Electric power fault location accurate identification method based on Beidou technology

Abstract

The invention relates to the technical field of power fault location, and discloses a Beidou technology-based power fault location accurate identification method, which comprises the following steps of: acquiring power grid line data, filling a missing value according to a context, and constructing a target region power network topology; constructing a regional convergence access device deployment optimization objective function and solving to obtain an optimal access device deployment position scheme and performing deployment; and converging the power sensor data of the target area by using the deployed access equipment, constructing a power fault recognition model, and performing power fault recognition and positioning on the power sensor data converged in the target area. Information fusion is carried out based on position information of adjacent power sensors and time sequence information of power sensor data, feature vectors representing abnormal states of the power sensor data are obtained, power fault identification is carried out on the power sensor data, and access equipment carries out power fault positioning according to transmission directions of the power sensor data.

Description

Technical Field

[0001] The present invention relates to the technical field of power fault locating, and in particular to a method for accurately identifying the location of a power fault based on Beidou technology. Background Art

[0002] With the continuous development of society and the increase in electricity demand, the power system is developing in the direction of large-scale, high efficiency and intelligence. However, there are various faults in the power system, such as line short circuit, equipment failure, etc. These faults bring serious safety hazards to the operation of the power system and may even cause accidents. How to quickly and accurately identify faults in the power system has become an urgent problem to be solved in the operation and management of the power system. As a satellite navigation system independently developed by my country, the Beidou satellite navigation system has the characteristics of global coverage, high precision, high reliability and high robustness. Beidou technology has a wide range of applications in positioning, navigation and time services, and its performance and coverage are constantly improving. It has become an important member of the global satellite navigation system. Applying Beidou technology to the accurate identification of the fault location of the power system is expected to achieve rapid positioning and accurate identification of the fault location of the power system through its high-precision positioning and navigation capabilities, thereby improving the safety and reliability of the power system. In view of this, the present invention proposes a method for accurately identifying the location of power faults based on Beidou technology. By combining Beidou technology with the power system, new ideas and methods are provided for the intelligent operation and management of the power system, and the modernization of the power system is promoted. Summary of the invention

[0003] In view of this, the present invention provides a method for accurately identifying the location of power faults based on Beidou technology, the purpose of which is to: 1) iterate the missing value filling result according to the deviation between the power grid line data after filling the missing values ​​and the original power grid line data, and obtain the missing value filling result with the highest structural similarity, so as to achieve the missing value filling of the power grid line data, and construct the target area power network topology, take the transmission station as the node, the power grid line between the transmission station as the edge, and deploy power sensors at the transmission station to sense the power transmission of the transmission station, construct the regional convergence access equipment deployment optimization objective function, take the coverage and coverage accuracy of the access equipment to the node as the optimization target, adopt the improved dung beetle optimization algorithm, and iterate the access equipment deployment location scheme in multiple ways in the iterative manner of different types of dung beetles, 1) Obtain the optimal access device deployment location plan, and use the access device to aggregate the power sensor data in the coverage area; 2) Use the deployed access device to aggregate the power sensor data in the target area, build a power fault identification model, perform information fusion based on the location information of adjacent power sensors and the timing information of the power sensor data, obtain the feature vector that characterizes the abnormal state of the power sensor data, and identify the power fault of the power sensor data aggregated in the target area. The access device locates the power sensor that has identified the power fault according to the transmission direction of the power sensor data, identifies the power sensor data with the power fault, and uses the node location of the power sensor as the power fault area location, so as to realize the accurate identification of the power fault location based on the power sensor data information and Beidou positioning technology.

