Power grid fault diagnosis system and method based on big data

Abstract

The invention discloses a power grid fault diagnosis system and method based on big data, belongs to the technical field of fault diagnosis, and solves the technical problems that in the prior art, only a line section where a fault occurs can be diagnosed, the type of the fault cannot be judged, and subsequent fault processing is not facilitated. The method comprises the following steps: constructing a fault anomaly library through historical data, and analyzing real-time data based on the fault anomaly library to obtain a preliminary diagnosis result; technicians can prepare in advance; integrating the real-time data into an index feature sequence, calling a fault diagnosis model, and inputting the index feature sequence into the fault diagnosis model to obtain a fault tag; matching a corresponding fault type based on the fault tag; the accuracy and the reliability of diagnosis can be improved; fault position information is obtained according to the data sensor and fault type analysis, and a technician closest to the fault position is dispatched to process the fault; and the fault can be timely processed, and greater influence caused by untimely processing can be avoided.

Description

Technical Field

[0001] The present invention belongs to the field of fault diagnosis, and relates to power grid fault diagnosis technology, and specifically to a power grid fault diagnosis system and method based on big data. Background Art

[0002] Grid failures may damage power equipment, causing power supply interruptions or deterioration in power quality; grid failures may cause excessive current or overheating, posing a fire hazard; grid failures may cause large-scale power outages, causing inconvenience to users; therefore, grid fault diagnosis is very important. Through grid fault diagnosis, problems with deteriorating power supply quality can be discovered and resolved in a timely manner, and faults can be discovered and repaired in a timely manner to ensure the continuity and stability of power supply; through the diagnosis of grid faults, the operating status of the power system can be better understood, providing a basis for optimizing the operation of the power system, helping to promote the development of the power industry and improve the reliability and safety of the power system.

[0003] The prior art (invention patent with application number 2016108639120) discloses a distribution network fault diagnosis method, which includes: constructing a full model of the power grid; obtaining fault information of the feeder, and locating the feeder where the fault is located according to the fault information and the full model of the power grid; obtaining fault indication information of the fault indicator, and locating the first fault interval where the fault is located according to the fault indication information and the full model of the power grid; obtaining the distribution transformer power outage event of the distribution transformer, and locating the first tripping device where the fault is located according to the distribution transformer power outage time and the full model of the power grid; obtaining the measurement sudden drop information of the outgoing line switch, and locating the second fault interval and the second tripping device where the fault is located according to the measurement sudden drop information and the full model of the power grid; through the fault indication information, the distribution transformer power outage time and the measurement sudden drop information, combined with the full model of the power grid, the line section where the fault occurs can be accurately obtained, which effectively improves the fault diagnosis efficiency; however, the prior art can only diagnose the line section where the fault occurs, and cannot determine the type of the fault, which is not conducive to subsequent fault processing. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a power grid fault diagnosis system and method based on big data, which is used to solve the technical problem that the prior art can only diagnose the line segment where the fault occurs, but cannot determine the type of fault, which is not conducive to subsequent fault processing.

[0005] To achieve the above-mentioned object, the present invention provides a power grid fault diagnosis system based on big data, comprising: a fault diagnosis module and a data acquisition module and a fault processing module connected thereto; A data acquisition module collects real-time data through a number of data sensors connected thereto, and transmits the real-time data to a fault diagnosis module; and obtains historical data through an intelligent terminal connected thereto, and transmits the historical data to the fault diagnosis module; wherein the real-time data includes real-time voltage, real-time current, real-time temperature, and the status of the protector and the circuit breaker; and the historical data includes historical fault data, historical abnormal data, and historical fault types; The fault diagnosis module divides the power grid into regions according to the power grid distribution map to obtain the power grid region i; builds a fault anomaly library based on historical data; determines whether the real-time data is abnormal based on the historical data; if yes, obtains a preliminary diagnosis result based on the fault anomaly library analysis; if no, continuously monitors the real-time data; wherein i=1,2,…,n, n is a positive integer; and, The fault diagnosis model is retrieved to analyze the fault of the power grid and obtain the fault type; the fault location information is analyzed according to the fault type and real-time data; an early warning signal is generated based on the fault location information and the fault type, and the early warning signal is transmitted to the fault processing module; wherein the fault diagnosis model is constructed based on the artificial intelligence model; The fault processing module extracts the fault location information and fault type based on the early warning signal; dispatches the technician closest to the fault location to process the fault information based on the fault location information; encrypts and stores the fault type and processing measures in the smart terminal; and determines whether the preliminary diagnosis result is consistent with the fault type; if yes, continuously monitors the fault; if not, updates the fault anomaly library.

