Smart grid fault detection system and method based on multi-source heterogeneous data

Abstract

The application discloses a smart grid fault detection system and method based on multi-source heterogeneous data, relates to the technical field of grid fault detection, and comprises the following steps: evaluating the fault detection accuracy of monitoring parameters in fault range data on marked grid nodes; analyzing the node approximation degree between different grid nodes, evaluating the abnormal state of fault range data on grid nodes for fault detection; evaluating the fault capture accuracy of monitoring parameters in fault range data on target grid nodes; performing abnormal identification on grid nodes in the grid to obtain abnormal grid nodes, obtaining a grid topology graph of the grid, and obtaining abnormal area data in the grid according to the distribution situation of abnormal grid nodes in the grid topology graph; performing fault detection on grid equipment to obtain abnormal grid equipment, and performing maintenance on abnormal grid equipment in the grid, so that the fast and accurate identification of faults in the grid is realized, and the efficiency and accuracy of grid fault detection are greatly improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power grid fault detection, in particular to an intelligent power grid fault detection system and method based on multi-source heterogeneous data. BACKGROUND

[0002] In the operation of the power grid, data of different types, formats and structures will be generated in different sources and different collection devices. Using multi-source heterogeneous data in power grid fault detection can effectively integrate the data, thereby solving the problems caused by a single data source. Through the fusion of multi-source data, the information of each party can be verified, greatly improving the accuracy of power grid fault positioning and diagnosis, avoiding misjudgment. At the same time, through multi-source heterogeneous data, different data of the power grid can be fused to construct a comprehensive portrait of the power grid fault, so that the power grid fault can be quickly and comprehensively mastered.

[0003] At present, the method for detecting faults in the power grid mainly monitors the equipment in the power grid. When the parameters related to the fault in the equipment are not within the preset range, the equipment is determined to be abnormal, thereby obtaining the abnormal equipment in the power grid. However, the power grid structure is complex, and the working environments of different equipment in the power grid are different. It is difficult to accurately grasp the state of the equipment using the preset threshold. At the same time, the power grid is connected through transmission lines, so a single equipment failure may cause several related equipment to malfunction. The traditional method cannot identify the location of the power grid failure. In addition, the power grid needs to transport electricity and is easily affected by natural disasters, so the integrity of the appearance needs to be checked. However, most of the equipment in the power grid is far apart, and some are even in places with poor transportation. If the abnormal area of the power grid is manually checked, it will undoubtedly consume a lot of manpower and resources, and it is also difficult to accurately identify the state of the equipment. SUMMARY

[0004] The purpose of the present application is to provide an intelligent power grid fault detection system and method based on multi-source heterogeneous data to solve the problems in the prior art.

[0005] To achieve the above purpose, the present application provides the following technical scheme: an intelligent power grid fault detection method based on multi-source heterogeneous data, the method comprising:

[0006] Step S1: obtaining historical fault records and fault range data of a marked power grid node in the power grid, evaluating the fault detection accuracy of the monitoring parameters in the fault range data on the marked power grid node, and obtaining marked fault range data;

[0007] Step S2: Obtain fault range data and node data of power grid nodes in the power grid, analyze the node approximation degree between different power grid nodes, obtain an approximate power grid node set of the power grid nodes, obtain a marked power grid node from the approximate power grid node set, and mark it as an approximate marked power grid node, evaluate the abnormal state of the fault range data for fault detection of the power grid nodes, and obtain a target power grid node;

[0008] Step S3: Obtain the marked fault range data in the approximate marked power grid node of the target power grid node, evaluate the fault capture accuracy of the monitoring parameters in the fault range data on the target power grid node, and obtain target fault range data of the target power grid node;

[0009] Step S4: According to the target fault range data and the marked fault range data, the fault range data of the power grid nodes in the power grid is optimized and adjusted, the abnormal identification of the power grid nodes in the power grid is performed, the abnormal power grid nodes are obtained, the power grid topology graph of the power grid is obtained, and the abnormal area data in the power grid is obtained according to the distribution situation of the abnormal power grid nodes in the power grid topology graph;

[0010] Step S5: According to the abnormal area data, the unmanned aerial vehicle is dispatched to shoot the power grid equipment in the power grid, the equipment image is obtained, the fault detection of the power grid equipment is performed, the abnormal power grid equipment is obtained, and the abnormal power grid equipment in the power grid is maintained.

[0011] Further, step S1 includes:

[0012] Step S11: Obtain a power grid node that has occurred a fault in the power grid, mark it as a marked power grid node, obtain fault range data of the marked power grid node, and mark a preset range of each monitoring parameter in the marked power grid node from the fault range data;

[0013] Obtain each historical fault record of the marked power grid node, and obtain the data of each monitoring parameter in the marked power grid node from the historical fault record;

[0014] Step S12: When the maximum value or the minimum value of a certain monitoring parameter in a certain historical fault record is outside the preset fault range data range, the platform issues a fault warning to the marked power grid node, and subsequent inspection of the marked power grid node does not detect a fault, then the certain historical fault record is recorded as a first historical fault record;

[0015] Obtain each first historical fault record of the marked power grid node, and when the maximum value or the minimum value of the a-th monitoring parameter in the first historical fault record is outside the fault range data range, determine that the a-th monitoring parameter has a fault detection abnormality in the marked power grid node;

[0016] Obtain the total number E of the a-th monitoring parameter that has a fault abnormality in the marked power grid node sumObtain the total number γ of each first historical fault record. sum Calculate the outlier value Q of the monitoring parameter a. a =E sum / γ sum When detecting outlier Q a If the value exceeds the preset abnormal detection threshold, then the monitoring parameter a is determined to be inaccurate for detecting faults at marked power grid nodes.

[0017] Obtain the range constructed by the minimum and maximum values ​​of the a-th monitoring parameter in each first historical fault record, and denot it as the marked range of the a-th monitoring parameter in the marked power grid node;

[0018] Step S13: Remove each first historical fault record from the other historical fault records to obtain the second historical fault record that marks the power grid node;

[0019] Calculate the range distance value S of the βth monitoring parameter in the second historical fault record. β :

[0020] ,

[0021] Among them, H (max,β) and H (min,β) The range H of the βth monitoring parameter in the fault range data is respectively. β The maximum and minimum values ​​of H; △ For the range H β The average of the maximum and minimum values; H' (max,β) and H' (min,β) These are the maximum and minimum values ​​of the βth monitoring parameter in the second historical fault record, respectively;

[0022] When the range distance value S β If the distance is less than the preset range threshold, it is determined that the βth monitoring parameter has a distance anomaly in the second historical fault record.