[0004] To achieve the above object, the present invention provides a method for accurately identifying the location of power faults based on Beidou technology, comprising the following steps:

[0005] S1: Collect power grid line data and fill missing values ​​according to the context to obtain preprocessed power grid line data, build the target area power network topology based on the preprocessed power grid line data, and deploy power sensors in the target area power network topology;

[0006] S2: Construct the regional convergence access device deployment optimization objective function, where the regional convergence access device deployment optimization objective function takes the access device deployment location as the independent variable, and takes the access device's coverage of the target area power network topology and coverage accuracy as the optimization target. The access device is a regional data aggregation access point with Beidou positioning function, which is responsible for aggregating and transmitting power sensor data within the coverage area;

[0007] S3: Taking maximizing the coverage and coverage accuracy of the access equipment to the target area power network topology as the optimization goal, the regional aggregation access equipment deployment optimization objective function is optimized and solved, and the optimal access equipment deployment location plan is obtained and deployed, wherein the improved dung beetle optimization algorithm is the main implementation method of the optimization solution;

[0008] S4: Use the deployed access devices to aggregate the power sensor data of the target area, build a power fault identification model, identify the power fault on the power sensor data aggregated in the target area, identify the power sensor data with power faults, use the access devices to locate the power sensors that have identified the power faults, and obtain the location of the power fault area.

[0009] As a further improvement method of the present invention:

[0010] Optionally, in step S1, collecting power grid line data and filling missing values ​​according to context to obtain preprocessed power grid line data includes:

[0011] Collect power grid line data, where the collected power grid line data is in the form of:

[0012] {L n |n∈[1,N]}

[0013] L n =(L n (1),L n (2),...,L n (m),...,L n (M))

[0014] in:

[0015] L n represents the collected grid line data of the nth grid line, N represents the total number of grid lines; the grid line is the transmission line between transmission stations;

[0016] L n (1),L n (2),...,L n (m),...,L n (M) represents the current value collected by the nth power line at M collection moments, L n (m) represents the current value of the nth power line at the mth acquisition moment, m∈[1,M];

[0017] Fill in missing values ​​in the power grid line data according to the context, where the power grid line data L n Missing value L n The filling process of (q) is:

[0018] S11: Initialize and generate U missing value filling results, where the u-th missing value filling result generated by initialization is:

[0019] S12: Set the current iteration number of the missing value filling result to t, the initial value of t is 0, and the maximum value is Max, then the missing value filling result The result of the tth iteration is

[0020] S13: Fill the missing value filling result into the power grid line data L n The corresponding power grid line data is obtained, where the missing values ​​are filled in. The power grid line data is

[0021] Calculate power grid line data With the grid line data L n The error between Fitness:

[0022]

[0023] in:

[0024] Indicates the missing value filling result Adaptability;

[0025] μ n Indicates the power grid line data L n The mean of Indicates power grid line data The mean of

[0026] σ n Indicates the power grid line data L n The standard deviation of Indicates power grid line data The standard deviation of

[0027] ∈1,∈2 represent adjustment coefficients;

[0028] S14: Select the missing value filling result with the lowest fitness among the U missing value filling results as the optimal filling result obtained in the tth iteration

[0029] S15: Iterate the missing value filling results, where the missing value filling results The iteration formula is:

[0030]

[0031] in:

[0032] rand(0,1) represents a random number between 0 and 1;

[0033] c1, c2 represent control parameters;

[0034] exp(·) represents an exponential function with a natural constant as the base;

[0035] express The missing value filling result with the smallest fitness;

[0036] S16: Let t = t + 1, return to step S13, until the maximum number of iterations is reached, and select the missing value filling result with the lowest fitness among the U missing value filling results obtained in the final iteration, and use the selected missing value filling result as the power grid line data L n Missing value L n (q) fill value, get the power grid line data L n The preprocessing result of ′ n ;

[0037] Construct the target area power network topology based on the preprocessed power grid line data.

[0038] Optionally, constructing the target area power network topology based on the preprocessed power grid line data includes:

[0039] The target area power network topology is constructed based on the preprocessed power grid line data, where the construction process of the target area power network topology is as follows:

[0040] The transmission stations in the power network are used as nodes in the target area power network, and the power grid lines between the transmission stations are used as edges between nodes to form the target area power network topology:

[0041] G=(E,V)

[0042] E={e s |s∈[1,S]}

[0043]

[0044] in:

[0045] G represents the target area power network topology;

[0046] E represents the node set in the target area power network, e s represents the sth node in the target area power network, S represents the total number of nodes in the target area power network;

[0047] V represents the edge set in the target area power network, Represents node e s With e i The edge between

[0048] Calculate the weights of the edges in the target area power network topology, where the edges The transmission weight is:

[0049]

[0050] in:

[0051] Represents edge The transmission weight of

[0052] mean(·) means calculating the mean. express Corresponding pre-processed power grid line data;

[0053] Deploy power sensors in the target area power network topology.

