nodes mapped to the working mode, wherein the linkage monitoring nodes include monitoring nodes of single devices and summary nodes of result supervision;

[0057] A working mode acquisition module 13, wherein the working mode acquisition module 13 is used to acquire the working mode of the distribution network in real time when the distribution network is working, and locate the linkage monitoring node according to the working mode;

[0058] A timing compensation module 14, which is used to read the timing input data of the distribution network grid and perform timing compensation on the located linkage monitoring nodes;

[0059] A data monitoring module 15, the data monitoring module 15 is used to perform data monitoring of the single devices and output results of the distribution network frame according to the linkage monitoring node, and generate a time series data set;

[0060] An abnormality determination module 16, the abnormality determination module 16 is used to send the timing compensation and the timing data set to an abnormality decision channel to perform abnormality determination, and the abnormality decision channel is constructed by the construction data;

[0061] The device abnormality reporting module 17 is used to report the abnormality of the equipment in the distribution network according to the abnormality determination result.

[0062] Furthermore, the abnormality determination module 16 further includes:

[0063] Based on the construction data, the basic construction of the abnormal decision channel is completed, wherein the construction process includes calling the grid attribute data to construct the abnormal decision sub-network of the single device;

[0064] After the basic construction is completed, the abnormal decision sub-network feedback link under each working mode is carried out through the linkage monitoring node to complete the construction compensation of the abnormal decision channel;

[0065] After the timing compensation and the timing data set are sent to the abnormal decision channel, after each abnormal decision sub-network is compensated according to the timing compensation, the output data of the abnormal decision sub-network is linked and transmitted through the feedback link as abnormal detection compensation data linked to the abnormal decision sub-network;

[0066] The anomaly detection of all abnormal decision sub-networks is summarized through the summary sub-network, and the node results of the summary node are verified;

[0067] Complete anomaly determination based on verification and summary results.

[0068] Furthermore, the abnormality determination module 16 further includes:

[0069] When the link abnormality decision sub-network receives the corresponding output data, the output data is used as input determination data, and the collected output data is used as result determination data to execute the processing determination of the current single device;

[0070] Record the processing judgment results and generate abnormal data sets for individual devices;

[0071] Performing abnormality determination of a single device according to the abnormal data set, wherein the abnormality determination includes point discrete determination and trend abnormality determination;

[0072] Complete the abnormal detection summary based on the judgment results.

[0073] Furthermore, the system further includes an abnormality determination compensation module, and the abnormality determination compensation module includes:

[0074] Collect environmental data of the distribution network through the environmental sensor group and establish an environmental data set;

[0075] Initializing the equipment environment impact of the distribution network frame according to the construction data;

[0076] Expanding the fluctuation tolerance of the environmental impact through the device environmental impact initialization result and the environmental data set;

[0077] Generate a tolerance threshold for anomaly detection based on the fluctuation tolerance expansion result;

[0078] The tolerance judgment threshold is used to perform abnormal judgment compensation on the abnormal decision channel.

[0079] Furthermore, the system further includes a single device maintenance management module, and the single device maintenance management module includes:

[0080] Continuously monitor the individual devices for abnormal judgment and compensation, and generate tolerance trigger values;

[0081] Performing trigger accumulation verification within a time window on the tolerance trigger value;

[0082] If the cumulative verification of triggers within any time window meets the preset window threshold, a window warning message is generated;

[0083] The maintenance and management of the single equipment is performed according to the window warning information.

[0084] Furthermore, the system further comprises a constraint unit coupling module, and the constraint unit coupling module comprises:

[0085] Extracting equipment features according to the grid working data to obtain equipment feature extraction results;

[0086] Establishing a sensitive trigger factor of a device based on the device feature extraction result;

[0087] The sensitive trigger factor is used to construct a constraint unit, and the constraint unit is coupled to the abnormal decision channel to complete the abnormality determination.

[0088] Furthermore, the system further includes a channel feedback optimization module, and the channel feedback optimization module includes:

[0089] Make maintenance judgments on reported equipment abnormalities and record maintenance judgment data;

[0090] Performing time series clustering on the maintenance determination data, and performing common deviation determination on the time series clustering results;

[0091] Compensation parameters are generated through the common deviation determination results, and channel feedback optimization of the abnormal decision channel is performed based on the compensation parameters.

[0092] The specific example of a distribution network grid automatic fault detection method in the aforementioned embodiment 1 is also applicable to a distribution network grid automatic fault detection system in this embodiment. Through the aforementioned detailed description of a distribution network grid automatic fault detection method, technical personnel in this field can clearly know a distribution network grid automatic fault detection system in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here.

[0093] It should be understood that various forms of the processes shown above can be used to reorder, add or delete steps, as long as the expected results of the technical solutions disclosed in this application can be achieved, and this document does not limit this.

[0094] Note that the above are only preferred embodiments of the present application and the technical principles used. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present application. Therefore, although the present application is described in more detail through the above embodiments, the present application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the appended claims.