[0023] The range anomaly value B of the β-th monitoring parameter is obtained by comparing the total number of second historical fault records showing distance anomalies in the marked power grid node with the total number of second historical fault records for the marked power grid node. β ;

[0024] When the range outlier B β If the value exceeds the preset abnormal threshold, it is determined that the β-th monitoring parameter is inaccurate in detecting faults in the marked power grid node. The range constructed by the maximum and minimum values ​​of the β-th monitoring parameter within the fault range data is obtained from several second historical fault records and recorded as the marked range of the β-th monitoring parameter in the marked power grid node.

[0025] Step S14: Obtain the marked range of the number of monitoring parameters in the marked power grid node which is determined to be inaccurate in fault detection, and replace the range of the number of monitoring parameters in the fault range data of the marked power grid node to obtain the marked fault range data of the marked power grid node.

[0026] Further, step S2 includes:

[0027] Step S21: Obtain the power grid node fault range data and node data, and obtain the data of each node index from the node data;

[0028] Step S22: Analyze the node approximation degree between different power grid nodes to obtain the approximate power grid node set of the power grid node, and the specific analysis process is:

[0029] Obtain the range L d d,min d,max e d e d,min e d,max of the dth monitoring parameter of the power grid node and the e th power grid node respectively, where L d,min d,max are the minimum threshold and maximum threshold of the dth monitoring parameter in the fault range data of the power grid node, and L e d,min e d,max are the minimum threshold and maximum threshold of the dth monitoring parameter in the fault range data of the e th power grid node;

[0030] When there is a certain value in the range L d e d , obtain the intersection Y d e d between L e d e d,min e d,max , where Y e d,min =max{L d,min , L d,max}, Y e d,max =min{L d,max , L e d,max , and obtain the range L d ​​​​​​​​​​​L e d |Y e d |Y e d,max |Y e d,min |Y e d |Y

[0031] |L d |L d |L d,max |L d,min |L e d |L e d |L e d,max |L e d,min ;

[0032] C e d ;

[0033] ,

[0034] C e ;

[0035] F e ;

[0036] P e =η C ×C e +η F ×F e , wherein η C , η F are preset failure weight coefficients and node weight coefficients respectively, η C > 0, η F > 0, η C + η F = 1;

[0037] When the node approximation value Pe greater than a preset approximate threshold value, it is determined that the e-th power grid node is approximate to the power grid node, and the e-th power grid node is recorded as an approximate power grid node of the power grid node;

[0038] Step S24: Obtain a plurality of approximate power grid nodes of the power grid node in the power grid and aggregate them to obtain an approximate marked power grid node;

[0039] The marked power grid node is obtained from the approximate power grid node set and recorded as an approximate marked power grid node. The total number of approximate marked power grid nodes of the power grid node is divided by the total number of the plurality of approximate power grid nodes to obtain an outlier value of the power grid node. When the outlier value is greater than a preset outlier threshold value, it is determined that the fault range data of the power grid node is abnormal, and the fault detection of the power grid node is abnormal. The power grid node is recorded as a target power grid node.

[0040] Further, step S3 includes:

[0041] Step S31: Obtain each approximate marked power grid node of the target power grid node, and obtain the marked fault range data of each approximate marked power grid node;

[0042] Step S32: Obtain the fault range data of the target power grid node, and evaluate the g-th monitoring parameter in the fault range data for the target power grid fault capture accuracy. The specific evaluation process is:

[0043] Obtain a plurality of approximate marked power grid nodes in which the g-th monitoring parameter is determined to be inaccurate in fault detection, obtain the sum of the node approximation values between the plurality of approximate marked power grid nodes and the target power grid node, and obtain an approximate total value U sum ;

[0044] Obtain a plurality of approximate marked power grid nodes in which the g-th monitoring parameter is determined to be inaccurate in fault detection, obtain the sum of the node approximation values between the plurality of approximate marked power grid nodes and the target power grid node, and obtain an approximate total value U (g,sum) , calculate the parameter outlier value Q g of the g-th monitoring parameter in the target power grid node (g,sum) / U sum ;

[0045] Step S33: When the parameter outlier value Q g is less than a preset parameter outlier threshold value, it is determined that the range of the g-th monitoring parameter in the preset fault range data is accurate for the target power grid node fault capture. Otherwise, it is determined that the range of the g-th monitoring parameter in the preset fault range data is inaccurate for the target power grid node fault capture. The g-th monitoring parameter is recorded as an abnormal monitoring parameter of the target power grid node;

[0046] Step S34: Obtain a certain abnormal monitoring parameter of the target power grid node, obtain the average value of the maximum value and the minimum value of the certain abnormal monitoring parameter in the marking fault range data of a certain approximate marking power grid node, and record it as a range reference value ζ;

[0047] respectively, and record it as the upper limit distance τ of a certain abnormal monitoring parameter of a certain approximate marking power grid node max and the lower limit distance τ min ;

[0048] Calculate the range reference value ζ' of a certain abnormal monitoring parameter of the target power grid node:

[0049] ,

[0050] wherein, ζ i is the range reference value ζ of a certain abnormal monitoring parameter of the i-th approximate marking power grid node of the target power grid node; P i is the node approximation value between the target power grid node and the i-th approximate marking power grid node; j is the total number of each approximate marking power grid node;

[0051] Step S35: Calculate the upper limit distance τ' of a certain abnormal monitoring parameter of the target power grid node max and the lower limit distance τ' min , obtain the target range W of a certain abnormal monitoring parameter of the target power grid node [ζ' - τ' min , ζ' + τ' max ];

[0052] Obtain the target range of each abnormal monitoring parameter of the target power grid node, use the target range of each abnormal monitoring parameter, replace the range of each abnormal monitoring parameter in the fault range data of the target power grid node, and record the replaced fault range data as the target fault range data of the target power grid node;

[0053] The target power grid node in the above step is obtained according to the marked power grid node that detects the abnormality of the fault, because there is a part of the power grid nodes in the power grid that have no fault, and the accuracy of the fault detection of the fault range data set by the platform cannot be accurately determined according to the actual historical record, but the power grid node that may have a fault detection abnormality, i.e., the target power grid node of the application, can be obtained according to the fault range data between the power grid node and the marked power grid node and the similar situation on the node. The target power grid node is obtained according to the marked fault range data in the approximate marked power grid node, according to the approximate relationship between the target power grid node and different approximate marked power grid nodes, and the fault detection range of each monitoring parameter in different approximate marked power grid nodes, so as to obtain the range that can capture the fault of the target power grid node to the greatest extent, greatly guaranteeing the accuracy of the subsequent abnormality determination of the power grid node in the power grid.