[0054] Optionally, the deploying of power sensors in the target area power network topology includes:

[0055] The power sensors are deployed at the locations of S nodes in the target area power network topology to form a power sensor set:

[0056] h={h s |s∈[1,S]}

[0057] in:

[0058] h represents the power sensor set, h s represents the sth node e deployed in the target area power network topology s Power sensors at the location.

[0059] Optionally, constructing a regional convergence access device deployment optimization objective function in step S2 includes:

[0060] The regional convergence access equipment deployment optimization objective function is constructed, where the regional convergence access equipment deployment optimization objective function takes the access equipment deployment location as the independent variable, and takes the access equipment's coverage of the target area power network topology and coverage accuracy as the optimization target. The access equipment is a regional data convergence access point with Beidou positioning function, which is responsible for aggregating and transmitting power sensor data within the coverage area. The constructed regional convergence access equipment deployment optimization objective function is:

[0061]

[0062] θ=(θ1,θ2,...,θ k ,...,θ K )

[0063] in:

[0064] F(·) represents the optimization objective function of regional aggregation access equipment deployment, θ represents the access equipment deployment location plan, θ1,θ2,...,θ k ,...,θ K represents the deployment positions of K access devices in the access device deployment position scheme θ, and F(θ) represents the objective function value of the access device deployment position scheme θ;

[0065] W s Represents node e s The transmission weight, Ω s Represents the target area power network topology with node e s is the set of edges with endpoints, c∈Ω s , c represents the set Ω s For any edge in , w(c) represents the transmission weight of edge c; Used to indicate the coverage accuracy of access equipment;

[0066] α(θ k ,e s ) means deployed in θ k The access device of node e s The coverage of max{α(θ k ,e s )|k∈[1,K]} represents the access device deployment location solution θ for node e s coverage;

[0067] R represents the aggregation radius of the access device;

[0068] dis(θ k ,e s ) represents θ k With node e s The distance between the locations.

[0069] Optionally, in the step S3, maximizing the coverage and coverage accuracy of the access equipment to the target area power network topology is taken as the optimization goal, and the regional convergence access equipment deployment optimization objective function is optimized and solved, including:

[0070] Taking maximizing the coverage and coverage accuracy of the access equipment to the target area power network topology as the optimization goal, the regional convergence access equipment deployment optimization objective function is optimized and solved, wherein the improved dung beetle optimization algorithm is the main implementation method of the optimization solution, and the optimization solution process of the regional convergence access equipment deployment optimization objective function is as follows:

[0071] S31: Initialize and generate the positions of D dung beetles, wherein the initialization generation formula for the position of the dth dung beetle is:

[0072]

[0073] in:

[0074] represents the initialization position of the d-th dung beetle, d∈[1,D]; in the embodiment of the present invention, is a randomly generated position;

[0075] S32: Set the current iteration number of the dung beetle position to z, and the maximum iteration number to Z, then the zth iteration position of the d dung beetle is

[0076]

[0077] in:

[0078] The k-th dimension coordinate representing the iteration position corresponds to the deployment position of the k-th access device;

[0079] S33: Using the iteration position of the dung beetle as the input value of the regional convergence access device deployment optimization objective function to obtain the objective function value corresponding to the iteration position, where the iteration position The corresponding objective function value is

[0080] Sort the D groups of iteration positions obtained in the zth iteration in descending order of the objective function value, and evenly divide the sorting results to obtain 3 groups of iteration position sequences, where each group of iteration position sequences has D / 3 groups of iteration positions;

[0081] S34: using the iteration position in the first group of iteration position sequences as the position of the rolling ball dung beetle, and performing rolling ball dung beetle position iteration:

[0082]

[0083] in:

[0084] σ represents the deflection coefficient, and σ is set to 0.2;

[0085] It represents the iteration position with the minimum objective function value among the D group iteration positions obtained in the z-th iteration;

[0086] The iteration position in the second set of iteration position sequence is used as the position of the dancing dung beetle, and the dancing dung beetle position iteration is performed:

[0087]

[0088] in:

[0089] τ represents a random angle in (0,π / 2);

[0090] The iterative position in the third set of iterative position sequences is used as the position for breeding dung beetles, and the breeding dung beetle position is iterated:

[0091]

[0092] in:

[0093] It represents the iteration position with the largest objective function value among the D group iteration positions obtained in the z-th iteration;

[0094] S35: Let d=d+1, and return to step S33 until the maximum number of iterations is reached, and the dung beetle position with the largest objective function value after the maximum number of iterations is reached is used as the optimal access device deployment position solution, and the access device is deployed.

[0095] Optionally, in step S4, the power sensor data of the target area is aggregated by the deployed access device to construct a power fault identification model, including:

[0096] The deployed access devices are used to aggregate the power sensor data in the target area and upload the data sensed by the power sensors to the access devices. s The power sensor data sensed is H s :

[0097] H s =(H s (1),H s (2),...,H s (a),...,H s (A)

[0098] in:

[0099] H s (1),H s (2),...,H s (a),...,H s (A) represents the power sensor h s The perceived node es The power transmission sequence,

[0100] H s (a) represents node e s The amount of power transmitted at the ath transmission moment, A represents the number of transmission moments, a∈[1,A];

[0101] Constructing a power fault identification model, wherein the power fault identification model includes an input layer, a location association information extraction layer, a time sequence information extraction layer, an information fusion layer, and a fault identification layer;

[0102] The input layer is used to receive the power sensor data to be identified and the associated adjacent power sensor data, wherein there is an uninterrupted power grid line between the power sensor and the node where the adjacent power sensor is deployed, and the adjacent power sensor data is the data sensed by the adjacent power sensor. The position association information extraction layer is used to extract the position association information between the adjacent power sensor and the power sensor to be identified. The timing information extraction layer is used to extract the timing information of the power sensor data and the associated adjacent power sensor data respectively. The information fusion layer is used to fuse the position association information and the timing information to form a feature vector that characterizes the abnormal state of the power sensor data. The fault identification layer is used to convert the feature vector into a fault probability. If the fault probability is higher than a preset threshold, it indicates that there is a power fault in the power sensor data.

[0103] Power fault identification is performed on the power sensor data gathered in the target area to identify the power sensor data with power faults.

[0104] Optionally, the using the power fault identification model to identify power faults in the power sensor data gathered in the target area to identify the power sensor data with power faults includes:

[0105] The power fault identification model is used to identify power faults in the power sensor data collected in the target area, where the power sensor data H s The power fault identification process is as follows:

[0106] S41: The input layer receives the power sensor data H to be identified s and associated proximity power sensor data Where B represents the power sensor data H s The total number of nearby power sensor data, Indicates power sensor data H s The bth neighboring power sensor data, b∈[1,B];

[0107] S42: The location association information extraction layer extracts the neighboring power sensor data corresponding to the neighboring power sensor and the power sensor h sThe location association information between the adjacent power sensor data Corresponding nearby power sensor With power sensor s The location association information between them is:

[0108]

[0109] in:

[0110] Indicates proximity to power sensors With power sensor s The location association information between them;

[0111] Indicates proximity to power sensors With power sensor s the distance between deployed positions;

[0112] Indicates proximity to power sensors With power sensor s The transmission weight of the power grid lines between the corresponding nodes;

[0113] S43: The timing information extraction layer extracts timing information of the power sensor data and the associated adjacent power sensor data respectively, wherein the power sensor data H s The result of extracting the timing information is:

[0114]

[0115] in:

[0116] W 1 represents the time series information convolution matrix, * represents the convolution operator;

[0117] F2(H s ) represents the power sensor data H s The time series information extraction results;