[0006] Preferably, the fault diagnosis module communicates and / or is electrically connected to the data acquisition module and the fault processing module respectively; the data acquisition module communicates and / or is electrically connected to the data sensor and the smart terminal respectively; wherein the data sensor includes a voltage sensor, a current sensor and a temperature sensor; and the smart terminal includes a mobile phone or a computer.

[0007] Preferably, the construction of a fault anomaly library based on historical data includes: Retrieve historical fault data and historical abnormal data, and extract historical fault types; According to the historical fault data, the corresponding historical abnormal data and historical fault types are matched to generate a fault abnormality library.

[0008] The present invention constructs a fault anomaly library based on historical data, and can perform preliminary diagnosis on the abnormal data later to obtain preliminary diagnosis information; according to the preliminary diagnosis information, technical personnel can be notified in advance of power grid failure, which is conducive to technical personnel making preparations in advance.

[0009] Preferably, judging whether the real-time data is abnormal based on the historical data includes: Extract historical fault data and historical abnormal data from historical data; extract real-time data; Determine whether the real-time data is consistent with the historical fault data or historical abnormal data; if so, obtain preliminary diagnosis results based on the fault anomaly library analysis; if not, continue to monitor the real-time data.

[0010] The present invention judges the real-time data according to the historical fault data and the historical abnormal data; when the real-time data is the same as the historical fault data or the historical abnormal data, it is determined that an abnormal phenomenon occurs in the power grid; the judgment based on the historical data has certain data support, and the abnormal situation can be discovered in time to reduce the impact.

[0011] Preferably, the preliminary diagnosis result obtained based on the fault anomaly library analysis includes: Retrieve historical abnormal data and historical fault data from historical data; extract real-time voltage, real-time current and real-time temperature from real-time data; match corresponding historical data based on real-time data; The fault type corresponding to the historical data is queried according to the fault anomaly library, and the real-time data is matched with the fault type to obtain the preliminary diagnosis result.

[0012] The present invention obtains preliminary diagnosis results based on the fault anomaly library analysis, which enables technical personnel to prepare in advance. After the fault type is diagnosed, the fault can be handled in time, which is conducive to reducing the impact caused by the fault.

[0013] Preferably, the retrieving the fault diagnosis model to analyze the fault of the power grid includes: Integrate real-time voltage data, real-time current data and real-time temperature data into a real-time feature sequence; integrate the status of the protector and circuit breaker into an equipment feature sequence; integrate the real-time feature sequence and the equipment feature sequence into an index feature sequence; The fault diagnosis model is called, and the indicator feature sequence is input into the fault diagnosis model to obtain the fault label; the corresponding fault type is matched based on the fault label; wherein the fault label is set to a positive integer; the fault diagnosis model is constructed based on the artificial intelligence model.

[0014] Preferably, the fault diagnosis model is constructed based on an artificial intelligence model, including: Acquire a standard data set; wherein the standard data set includes standard input data consistent with the content attributes of the indicator feature sequence, and standard output data consistent with the content attributes of the fault label; Divide the standard data set into a training set, a validation set, and a test set; use the training set to train the artificial intelligence model; use the validation set to optimize the parameters of the artificial intelligence model; use the test set to evaluate the performance indicators of the artificial intelligence model to determine whether the performance indicators of the artificial intelligence model are greater than the indicator threshold; if yes, mark the trained artificial intelligence model as a fault diagnosis model; if not, retrain the artificial intelligence model; wherein the artificial intelligence model is a BP neural network model or a convolutional neural network model.