[0054] Further, step S4 comprises:

[0055] Step S41: obtaining the fault range data of each power grid node in the power grid, obtaining the target fault range data of each target power grid node in the power grid, and obtaining the marked fault range data of each marked power grid node in the power grid;

[0056] The fault range data of the power grid node in the power grid is optimized and adjusted, specifically:

[0057] The marked fault range data of each marked power grid node and the target fault range data of each target power grid node are used to replace the fault range data of the corresponding power grid node in each power grid node;

[0058] Step S42: using the replaced fault range data of each power grid node to detect the fault of the power grid, when the maximum value and the minimum value of a monitoring parameter in a power grid node are not in the range of the optimized fault range data, determining that there is an abnormality in the power grid node, and recording the power grid node as an abnormal power grid node;

[0059] Step S43: obtaining the power grid topology of the power grid, obtaining the abnormal area in the power grid according to the distribution situation of the abnormal power grid node in the power grid topology, obtaining and collecting each abnormal area in the power grid to obtain the abnormal area data of the power grid.

[0060] Further, step S5 comprises:

[0061] Step S51: obtaining the abnormal area data of the power grid, obtaining each abnormal area of the power grid from the abnormal area data, obtaining the power grid equipment in the abnormal area, and using a drone to shoot the power grid equipment in the abnormal area to obtain the equipment image;

[0062] Step S52: pre-process the device image, input the pre-processed device image into the detection model using a preset image recognition model, perform fault detection on the power grid equipment, obtain abnormal power grid equipment, and dispatch maintenance personnel to maintain the abnormal power grid equipment in the power grid.

[0063] According to the above method, an intelligent power grid fault detection system based on multi-source heterogeneous data is further proposed, which comprises a range acquisition module, an anomaly identification module and a fault detection module.

[0064] The range acquisition module is used to acquire the marked fault range data of the marked power grid node in the power grid, evaluate the abnormal state of the fault range data for fault detection of the power grid node, obtain the target power grid node, and acquire the target fault range data of the target power grid node.

[0065] The anomaly identification module is used to identify the anomaly of the power grid node in the power grid according to the target fault range data and the marked fault range data, obtain the abnormal power grid node, and obtain the abnormal area data in the power grid according to the distribution situation of the abnormal power grid node in the power grid topology graph.

[0066] The fault detection module is used to use a drone to shoot the power grid equipment in the power grid, acquire the device image, perform fault detection on the power grid equipment, obtain the abnormal power grid equipment, and maintain the abnormal power grid equipment in the power grid.

[0067] Further, the range acquisition module comprises a marked range acquisition unit, a fault detection anomaly evaluation unit and a target range acquisition unit.

[0068] The marked range acquisition unit is used to evaluate the fault detection accuracy of the monitoring parameters in the fault range data on the marked power grid node, and mark the fault range data.

[0069] The fault detection anomaly evaluation unit is used to acquire the approximate marked power grid node of the power grid node, and evaluate the abnormal state of the fault range data for fault detection of the power grid node according to the marked fault range data of the approximate marked power grid node, to obtain the target power grid node.

[0070] The target range acquisition unit is used to evaluate the fault capture accuracy of the monitoring parameters in the fault range data on the target power grid node, to obtain the target fault range data of the target power grid node.

[0071] Further, the anomaly identification module comprises a range optimization unit and an anomaly identification unit.

[0072] The range optimization unit is used to acquire the marked fault range data of the marked power grid node, acquire the target fault range data of the target power grid node, and optimize and adjust the fault range data of the power grid node in the power grid.

[0073] An abnormality identification unit is configured to identify abnormality of the power grid nodes in the power grid according to the optimized fault range data of the power grid nodes in the power grid, to obtain abnormal power grid nodes, and to obtain abnormal area data in the power grid in combination with the power grid topology.

[0074] Further, the fault detection module comprises a fault detection unit.

[0075] The fault detection unit is configured to obtain images of power grid equipment in the abnormal area in the abnormal area data according to the abnormal area data, to detect faults of the power grid equipment, to obtain abnormal power grid equipment, and to maintain the abnormal power grid equipment in the power grid.

[0076] Compared with the prior art, the present application has the beneficial effects that: the present application realizes accurate detection of power grid faults, starting from the pivotal power grid nodes in the actual power grid, considering that the range of abnormal detection of the platform on the power grid nodes in the actual situation may not conform to the actual situation, obtaining the historical fault records of the marked power grid nodes, and evaluating the accuracy of the fault detection of the fault range data of the platform on the power grid nodes according to the historical fault records, thereby obtaining the marked fault range data of the marked power grid nodes, and obtaining the target power grid nodes with fault detection risk according to the node approximation degree between the power grid nodes without faults and the marked power grid nodes, adjusting the fault range data of the platform on the target power grid nodes according to the fault detection accuracy of the target power grid nodes according to the monitoring parameters in the fault range data, and optimizing and adjusting the fault range data of the power grid nodes, accurately detecting the faults of the power grid nodes, and quickly and accurately obtaining the area with equipment faults in the power grid according to the power grid topology, and quickly detecting the faults of the power grid equipment in the abnormal area according to the images taken by the unmanned aerial vehicle, realizing the rapid and accurate identification of the faults in the power grid through the electrical quantity characteristics, topological relationship and image information in the power grid, and greatly improving the efficiency and accuracy of the power grid fault detection. BRIEF DESCRIPTION OF DRAWINGS

[0077] Fig. 1 is the method logic diagram of the intelligent power grid fault detection method based on multi-source heterogeneous data of the present application;