[0118] S44: The information fusion layer fuses the position association information and the time sequence information to form a feature vector representing the abnormal state of the power sensor data:

[0119]

[0120] in:

[0121] F3(H s ) represents the power sensor data H s The feature vector of abnormal state;

[0122] S45: The fault identification layer transforms the feature vector F3(H s ) is converted into the failure probability P(H s ), if the failure probability P(H s ) is higher than the preset threshold, it means that the power sensor data H s There is a power failure, where the failure probability P(H s ) is:

[0123]

[0124] in:

[0125] W 2 Represents the probability mapping convolution matrix;

[0126] The access device locates the power sensor that has identified the power fault according to the transmission direction of the power sensor data, and uses the node location of the power sensor as the power fault area location.

[0127] In order to solve the above problem, the present invention provides an electronic device, the electronic device comprising:

[0128] A memory storing at least one instruction;

[0129] Communication interface, enabling electronic equipment to communicate; and

[0130] The processor executes the instructions stored in the memory to implement the above-mentioned method for accurately identifying the location of power faults based on Beidou technology.

[0131] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned method for accurately identifying the location of power faults based on Beidou technology.

[0132] Compared with the prior art, the present invention proposes a method for accurately identifying the location of power faults based on Beidou technology, which has the following advantages:

[0133] First, this scheme proposes an access device deployment scheme and constructs a regional convergence access device deployment optimization objective function, in which the regional convergence access device deployment optimization objective function takes the access device deployment location as the independent variable, and takes the access device's coverage of the target area power network topology and coverage accuracy as the optimization target. The access device is a regional data aggregation access point with Beidou positioning function, which is responsible for aggregating and transmitting power sensor data within the coverage area. The constructed regional convergence access device deployment optimization objective function is:

[0134]

[0135] θ=(θ1,θ2,...,θ k ,...,θ K )

[0136] Where: F(·) represents the optimization objective function of regional aggregation access equipment deployment, θ represents the access equipment deployment location plan, θ1,θ2,...,θ k ,...,θ K represents the deployment positions of K access devices in the access device deployment position scheme θ, and F(θ) represents the objective function value of the access device deployment position scheme θ; W s Represents node e s The transmission weight, Ω s Represents the target area power network topology with node e s is the set of edges with endpoints, c∈Ω s , c represents the set Ω s For any edge in , w(c) represents the transmission weight of edge c; Used to indicate the coverage accuracy of access equipment; α(θ k ,e s ) means deployed in θ k The access device of node e s The coverage of max{α(θ k ,e s )|k∈[1,K]} represents the access device deployment location solution θ for node e s The coverage of the access device; R represents the convergence radius of the access device; dis(θ k ,e s ) represents θ k With node e s The distance between the locations. This scheme iterates the missing value filling results according to the deviation between the grid line data after filling the missing values ​​and the original grid line data, and obtains the missing value filling result with the highest structural similarity, realizes the missing value filling of the grid line data, and constructs the target area power network topology, taking the transmission station as the node, the grid line between the transmission station as the edge, and deploying power sensors at the transmission station to sense the power transmission of the transmission station, constructing the regional aggregation access equipment deployment optimization objective function, taking the coverage and coverage accuracy of the access equipment to the node as the optimization target, adopting the improved dung beetle optimization algorithm, and iterating the access equipment deployment location plan in multiple ways in the iterative method of different types of dung beetles, to obtain the optimal access equipment deployment location plan, and using the access equipment to aggregate the power sensor data in the transmission coverage area.