[0015] The present invention constructs a fault diagnosis model by training an artificial intelligence model, and uses the fault diagnosis model to analyze real-time data to obtain a fault label; this is beneficial to reducing the time of judgment, improving the efficiency of judgment, and increasing the accuracy and reliability of judgment.

[0016] Preferably, analyzing the fault location information according to the fault type and real-time data includes: Retrieve real-time data and matching data sensors; determine the power grid area i where the fault occurs based on the data sensors; Extract the fault type and determine whether the equipment is faulty based on the fault type; if yes, check the equipment in the power grid area i to get the faulty equipment; if not, check the wires in the power grid area i to get the faulty line section; conduct risk assessment on the faulty equipment and faulty line section, and replace the faulty equipment and faulty line section.

[0017] The present invention checks the equipment and lines in the power grid area according to the fault type and the location of the data sensor, which is conducive to accurately and quickly finding the fault location, handling the fault in time, and avoiding major losses caused by prolonged fault time.

[0018] It is worth noting that the risk assessment of faulty equipment and faulty power line sections includes: Obtaining evaluation information of the faulty equipment or faulty power line section; wherein the evaluation information includes: fault severity, number of faults, and service life; The fault severity, fault frequency and service life are marked as YZ, GC and SN respectively; the risk factor of the faulty equipment or faulty power line section is calculated by the formula FX=α×YZ+β×GC+γ×ln(SN+1); wherein α, β and γ are proportional coefficients greater than 0; When the risk factor is greater than the set coefficient threshold, the faulty equipment or faulty power line section is replaced; otherwise, the risk assessment of the faulty equipment or faulty power line section is continuously performed.

[0019] The present invention also provides a power grid fault diagnosis system and method based on big data, including: Step S1: collect real-time data through a number of data sensors; obtain historical data through an intelligent terminal; wherein the real-time data includes real-time voltage, real-time current, real-time temperature, and the status of the protector and circuit breaker; the historical data includes historical fault data, historical abnormal data, and historical fault types; Step S2: Divide the power grid into regions according to the power grid distribution map to obtain power grid region i; construct a fault anomaly library based on historical data; wherein i=1,2,…,n, and n is a positive integer; Step S3: Determine whether the real-time data is abnormal based on the historical data; if yes, obtain a preliminary diagnosis result based on the fault anomaly library analysis and proceed to step S4; if no, continue to monitor the real-time data and proceed to step S1; Step S4: Retrieving the fault diagnosis model to analyze the fault of the power grid and obtain the fault type; analyzing the fault location information according to the fault type and real-time data; and generating an early warning signal based on the fault location information and the fault type; Step S5: Extract fault location information and fault type based on the early warning signal; dispatch the technician closest to the fault location to process the fault information according to the fault location information; encrypt and store the fault type and processing measures in the smart terminal; determine whether the preliminary diagnosis result is consistent with the fault type, if so, continue to monitor the fault; otherwise, update the fault anomaly library.

[0020] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention collects real-time data and historical data through a data acquisition module, integrates the historical data, and obtains a fault anomaly library; when the real-time data is consistent with the historical fault data or the historical anomaly data, the fault type corresponding to the historical data is queried according to the fault anomaly library, and the real-time data is matched with the fault type to obtain a preliminary diagnosis result; the judgment based on the historical data has certain data support, and abnormal situations can be discovered in time, which is conducive to technical personnel to prepare in advance. When a fault occurs, the fault can be handled in time to reduce the impact caused by the power grid fault.

[0021] 2. Integrate real-time voltage data, real-time current data and real-time temperature data into a real-time feature sequence; integrate the status of the protector and circuit breaker into an equipment feature sequence; integrate the real-time feature sequence and the equipment feature sequence into an indicator feature sequence; call the fault diagnosis model, input the indicator feature sequence into the fault diagnosis model, and obtain the fault label; match the corresponding fault type based on the fault label; extract the fault location information and fault type based on the early warning signal; dispatch the technician closest to the fault location to process the fault information based on the fault location information; encrypt and store the fault type and treatment measures in the smart terminal; build a fault diagnosis model by training the artificial intelligence model, and use the fault diagnosis model to analyze the real-time data to obtain the fault label; it is helpful to reduce the judgment time, improve the judgment efficiency, and increase the accuracy and reliability of the judgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 It is a schematic diagram of the overall framework of the system of the present invention; Figure 2 It is a schematic diagram of the specific steps of the system of the present invention; Figure 3 A schematic flow chart of the steps of preliminary diagnosis of the system of the present invention; Figure 4 A schematic diagram of specific process steps for fault diagnosis of the system of the present invention; Figure 5 Schematic diagram of the specific steps of the method of the present invention. DETAILED DESCRIPTION