[0078] Fig. 2 is the module flow chart of the intelligent power grid fault detection system based on multi-source heterogeneous data of the present application. DETAILED DESCRIPTION

[0079] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0080] Embodiment: As shown in the figure, the present application provides a technical solution, a smart grid fault detection method based on multi-source heterogeneous data, the method comprising: Figs. 1-2

[0081] Step S1: obtaining the historical fault records and fault range data of the marked grid node in the power grid, evaluating the monitoring parameters in the fault range data for the fault detection accuracy of the marked grid node, and obtaining the marked fault range data;

[0082] In step S1, the following steps are included:

[0083] Step S11: obtaining the grid node that has occurred a fault in the power grid and marking it as a marked grid node, obtaining the fault range data of the marked grid node, and marking the preset range of each monitoring parameter in the fault range data;

[0084] For example, the monitoring parameters include the voltage and current of the grid node and the like;

[0085] Obtaining each historical fault record of the marked grid node, and obtaining the data of each monitoring parameter in the marked grid node from the historical fault record;

[0086] Step S12: when the maximum or minimum value of a certain monitoring parameter in a certain historical fault record is outside the preset fault range data range, the platform issues a fault warning to the marked grid node, and no fault is detected when the marked grid node is subsequently inspected, then the certain historical fault record is recorded as a first historical fault record;

[0087] Obtaining each first historical fault record of the marked grid node, and when the maximum or minimum value of the a-th monitoring parameter in the first historical fault record is outside the fault range data range, determining that the a-th monitoring parameter has a fault detection anomaly in the marked grid node;

[0088] Obtaining the total number E of the a-th monitoring parameter that has a fault anomaly in the marked grid node sum , obtaining the total number γ of each first historical fault record sum , calculating the detection anomaly value Q of the a-th monitoring parameter a =E sum / γ sum , when the detection anomaly value Q a ​If the value exceeds the preset abnormal detection threshold, then the monitoring parameter a is determined to be inaccurate for detecting faults at marked power grid nodes.

[0089] Obtain the range constructed by the minimum and maximum values ​​of the a-th monitoring parameter in each first historical fault record, and denot it as the marked range of the a-th monitoring parameter in the marked power grid node;

[0090] Step S13: Remove each first historical fault record from the other historical fault records to obtain the second historical fault record that marks the power grid node;

[0091] Calculate the range distance value S of the βth monitoring parameter in the second historical fault record. β :

[0092] ,

[0093] Among them, H (max,β) and H (min,β) The range H of the βth monitoring parameter in the fault range data is respectively. β The maximum and minimum values ​​of H; △ For the range H β The average of the maximum and minimum values; H' (max,β) and H' (min,β) These are the maximum and minimum values ​​of the βth monitoring parameter in the second historical fault record, respectively;

[0094] When the range distance value S β If the distance is less than the preset range threshold, it is determined that the βth monitoring parameter has a distance anomaly in the second historical fault record.

[0095] The range anomaly value B of the β-th monitoring parameter is obtained by comparing the total number of second historical fault records showing distance anomalies in the marked power grid node with the total number of second historical fault records for the marked power grid node. β ;

[0096] When the range outlier B β If the value exceeds the preset abnormal threshold, it is determined that the β-th monitoring parameter is inaccurate in detecting faults in the marked power grid node. The range constructed by the maximum and minimum values ​​of the β-th monitoring parameter within the fault range data is obtained from several second historical fault records and recorded as the marked range of the β-th monitoring parameter in the marked power grid node.

[0097] Step S14: Obtain the marking range of several monitoring parameters in the marked power grid node that are determined to be inaccurate in fault detection, and replace the range of several monitoring parameters in the fault range data of the marked power grid node to obtain the marked fault range data of the marked power grid node.

[0098] Step S2: Obtain fault range data and node data of power grid nodes in the power grid, analyze the node approximation degree between different power grid nodes, obtain an approximate power grid node set of the power grid nodes, obtain a marked power grid node from the approximate power grid node set and mark it as an approximate marked power grid node, evaluate the abnormal state of the fault range data for fault detection of the power grid nodes, and obtain a target power grid node;

[0099] Step S2 includes:

[0100] Step S21: Obtain fault range data and node data of power grid nodes, and obtain data of each node index from the node data;

[0101] For example, each node index includes a node degree (the number of edges directly connected to the power grid node), a node voltage level (the specific level of voltage in the power grid node, such as 10kv, 110kv, etc.), and the like;

[0102] Step S22: Analyze the node approximation degree between different power grid nodes to obtain an approximate power grid node set of the power grid nodes, and the specific analysis process is as follows:

[0103] Obtain the range L d of the dth monitoring parameter of the power grid node and the e th power grid node, respectively d,min , d,max and L e d , e d,min , e d,max , respectively, where L d,min , L d,max are the minimum threshold and the maximum threshold of the dth monitoring parameter in the fault range data of the power grid node, L e d,min , e d,max are the minimum threshold and the maximum threshold of the dth monitoring parameter in the fault range data of the e th power grid node, respectively;

[0104] When there is a value within the range L d and L e d , obtain the intersection Y d between L e and L d e d , e d,min , e d,max , where Y e d,min=max{L d,min L d,max}, Y e d,max =min{L d,max L e d,max}, get range L d With L e d Intersection distance between |Y e d |=Y e d,max -Y e d,min Conversely, the intersection distance length |Y e d |=0;

[0105] Calculation range L d Distance length |L d |=L d,max -L d,min Calculation range L e d Distance length |L e d |=L e d,max -L e d,min ;

[0106] Calculate the approximate range C of the d-th monitoring parameter between the e-th power grid node and the d-th power grid node. e d :

[0107] ,

[0108] Obtain the average value of the approximate range values ​​of various monitoring parameters between the power grid node and the e-th power grid node, and denot it as the approximate fault range value C between the power grid node and the e-th power grid node. e ;

[0109] Step S23: Based on the node data of the power grid nodes, construct the node feature vectors of the power grid nodes, obtain the node feature vector of the e-th power grid node, calculate the cosine similarity between the power grid node and the e-th power grid node, and obtain the node approximation value F. e ;

[0110] Calculate the approximate node value P between the e-th node and the power grid node. e =η C ×C e +η F ×F e , where η C ηF These are the preset fault weight coefficient and node weight coefficient, η. C >0, η F >0, η C +η F =1;

[0111] When the node approximation value P e If the value is greater than the preset approximation threshold, then the e-th power grid node is determined to be approximate with the power grid node, and the e-th power grid node is recorded as the approximate power grid node of the power grid node.