Claims

1. A method for automatic fault detection of a distribution network, characterized in that: The method comprises: Establishing grid communication with the distribution network grid, and completing information exchange based on the grid communication to obtain construction data of the distribution network grid, wherein the construction data includes grid structure data, grid attribute data, and grid working data; Configure a linkage monitoring node mapped with the working mode, wherein the linkage monitoring node includes a monitoring node of a single device and a summary node for result supervision; When the distribution network is working, the working mode of the distribution network is obtained in real time, and the linkage monitoring node is located according to the working mode; Reading the timing input data of the distribution network grid and performing timing compensation on the located linkage monitoring nodes; Perform data monitoring of the single devices and output results of the distribution network framework according to the linkage monitoring node to generate a time series data set; Sending the timing compensation and the timing data set to an abnormality decision channel to perform abnormality determination, wherein the abnormality decision channel is constructed by the construction data; Report equipment abnormalities in the distribution network frame according to the abnormality determination results; Wherein, the method further comprises: Expand the tolerance for fluctuations in environmental impacts; Generate a tolerance threshold for anomaly detection based on the fluctuation tolerance expansion result; Performing abnormal determination compensation of an abnormal decision channel by using the tolerance determination threshold; Continuously monitor the individual devices for abnormal judgment and compensation, and generate tolerance trigger values; Performing trigger accumulation verification within a time window on the tolerance trigger value; If the cumulative verification of triggers within any time window meets the preset window threshold, a window warning message is generated; The maintenance and management of the single equipment is performed according to the window warning information.

2. The method according to claim 1, characterized in that The method further comprises: Based on the construction data, the basic construction of the abnormal decision channel is completed, wherein the construction process includes calling the grid attribute data to construct the abnormal decision sub-network of the single device; After the basic construction is completed, the abnormal decision sub-network feedback link under each working mode is carried out through the linkage monitoring node to complete the construction compensation of the abnormal decision channel; After the timing compensation and the timing data set are sent to the abnormal decision channel, after each abnormal decision sub-network is compensated according to the timing compensation, the output data of the abnormal decision sub-network is linked and transmitted through the feedback link as abnormal detection compensation data linked to the abnormal decision sub-network; The anomaly detection of all abnormal decision sub-networks is summarized through the summary sub-network, and the node results of the summary node are verified; Complete anomaly determination based on verification and summary results.

3. The method according to claim 2, characterized in that The method further comprises: When the link abnormality decision sub-network receives the corresponding output data, the output data is used as input determination data, and the collected output data is used as result determination data to execute the processing determination of the current single device; Record the processing judgment results and generate abnormal data sets for individual devices; Performing abnormality determination of a single device according to the abnormal data set, wherein the abnormality determination includes point discrete determination and trend abnormality determination; Complete the abnormal detection summary based on the judgment results.

4. The method according to claim 3, characterized in that The method further comprises: Collect environmental data of the distribution network through the environmental sensor group and establish an environmental data set; Initializing the equipment environment impact of the distribution network frame according to the construction data; Expanding the fluctuation tolerance of the environmental impact through the device environmental impact initialization result and the environmental data set; Generate a tolerance threshold for anomaly detection based on the fluctuation tolerance expansion result; The tolerance judgment threshold is used to perform abnormal judgment compensation on the abnormal decision channel.

5. The method according to claim 1, characterized in that The method further comprises: Extracting equipment features according to the grid working data to obtain equipment feature extraction results; Establishing a sensitive trigger factor of a device based on the device feature extraction result; The sensitive trigger factor is used to construct a constraint unit, and the constraint unit is coupled to the abnormal decision channel to complete the abnormality determination.

6. The method according to claim 1, characterized in that The method further comprises: Make maintenance judgments on reported equipment abnormalities and record maintenance judgment data; Performing time series clustering on the maintenance determination data, and performing common deviation determination on the time series clustering results; Compensation parameters are generated through the common deviation determination results, and channel feedback optimization of the abnormal decision channel is performed based on the compensation parameters.

7. A distribution network automatic fault detection system, characterized in that: The system for performing the steps of the method according to any one of claims 1 to 6 comprises: A grid communication interaction module, the grid communication interaction module is used to establish grid communication with the distribution network grid, and complete information interaction based on the grid communication to obtain construction data of the distribution network grid, wherein the construction data includes grid structure data, grid attribute data, and grid working data; A linkage monitoring node configuration module, the linkage monitoring node configuration module is used to configure the linkage monitoring node mapped with the working mode, wherein the linkage monitoring node includes the monitoring node of the single device and the summary node of the result supervision; A working mode acquisition module, wherein the working mode acquisition module is used to acquire the working mode of the distribution network in real time when the distribution network is working, and locate the linkage monitoring node according to the working mode; A timing compensation module, which is used to read the timing input data of the distribution network grid and perform timing compensation on the located linkage monitoring nodes; A data monitoring module, the data monitoring module is used to perform data monitoring of the single device and output results of the distribution network frame according to the linkage monitoring node, and generate a time series data set; An abnormality determination module, the abnormality determination module is used to send the timing compensation and the timing data set to an abnormality decision channel to perform abnormality determination, and the abnormality decision channel is constructed by the construction data; The device abnormality reporting module is used to report the abnormality of the equipment of the distribution network according to the abnormality determination result.

Images