[0024] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] Example 1: Please refer to Figure 1-Figure 2 , a power grid fault diagnosis system based on big data, comprising: a data acquisition module, a fault diagnosis module and a fault processing module; the fault diagnosis module is connected with the data acquisition module and the fault processing module; A data acquisition module collects real-time data through a number of data sensors connected thereto, and transmits the real-time data to a fault diagnosis module; and obtains historical data through an intelligent terminal connected thereto, and transmits the historical data to the fault diagnosis module; wherein the real-time data includes real-time voltage, real-time current, real-time temperature, and the status of the protector and the circuit breaker; and the historical data includes historical fault data, historical abnormal data, and historical fault types; The fault diagnosis module divides the power grid into regions according to the power grid distribution map to obtain the power grid region i; builds a fault anomaly library based on historical data; determines whether the real-time data is abnormal based on the historical data; if yes, obtains a preliminary diagnosis result based on the fault anomaly library analysis; if no, continuously monitors the real-time data; wherein i=1,2,…,n, n is a positive integer; and, The fault diagnosis model is retrieved to analyze the fault of the power grid and obtain the fault type; the fault location information is analyzed according to the fault type and real-time data; an early warning signal is generated based on the fault location information and the fault type, and the early warning signal is transmitted to the fault processing module; wherein the fault diagnosis model is constructed based on the artificial intelligence model; The fault processing module extracts the fault location information and fault type based on the early warning signal; dispatches the technician closest to the fault location to process the fault information based on the fault location information; encrypts and stores the fault type and processing measures in the smart terminal; and determines whether the preliminary diagnosis result is consistent with the fault type; if yes, continuously monitors the fault; if not, updates the fault anomaly library.

[0026] For example, a fault diagnosis is now performed on a power grid in a certain place. The total area of ​​the region is obtained, and the total area of ​​the region is evenly divided according to a specified area threshold to obtain a number of regions i. Various data of the power grid in the region are monitored and fault diagnosis is performed based on the several regions i.

[0027] See also Figure 3 , retrieve historical fault data and historical abnormal data, and extract historical fault types; match the corresponding historical abnormal data and historical fault types according to the historical fault data to generate a fault abnormality library; determine whether the real-time data is consistent with the historical fault data or historical abnormal data; if yes, obtain a preliminary diagnosis result based on the fault abnormality library analysis; if not, continue to monitor the real-time data; retrieve the historical abnormal data and historical fault data in the historical data; extract the real-time voltage, real-time current and real-time temperature in the real-time data; match the corresponding historical data based on the real-time data; query the fault type corresponding to the historical data according to the fault abnormality library; match the real-time data with the fault type to obtain a preliminary diagnosis result.

[0028] For example, a fault diagnosis is now being conducted on a power grid in a certain place. The historical fault data and historical abnormal data of the place are retrieved, and the historical fault types of the place are extracted. The corresponding historical fault types are matched according to the historical fault data and historical abnormal data to generate a fault abnormality library.

[0029] By monitoring the real-time data of the local power grid, it was found that the data in areas 3 and 7 were abnormal. The real-time data of areas 3 and 7 at the time of the abnormalities were extracted, and the real-time data of areas 3 and 7 were matched with the fault anomaly library. The preliminary diagnosis result was that area 3 might fail, but area 7 would not fail.