[0112] Step S24: Obtain several approximate power grid nodes in the power grid and aggregate them to obtain approximate marked power grid nodes;

[0113] Marked power grid nodes are obtained from the set of approximate power grid nodes and recorded as approximate marked power grid nodes. The total number of approximate marked power grid nodes is divided by the total number of approximate power grid nodes to obtain the abnormal value of the power grid node. When the abnormal value is greater than the preset abnormal threshold, the fault range data of the power grid node is determined. If the fault detection of the power grid node shows an anomaly, the power grid node is recorded as the target power grid node.

[0114] Step S3: Obtain the marked fault range data in the approximate marked power grid node of the target power grid node, evaluate the accuracy of the monitoring parameters in the fault range data for fault capture of the target power grid node, and obtain the target fault range data of the target power grid node;

[0115] Step S3 includes:

[0116] Step S31: Obtain the approximate marked grid nodes of the target grid node and obtain the marked fault range data of each approximate marked grid node;

[0117] Step S32: Obtain the fault range data of the target power grid node, and evaluate the accuracy of the g-th monitoring parameter in the fault range data for capturing faults in the target power grid. The specific evaluation process is as follows:

[0118] Obtain several monitoring parameters from the approximate marked power grid nodes that are determined to have inaccurate operational detection. Then, obtain the sum of the approximate node values ​​between each approximate marked power grid node and the target power grid node to obtain the approximate total value U. sum ;

[0119] Obtain several approximate marked power grid nodes for which the g-th monitoring parameter is determined to be inaccurate in fault detection, and obtain the sum U of the node approximation values ​​between the several approximate marked power grid nodes and the target power grid node. (g,sum) Calculate the abnormal value Q of the g-th monitoring parameter in the target power grid node. g =U (g,sum) / Usum ;

[0120] Step S33: When the parameter abnormal value Q g is less than the preset parameter abnormal threshold, it is determined that the range of the gth monitoring parameter in the preset fault range data is accurate for capturing the fault of the target power grid node, otherwise, it is determined that the range of the gth monitoring parameter in the preset fault range data is inaccurate for capturing the fault of the target power grid node, and the gth monitoring parameter is recorded as an abnormal monitoring parameter of the target power grid node;

[0121] Step S34: An abnormal monitoring parameter of the target power grid node is obtained, and an average value of the maximum value and the minimum value of the abnormal monitoring parameter of the certain one approximate marker power grid node in the marker fault range data is obtained and recorded as a range reference value ζ;

[0122] The absolute value of the difference between the range reference value ζ and the difference between the maximum value and the minimum value of the abnormal monitoring parameter of the certain one approximate marker power grid node in the marker fault range data is obtained respectively, and recorded as an upper limit distance τ max and a lower limit distance τ min of the abnormal monitoring parameter of the certain one approximate marker power grid node;

[0123] The range reference value ζ´ of the abnormal monitoring parameter of the target power grid node is calculated:

[0124] ,

[0125] wherein, ζ i is the range reference value ζ of the abnormal monitoring parameter of the i th approximate marker power grid node of the target power grid node; P i is the node approximation value between the target power grid node and the i th approximate marker power grid node; j is the total number of each approximate marker power grid node;

[0126] Step S35: The upper limit distance τ´ max and the lower limit distance τ´ min of the abnormal monitoring parameter of the target power grid node are calculated, and the target range W of the abnormal monitoring parameter of the target power grid node is obtained as W = [ζ´-τ´ min , ζ´+τ´ max ];

[0127] For example, the specific calculation formula of the upper limit distance τ´ max and the lower limit distance τ´ min is as follows:

[0128] wherein, the calculation formula of the upper limit distance τ´ max is as follows:

[0129] ,

[0130] wherein, τ （i,max） is an upper limit distance of an abnormal monitoring parameter in the i-th approximate marked grid node of the target grid node;

[0131] wherein, the lower limit distance τ´ min is calculated by the following formula:

[0132] ,

[0133] wherein, τ （i,mmin） is a lower limit distance of an abnormal monitoring parameter in the i-th approximate marked grid node of the target grid node;

[0134] obtaining the target range of each abnormal monitoring parameter in the target grid node, using the target range of each abnormal monitoring parameter, replacing the range of each abnormal monitoring parameter in the fault range data in the target grid node, and recording the replaced fault range data as the target fault range data of the target grid node;

[0135] wherein, step S4 comprises:

[0136] Step S41: obtaining the fault range data of each grid node in the grid, obtaining the target fault range data of each target grid node in the grid, and obtaining the marked fault range data of each marked grid node in the grid;

[0137] optimizing and adjusting the fault range data of the grid nodes in the grid, specifically:

[0138] using the marked fault range data of each marked grid node and the target fault range data of each target grid node, replacing the fault range data of the corresponding grid node in each grid node;

[0139] Step S42: using the replaced fault range data of each grid node, detecting faults in the grid, when the maximum value and the minimum value of a monitoring parameter in a certain grid node are not within the range in the optimized fault range data, determining that a certain grid node is abnormal, and recording the certain grid node as an abnormal grid node;

[0140] Step S43: obtaining the grid topology graph of the grid, obtaining the abnormal area in the grid according to the distribution situation of the abnormal grid node in the grid topology graph, obtaining and collecting each abnormal area in the grid to obtain the abnormal area data of the grid;

[0141] For example, the grid topology graph is a topology graph constructed according to the topological relationship on the physical structure of the grid, which clearly shows the connection relationship between each part in the grid;

[0142] The power grid topology mainly includes power grid nodes and edges, wherein the power grid nodes include power grid power plant outlets, substations and the like;

[0143] The edges in the power grid include power transmission lines, power distribution lines, towers, transformers and the like;

[0144] For example, the power grid topology of the power grid is acquired, and according to the distribution situation of the abnormal power grid nodes in the power grid topology, an abnormal area in the power grid is acquired, and the specific acquisition process is as follows:

[0145] The abnormal power grid nodes on the power grid topology are acquired, and when a certain abnormal power grid node is connected to another power grid node through a path in the power grid topology, an area composed of the path between the certain abnormal power grid node and the other abnormal power grid node is recorded as an abnormal area;

[0146] Step S5: according to the abnormal area data, a UAV is dispatched to take pictures of the power grid equipment in the power grid, device images are acquired, fault detection is performed on the power grid equipment, abnormal power grid equipment is obtained, and the abnormal power grid equipment in the power grid is repaired;

[0147] Wherein, step S5 includes:

[0148] Step S51: acquire the abnormal area data of the power grid, acquire each abnormal area of the power grid from the abnormal area data, acquire the power grid equipment in the abnormal area, use a UAV to take pictures of the power grid equipment in the abnormal area, and obtain device images;

[0149] Step S52: pre-process the device images, and use a preset image recognition model to input the pre-processed device images into a detection model, perform fault detection on the power grid equipment, obtain abnormal power grid equipment, and dispatch maintenance personnel to repair the abnormal power grid equipment in the power grid;

[0150] For example, the pre-processing process of the device images includes size adjustment, image enhancement and the like;

[0151] For example, the detection model is constructed using a convolutional neural network model, a large number of images of power grid equipment are used for training, the images contain normal and damaged images, and the recognition accuracy of the model is evaluated by accuracy (the proportion of correctly recognized images) and precision (the proportion of truly damaged samples in the samples predicted as "damaged" by the model);

[0152] According to the above method, an intelligent power grid fault detection system based on multi-source heterogeneous data is also proposed, which includes a range acquisition module, an abnormality recognition module and a fault detection module;

[0153] The range acquisition module is configured to acquire marker fault range data of a marker power grid node in the power grid, evaluate an abnormal state of the fault range data in fault detection of the power grid node, obtain a target power grid node, and acquire target fault range data of the target power grid node.

[0154] The abnormality identification module is configured to identify abnormal power grid nodes in the power grid according to the target fault range data and the marker fault range data, obtain abnormal region data in the power grid according to a distribution situation of the abnormal power grid nodes in the power grid topology diagram.

[0155] The fault detection module is configured to use a UAV to capture images of power grid equipment in the power grid, acquire the equipment images, perform fault detection on the power grid equipment, obtain abnormal power grid equipment, and perform maintenance on the abnormal power grid equipment in the power grid.

[0156] The range acquisition module includes a marker range acquisition unit, a fault detection abnormality evaluation unit, and a target range acquisition unit.

[0157] The marker range acquisition unit is configured to evaluate a fault detection accuracy of a monitoring parameter in the fault range data on the marker power grid node, and acquire the marker fault range data.

[0158] The fault detection abnormality evaluation unit is configured to acquire an approximate marker power grid node of the power grid node, evaluate an abnormal state of the fault range data in fault detection of the power grid node according to marker fault range data of the approximate marker power grid node, and obtain the target power grid node.

[0159] The target range acquisition unit is configured to evaluate a fault capture accuracy of a monitoring parameter in the fault range data on the target power grid node, and obtain the target fault range data of the target power grid node.

[0160] The abnormality identification module includes a range optimization unit and an abnormality identification unit.

[0161] The range optimization unit is configured to acquire the marker fault range data of the marker power grid node, acquire the target fault range data of the target power grid node, and optimize and adjust the fault range data of the power grid node in the power grid.

[0162] The abnormality identification unit is configured to identify abnormal power grid nodes in the power grid according to the optimized fault range data of the power grid node in the power grid, obtain abnormal power grid nodes, and acquire abnormal region data in the power grid in combination with the power grid topology diagram.

[0163] The fault detection module includes a fault detection unit.

[0164] The fault detection unit is used for obtaining the power grid equipment image in the abnormal area in the abnormal area data according to the abnormal area data, performing fault detection on the power grid equipment, obtaining the abnormal power grid equipment, and performing maintenance on the abnormal power grid equipment in the power grid.

[0165] It will be obvious to a person skilled in the art that the application is not limited to the details of the foregoing exemplary embodiments and can be implemented in other concrete forms without departing from the spirit or essential characteristics of the application. The embodiments are considered in all respects to be illustrative and not restrictive, and the scope of the application is defined by the appended claims rather than the foregoing description, and it is intended to embrace all changes and modifications that fall within the meaning and scope of the equivalent elements of the claims. Any reference signs in the claims should not be considered as limiting the claims involved.