[0030] See also Figure 4 , integrate the real-time voltage data, real-time current data and real-time temperature data into a real-time feature sequence; integrate the status of the protector and the circuit breaker into an equipment feature sequence; integrate the real-time feature sequence and the equipment feature sequence into an indicator feature sequence; call the fault diagnosis model, input the indicator feature sequence into the fault diagnosis model, and obtain the fault label; match the corresponding fault type based on the fault label; retrieve the real-time data and the matching data sensor; determine the power grid area i where the fault occurs based on the data sensor; extract the fault type, and determine whether the equipment fails based on the fault type; if yes, check the equipment in the power grid area i to obtain the faulty equipment; if not, check the wires in the power grid area i to obtain the faulty wire section.

[0031] For example, a fault diagnosis is conducted on a power grid in a certain place. It is found that the data in areas 3 and 7 are abnormal. The data in areas 3 and 7 are respectively integrated into two sets of indicator feature sequences. The two sets of indicator feature sequences are input into the fault diagnosis model, and the fault label of area 3 is 2, and the fault label of area 7 is 0. According to the fault label matching corresponding fault types, the fault type of area 3 is a transformer failure, and there is no failure in area 7. The transformer in area 3 is inspected and repaired.

[0032] In some other preferred embodiments, risk assessment is performed on faulty equipment and faulty power line sections, including: Obtaining evaluation information of the faulty equipment or faulty power line section; wherein the evaluation information includes: fault severity, number of faults, and service life; The fault severity, fault frequency and service life are marked as YZ, GC and SN respectively; the risk factor of the faulty equipment or faulty power line section is calculated by the formula FX=α×YZ+β×GC+γ×ln(SN+1); wherein α, β and γ are proportional coefficients greater than 0; When the risk factor is greater than the set coefficient threshold, the faulty equipment or faulty power line section is replaced; otherwise, the risk assessment of the faulty equipment or faulty power line section is continuously performed.

[0033] It should be noted that the severity of the fault is determined according to the fault type, and the severity is divided into 10 levels in total; level 1 is the lowest severity and level 10 is the highest severity; the coefficient threshold and α, β and γ are set manually, and the value range is [1,100]; different proportional coefficient settings have different coefficient thresholds; risk assessment of the equipment is performed only when a fault occurs.

[0034] For example, suppose there are two faulty devices. The risks of the two faulty devices are evaluated separately. α=1, β=1, γ=10. The specific steps are as follows: Faulty equipment 1: fault severity YZ=5, fault count GC=3, service life SN=1.5; the risk factor is FX=12.05; Faulty equipment 2: Fault severity YZ=3, fault count GC=1, service life SN=1.5; the risk factor is FX=8.05; If the coefficient threshold is set to 10, then if the risk coefficient of faulty device 1 is greater than the coefficient threshold, the faulty device 1 is replaced; if the risk coefficient of faulty device 2 is less than the coefficient threshold, the fault of faulty device 2 is processed, and risk assessments are continuously performed on other faulty devices or faulty power line sections.

[0035] Example 2: Please refer to Figure 5 , a power grid fault diagnosis method based on big data, comprising the following steps: Step S1: collect real-time data through a number of data sensors; obtain historical data through an intelligent terminal; wherein the real-time data includes real-time voltage, real-time current, real-time temperature, and the status of the protector and circuit breaker; the historical data includes historical fault data, historical abnormal data, and historical fault types; Step S2: Divide the power grid into regions according to the power grid distribution map to obtain power grid region i; construct a fault anomaly library based on historical data; wherein i=1,2,…,n, and n is a positive integer; Step S3: Determine whether the real-time data is abnormal based on the historical data; if yes, obtain a preliminary diagnosis result based on the fault anomaly library analysis and proceed to step S4; if no, continue to monitor the real-time data and proceed to step S1; Step S4: Retrieving the fault diagnosis model to analyze the fault of the power grid and obtain the fault type; analyzing the fault location information according to the fault type and real-time data; and generating an early warning signal based on the fault location information and the fault type; Step S5: Extract fault location information and fault type based on the early warning signal; dispatch the technician closest to the fault location to process the fault information according to the fault location information; encrypt and store the fault type and processing measures in the smart terminal; determine whether the preliminary diagnosis result is consistent with the fault type, if so, continue to monitor the fault; otherwise, update the fault anomaly library.

[0036] Working principle of the present invention: The present invention collects real-time data through data sensors; collects historical data through intelligent terminals; divides the power grid into regions according to the power grid distribution map to obtain power grid region i; generates a fault anomaly library based on historical data and historical fault types; when real-time data is consistent with historical fault data or historical anomaly data; matches the real-time data with the corresponding historical data; queries the fault type corresponding to the historical data according to the fault anomaly library, matches the real-time data with the fault type, and obtains a preliminary diagnosis result, otherwise the real-time data is continuously monitored; Integrate real-time voltage data, real-time current data and real-time temperature data into a real-time feature sequence; integrate the status of the protector and the circuit breaker into an equipment feature sequence; integrate the real-time feature sequence and the equipment feature sequence into an index feature sequence; call the fault diagnosis model, input the index feature sequence into the fault diagnosis model, and obtain the fault label; match the corresponding fault type based on the fault label; Extract fault location information and fault type based on the early warning signal; dispatch the technician closest to the fault location to process the fault information based on the fault location information; encrypt and store the fault type and treatment measures in the smart terminal; update the fault anomaly library based on the fault type and preliminary diagnosis results.

[0037] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A power grid fault diagnosis system based on big data, comprising: The fault diagnosis module and the data acquisition module and fault processing module connected thereto are characterized in that: A data acquisition module collects real-time data through a number of data sensors connected thereto, and transmits the real-time data to a fault diagnosis module; and obtains historical data through an intelligent terminal connected thereto, and transmits the historical data to the fault diagnosis module; wherein the real-time data includes real-time voltage, real-time current, real-time temperature, and the status of the protector and the circuit breaker; and the historical data includes historical fault data, historical abnormal data, and historical fault types; The fault diagnosis module divides the power grid into regions according to the power grid distribution map to obtain the power grid region i; builds a fault anomaly library based on historical data; determines whether the real-time data is abnormal based on the historical data; if yes, obtains a preliminary diagnosis result based on the fault anomaly library analysis; if no, continuously monitors the real-time data; wherein i=1,2,…,n, n is a positive integer; and, The fault diagnosis model is retrieved to analyze the fault of the power grid and obtain the fault type; the fault location information is analyzed according to the fault type and real-time data; an early warning signal is generated based on the fault location information and the fault type, and the early warning signal is transmitted to the fault processing module; wherein the fault diagnosis model is constructed based on the artificial intelligence model; The fault processing module extracts the fault location information and fault type based on the early warning signal; dispatches the technician closest to the fault location to process the fault information based on the fault location information; encrypts and stores the fault type and processing measures in the smart terminal; and determines whether the preliminary diagnosis result is consistent with the fault type; if yes, continuously monitors the fault; if not, updates the fault anomaly library.

2. A power grid fault diagnosis system based on big data according to claim 1, characterized in that: The fault diagnosis module communicates and / or is electrically connected to the data acquisition module and the fault processing module respectively; the data acquisition module communicates and / or is electrically connected to the data sensor and the smart terminal respectively; wherein the data sensor includes a voltage sensor, a current sensor and a temperature sensor; the smart terminal includes a mobile phone or a computer.

3. A power grid fault diagnosis system based on big data according to claim 1, characterized in that: The method of constructing a fault anomaly library based on historical data includes: Retrieve historical fault data and historical abnormal data, and extract historical fault types; According to the historical fault data, the corresponding historical abnormal data and historical fault types are matched to generate a fault abnormality library.

4. A power grid fault diagnosis system based on big data according to claim 3, characterized in that: The determining whether the real-time data is abnormal based on the historical data includes: Extract historical fault data and historical abnormal data from historical data; extract real-time data; Determine whether the real-time data is consistent with the historical fault data or historical abnormal data; if so, obtain preliminary diagnosis results based on the fault anomaly library analysis; if not, continue to monitor the real-time data.

5. A power grid fault diagnosis system based on big data according to claim 4, characterized in that: The preliminary diagnosis results obtained based on the fault anomaly library analysis include: Retrieve historical abnormal data and historical fault data from historical data; extract real-time voltage, real-time current and real-time temperature from real-time data; match corresponding historical data based on real-time data; The fault type corresponding to the historical data is queried according to the fault anomaly library, and the real-time data is matched with the fault type to obtain the preliminary diagnosis result.