Claims

1. A smart grid fault detection method based on multi-source heterogeneous data, characterized in that, The method comprises: Step S1: obtaining historical fault records and fault range data of a marked power grid node in a power grid, evaluating the fault detection accuracy of the monitoring parameters in the fault range data on the marked power grid node, and obtaining marked fault range data; Step S2: obtaining fault range data and node data of power grid nodes in the power grid, analyzing the node approximation degree between different power grid nodes, obtaining an approximate power grid node set of the power grid nodes, obtaining a marked power grid node from the approximate power grid node set and recording it as an approximate marked power grid node, evaluating the abnormal state of the fault range data for fault detection of the power grid nodes, and obtaining a target power grid node; Step S3: obtaining the marked fault range data in the approximate marked power grid node of the target power grid node, evaluating the fault capture accuracy of the monitoring parameters in the fault range data on the target power grid node, and obtaining target fault range data of the target power grid node; Step S4: according to the target fault range data and the marked fault range data, optimizing and adjusting the fault range data of the power grid nodes in the power grid, identifying the abnormal power grid nodes in the power grid, obtaining the abnormal region data in the power grid according to the distribution of the abnormal power grid nodes in the power grid topology graph, and obtaining the abnormal region data in the power grid; Step S5: according to the abnormal region data, sending a drone to take pictures of the power grid equipment in the power grid, obtaining equipment images, detecting faults of the power grid equipment, obtaining abnormal power grid equipment, and repairing the abnormal power grid equipment in the power grid. 2.The smart grid fault detection method based on multi-source heterogeneous data according to claim 1, characterized in that, The step S1 comprises: Step S11: obtaining a power grid node that has failed in the power grid and recording it as a marked power grid node, obtaining the fault range data of the marked power grid node, and marking the preset range of each monitoring parameter in the marked power grid node from the fault range data; obtaining each historical fault record of the marked power grid node, and obtaining the data of each monitoring parameter in the marked power grid node from the historical fault record; Step S12: when the maximum value or the minimum value of a certain monitoring parameter in a certain historical fault record is outside the preset fault range data range, the platform issues a fault warning to the marked power grid node, and subsequent inspection of the marked power grid node does not detect a fault, then the certain historical fault record is recorded as a first historical fault record; obtaining each first historical fault record of the marked power grid node, when the maximum value or the minimum value of the a-th monitoring parameter in the first historical fault record is outside the fault range data range, determining that the a-th monitoring parameter has a fault detection abnormality in the marked power grid node; Obtaining the total number E of times that the a-th monitoring parameter exists in the marked power grid node fault anomaly sum , obtaining the total number γ of each first historical fault record sum , calculating the detection anomaly value Q of the a-th monitoring parameter a =E sum / γ sum , when the detection anomaly value Q a is greater than the preset detection anomaly threshold value, it is determined that the a-th monitoring parameter is inaccurate for detecting the fault of the marked power grid node; obtaining the range constructed by the minimum value and the maximum value of the a-th monitoring parameter in each first historical fault record, and recording it as the marked range of the a-th monitoring parameter in the marked power grid node; Step S13: excluding each first historical fault record from each historical fault record to obtain second historical fault records of the marked power grid node; calculating a range distance value S of the βth monitoring parameter in the second historical failure record β : ， where H (max,β) and H (min,β) are the maximum and minimum values of the range H β of the βth monitoring parameter in the fault range data; H △ is the average of the maximum and minimum values of the range H β ; H´ (max,β) and H´ (min,β) are the maximum and minimum values of the βth monitoring parameter in the second historical fault record, respectively. When the range distance value S β is less than the preset range distance threshold value, it is determined that the βth monitoring parameter has a distance anomaly in the second historical failure record. Obtain the total number of second historical fault records in which the βth monitoring parameter in the marked power grid node has distance anomaly, and the ratio of the total number of second historical fault records of the marked power grid node, to obtain the range anomaly value B of the βth monitoring parameter β ; When the range abnormal value B β If the range abnormal value B is greater than the preset range abnormal threshold, it is determined that the βth monitoring parameter is inaccurate for the fault detection of the marked grid node, and a range is constructed from the maximum value and the minimum value of the βth monitoring parameter in the fault range data in the several second historical fault records, and is recorded as the marked range of the βth monitoring parameter in the marked grid node. Step S14: obtaining a marked range of the number of monitoring parameters in the marked power grid node determined as inaccurate in fault detection, and replacing the range of the number of monitoring parameters in the fault range data of the marked power grid node to obtain the marked fault range data of the marked power grid node. 3.The smart grid fault detection method based on multi-source heterogeneous data according to claim 2, characterized in that, The step S2 comprises: Step S21: obtaining power grid node fault range data and node data, and obtaining data of each node index from the node data; Step S22: analyzing the node approximation degree between different power grid nodes to obtain an approximate power grid node set of the power grid nodes, and the specific analysis process is as follows: obtain a range L d of the dth monitoring parameter of the power grid node and the e th power grid node respectively d,min , d,max and L e d , e d,min , e d,max , wherein L d,min , L d,max are the minimum threshold value and the maximum threshold value of the dth monitoring parameter in the fault range data of the power grid node respectively, L e d,min , e d,max are the minimum threshold value and the maximum threshold value of the dth monitoring parameter in the fault range data of the e th power grid node respectively; When there is a certain value in the range L d and L e , get the intersection Y d between L d and L e d e d =[Y e d,min , Y e d,max ], wherein Y e d,min =max{L d,min , L d,max}, Y e d,max =min{L d,max , L e d,max}, get the intersection distance |Y d | between the range L e and L d e d =Y e d,max -Y e d,min , otherwise, the intersection distance length |Y e d |=0; The distance length |L d of the range L d = L d,max - L d,min , the distance length |L e of the range L d = L e - L d = L e - L d,max = L e - L d,min ; calculating an approximation of a range of the dth monitoring parameter between the power grid node and the e th power grid node e d : ， obtaining an average of the range approximations of the monitoring parameters between the power grid node and the e-th power grid node, and denoted as a fault range approximation C between the power grid node and the e-th power grid node e ; Step S23: According to the node data of the power grid node, the node feature vector of the power grid node is constructed, the node feature vector of the e th power grid node is obtained, the cosine similarity between the power grid node and the e th power grid node is calculated, and the node approximation F is obtained e ; Calculate the node approximation value P between the grid node and the e-th grid node e =η C ×C e +η F ×F e , wherein η C , η F are preset fault weight coefficients and node weight coefficients respectively, η C >0, η F >0, η C +η F =1; When the node approximation value P e is greater than a preset approximation threshold, it is determined that the e-th power grid node is approximated with the power grid node, and the e-th power grid node is recorded as an approximated power grid node of the power grid node. Step S24: obtaining a plurality of approximate power grid nodes of the power grid nodes and collecting them to obtain approximate marked power grid nodes; Obtaining the marked power grid node from the approximate power grid node set, and recording it as an approximate marked power grid node. The abnormal value of the power grid node is obtained by dividing the total number of the approximate marked power grid nodes of the power grid node by the total number of the plurality of approximate power grid nodes. When the abnormal value is greater than a preset abnormal threshold, it is determined that the fault range data of the power grid node has an abnormality in fault detection, and the power grid node is recorded as a target power grid node. 4.The smart grid fault detection method based on multi-source heterogeneous data according to claim 3, characterized in that, The step S3 comprises: Step S31: obtaining each approximate marked power grid node of the target power grid node, and obtaining the marked fault range data of each approximate marked power grid node; Step S32: obtaining the fault range data of the target power grid node, and evaluating the gth monitoring parameter in the fault range data for the target power grid fault capture accuracy, and the specific evaluation process is as follows: Obtaining a number of monitoring parameters determined as inaccurate in work detection in the approximate marked power grid node, obtaining a sum of node approximations between each approximate marked power grid node and the target power grid node, and obtaining an approximate total sum value U sum ; Obtaining a plurality of approximate marked power grid nodes determined as inaccurate fault detection for the gth monitoring parameter, obtaining a sum U of node approximations between the plurality of approximate marked power grid nodes and the target power grid node (g,sum) , calculating a parameter outlier Q of the gth monitoring parameter in the target power grid node g =U (g,sum) / U sum ; Step S33: When the parameter abnormal value Q g If the parameter abnormal value Q is less than the preset parameter abnormal threshold, it is determined that the range of the gth monitoring parameter in the preset fault range data is accurate for capturing the fault of the target power grid node, otherwise, it is determined that the range of the gth monitoring parameter in the preset fault range data is inaccurate for capturing the fault of the target power grid node, and the gth monitoring parameter is recorded as an abnormal monitoring parameter of the target power grid node. Step S34: obtaining a certain abnormal monitoring parameter of the target power grid node, obtaining the average value of the maximum value and the minimum value of the certain abnormal monitoring parameter in the marked fault range data of a certain approximate marked power grid node, and recording it as a range reference value ζ; respectively, and recorded as the upper limit distance τ of the abnormal monitoring parameter of the certain one approximate marked power grid node max and the lower limit distance τ min ; Calculating the range reference value ζ' of the certain abnormal monitoring parameter in the target power grid node: ， wherein ζ i is a range reference value ζ of an abnormal monitoring parameter in the i-th approximate marked grid node of the target grid node; P i is a node approximation value between the target grid node and the i-th approximate marked grid node; j is the total number of the approximate marked grid nodes. Step S35: calculating the upper limit distance τ´ of the abnormal monitoring parameter in the target power grid node max and the lower limit distance τ´ min , obtaining the target range W=[ζ´-τ´ min ,ζ´+τ´ max ] of the abnormal monitoring parameter in the target power grid node; Obtaining the target range of each abnormal monitoring parameter in the target power grid node, replacing the range of each abnormal monitoring parameter in the fault range data of the target power grid node using the target range of each abnormal monitoring parameter, and recording the replaced fault range data as the target fault range data of the target power grid node.