6. The power grid fault diagnosis system based on big data according to claim 1, characterized in that: The calling of the fault diagnosis model to analyze the fault of the power grid includes: Integrate real-time voltage data, real-time current data and real-time temperature data into a real-time feature sequence; integrate the status of the protector and circuit breaker into an equipment feature sequence; integrate the real-time feature sequence and the equipment feature sequence into an index feature sequence; The fault diagnosis model is called, and the indicator feature sequence is input into the fault diagnosis model to obtain a fault label; the corresponding fault type is matched based on the fault label; wherein the fault label is set to a positive integer.

7. A power grid fault diagnosis system based on big data according to claim 6, characterized in that: The fault diagnosis model is constructed based on an artificial intelligence model and includes: Acquire a standard data set; wherein the standard data set includes standard input data consistent with the content attributes of the indicator feature sequence, and standard output data consistent with the content attributes of the fault label; Divide the standard data set into a training set, a validation set, and a test set; use the training set to train the artificial intelligence model; use the validation set to optimize the parameters of the artificial intelligence model; use the test set to evaluate the performance indicators of the artificial intelligence model to determine whether the performance indicators of the artificial intelligence model are greater than the indicator threshold; if yes, mark the trained artificial intelligence model as a fault diagnosis model; if not, retrain the artificial intelligence model; wherein the artificial intelligence model is a BP neural network model or a convolutional neural network model.

8. The power grid fault diagnosis system based on big data according to claim 1, characterized in that: Analyzing the fault location information according to the fault type and real-time data includes: Retrieve real-time data and matching data sensors; determine the power grid area i where the fault occurs based on the data sensors; Extract the fault type and determine whether the equipment is faulty based on the fault type; if yes, check the equipment in the power grid area i to get the faulty equipment; if not, check the wires in the power grid area i to get the faulty line section; conduct risk assessment on the faulty equipment and faulty line section, and replace the faulty equipment and faulty line section.

9. A power grid fault diagnosis system based on big data according to claim 8, characterized in that: The risk assessment of faulty equipment and faulty power line sections includes: Obtaining evaluation information of the faulty equipment or faulty power line section; wherein the evaluation information includes: fault severity, number of faults, and service life; The fault severity, fault frequency and service life are marked as YZ, GC and SN respectively; the risk factor of the faulty equipment or faulty power line section is calculated by the formula FX=α×YZ+β×GC+γ×ln(SN+1); wherein α, β and γ are proportional coefficients greater than 0; When the risk factor is greater than the set coefficient threshold, the faulty equipment or faulty power line section is replaced; otherwise, the risk assessment of the faulty equipment or faulty power line section is continuously performed.

10. A power grid fault diagnosis method based on big data, applied to a power grid fault diagnosis system based on big data as claimed in any one of claims 1 to 9, characterized in that: The following steps are involved: Step S1: collect real-time data through a number of data sensors; obtain historical data through an intelligent terminal; wherein the real-time data includes real-time voltage, real-time current, real-time temperature, and the status of the protector and circuit breaker; the historical data includes historical fault data, historical abnormal data, and historical fault types; Step S2: Divide the power grid into regions according to the power grid distribution map to obtain power grid region i; construct a fault anomaly library based on historical data; wherein i=1,2,…,n, and n is a positive integer; Step S3: Determine whether the real-time data is abnormal based on the historical data; if yes, obtain a preliminary diagnosis result based on the fault anomaly library analysis and proceed to step S4; if no, continue to monitor the real-time data and proceed to step S1; Step S4: Retrieving the fault diagnosis model to analyze the fault of the power grid and obtain the fault type; analyzing the fault location information according to the fault type and real-time data; and generating an early warning signal based on the fault location information and the fault type; Step S5: Extract fault location information and fault type based on the early warning signal; dispatch the technician closest to the fault location to process the fault information according to the fault location information; encrypt and store the fault type and processing measures in the smart terminal; determine whether the preliminary diagnosis result is consistent with the fault type, if so, continue to monitor the fault; otherwise, update the fault anomaly library.

Images