5. The method for smart grid fault detection based on multi-source heterogeneous data according to claim 4, characterized in that, The step S4 comprises: Step S41: obtaining the fault range data of each power grid node in the power grid, obtaining the target fault range data of each target power grid node in the power grid, and obtaining the marked fault range data of each marked power grid node in the power grid; Optimizing and adjusting the fault range data of the power grid nodes in the power grid, specifically: Replacing the fault range data of the corresponding power grid nodes in each power grid node using the marked fault range data of each marked power grid node and the target fault range data of each target power grid node; Step S42: using the replaced fault range data of each power grid node to perform fault detection on the power grid. When the maximum value and the minimum value of a certain monitoring parameter in a certain power grid node are not within the range in the fault range data after optimization, it is determined that the certain power grid node has an abnormality, and the certain power grid node is recorded as an abnormal power grid node; Step S43: Obtain a power grid topology of the power grid, and according to a distribution state of the abnormal power grid nodes in the power grid topology, obtain abnormal areas in the power grid, collect the abnormal areas in the power grid, and obtain abnormal area data of the power grid. 6.The smart grid fault detection method based on multi-source heterogeneous data according to claim 5, characterized in that, The step S5 comprises: Step S51: Obtain the abnormal area data of the power grid, obtain each abnormal area in the power grid from the abnormal area data, obtain power grid equipment in the abnormal area, use a UAV to take pictures of the power grid equipment in the abnormal area, and obtain equipment images; Step S52: Preprocess the equipment images, input the preprocessed equipment images into a detection model using a preset image recognition model, perform fault detection on the power grid equipment, obtain abnormal power grid equipment, and dispatch maintenance personnel to maintain the abnormal power grid equipment in the power grid.

7. An intelligent power grid fault detection system based on multi-source heterogeneous data, configured to perform the intelligent power grid fault detection method based on multi-source heterogeneous data according to any one of claims 1-6, characterized in that, The system comprises a range obtaining module, an abnormality identification module, and a fault detection module; The range obtaining module is configured to obtain the marked fault range data of the marked power grid nodes in the power grid, evaluate an abnormal state of the fault range data in fault detection on the power grid nodes, obtain target power grid nodes, and obtain target fault range data of the target power grid nodes. The abnormality identification module is configured to identify abnormal power grid nodes in the power grid according to the target fault range data and the marked fault range data, obtain abnormal power grid nodes, and obtain abnormal area data in the power grid according to a distribution state of the abnormal power grid nodes in a power grid topology. The fault detection module is configured to use a UAV to take pictures of power grid equipment in the power grid, obtain equipment images, perform fault detection on the power grid equipment, obtain abnormal power grid equipment, and maintain the abnormal power grid equipment in the power grid.

8. The intelligent power grid fault detection system based on multi-source heterogeneous data according to claim 7, characterized in that, The range obtaining module comprises a marked range obtaining unit, a fault detection abnormality evaluation unit, and a target range obtaining unit. The marked range obtaining unit is configured to evaluate the fault detection accuracy of the monitoring parameters in the fault range data on the marked power grid nodes, and mark the fault range data. The fault detection abnormality evaluation unit is configured to obtain approximate marked power grid nodes, and evaluate an abnormal state of the fault range data in fault detection on the power grid nodes according to the marked fault range data of the approximate marked power grid nodes, and obtain target power grid nodes. The target range obtaining unit is configured to evaluate the fault capture accuracy of the monitoring parameters in the fault range data on the target power grid nodes, and obtain target fault range data of the target power grid nodes.

9. The intelligent power grid fault detection system based on multi-source heterogeneous data according to claim 7, characterized in that, The abnormality identification module comprises a range optimization unit and an abnormality identification unit. The range optimization unit is configured to obtain the marked fault range data of the marked power grid nodes, obtain target fault range data of the target power grid nodes, and optimize and adjust the fault range data of the power grid nodes in the power grid. The abnormality identification unit is configured to identify abnormal power grid nodes in the power grid according to the optimized fault range data of the power grid nodes in the power grid, obtain abnormal power grid nodes, and obtain abnormal area data in the power grid in combination with a power grid topology.

10. The intelligent power grid fault detection system based on multi-source heterogeneous data according to claim 7, characterized in that, The fault detection module comprises a fault detection unit. The fault detection unit is configured to acquire an image of a power grid device in an abnormal area in the abnormal area data according to the abnormal area data, perform fault detection on the power grid device, obtain an abnormal power grid device, and perform maintenance on the abnormal power grid device in the power grid.

Images