[0137] At the same time, this scheme proposes a power fault identification method and fault location method based on power sensor data, and uses the power fault identification model to identify power faults on the power sensor data gathered in the target area, where the power sensor data H s The power fault identification process is as follows: the input layer receives the power sensor data H to be identified s and the associated neighboring power sensor data; the location association information extraction layer extracts the neighboring power sensor data corresponding to the neighboring power sensor and the power sensor h s The location association information between the adjacent power sensor data Corresponding nearby power sensor With power sensor s The location association information between them is:

[0138]

[0139] in: Indicates proximity to power sensors With power sensor s The location association information between them;

[0140] Indicates proximity to power sensors With power sensor s the distance between deployed positions; Indicates proximity to power sensors With power sensor s The transmission weight of the power grid line between the corresponding nodes; the time series information extraction layer extracts the time series information of the power sensor data and the associated adjacent power sensor data, where the power sensor data H s The result of extracting the timing information is:

[0141]

[0142] Where: W 1 represents the time series information convolution matrix, * represents the convolution operator; F2(H s ) represents the power sensor data H s The information fusion layer fuses the position association information and the time series information to form a feature vector that represents the abnormal state of the power sensor data:

[0143]

[0144] Among them: F3(H s ) represents the power sensor data H s The fault recognition layer converts the feature vector F3(H s) is converted into the failure probability P(H s ), if the failure probability P(H s ) is higher than the preset threshold, it means that the power sensor data H s There is a power failure, where the failure probability P(H s ) is:

[0145]

[0146] Where: W 2 Represents the probability mapping convolution matrix; the access device locates the power sensor that has identified the power fault according to the transmission direction of the power sensor data, and uses the node location of the power sensor as the location of the power fault area. This solution uses the deployed access device to gather the power sensor data in the target area, builds a power fault identification model, and performs information fusion based on the location information of the neighboring power sensors and the timing information of the power sensor data to obtain the feature vector that characterizes the abnormal state of the power sensor data, and identifies the power fault for the power sensor data gathered in the target area. The access device locates the power sensor that has identified the power fault according to the transmission direction of the power sensor data, identifies the power sensor data with the power fault, and uses the node location of the power sensor as the location of the power fault area, realizing accurate identification of the power fault location based on the power sensor data information and Beidou positioning technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0147] Figure 1 A flowchart of a method for accurately identifying power fault locations based on Beidou technology provided by an embodiment of the present invention;

[0148] Figure 2 A schematic diagram of the structure of an electronic device for implementing a method for accurately identifying power fault locations based on Beidou technology provided in one embodiment of the present invention.

[0149] In the figure: 1 electronic device, 10 processor, 11 memory, 12 program, 13 communication interface.

[0150] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0151] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0152] The embodiment of the present application provides a method for accurately identifying the location of power faults based on Beidou technology. The execution subject of the method for accurately identifying the location of power faults based on Beidou technology includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided in the embodiment of the present application. In other words, the method for accurately identifying the location of power faults based on Beidou technology can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0153] Embodiment 1:

[0154] S1: Collect power grid line data and fill missing values ​​according to the context to obtain preprocessed power grid line data, build the target area power network topology based on the preprocessed power grid line data, and deploy power sensors in the target area power network topology.

[0155] In the step S1, the power grid line data is collected and missing values ​​are filled according to the context to obtain preprocessed power grid line data, including:

[0156] Collect power grid line data, where the collected power grid line data is in the form of:

[0157] {L n |n∈[1,N]}

[0158] L n =(L n (1),L n (2),...,L n (m),...,L n (M))

[0159] in:

[0160] L n represents the collected grid line data of the nth grid line, N represents the total number of grid lines; the grid line is the transmission line between transmission stations;

[0161] L n (1),L n (2),...,L n (m),...,L n (M) represents the current value collected by the nth power line at M collection moments, L n (m) represents the current value of the nth power line at the mth acquisition moment, m∈[1,M];

[0162] Fill in missing values ​​in the power grid line data according to the context, where the power grid line data L nMissing value L n The filling process of (q) is:

[0163] S11: Initialize and generate U missing value filling results, where the u-th missing value filling result generated by initialization is:

[0164] S12: Set the current iteration number of the missing value filling result to t, the initial value of t is 0, and the maximum value is Max, then the missing value filling result The result of the tth iteration is

[0165] S13: Fill the missing value filling result into the power grid line data L n The corresponding power grid line data is obtained, where the missing values ​​are filled in. The power grid line data is

[0166] Calculate power grid line data With the grid line data L n The error between Fitness:

[0167]

[0168] in:

[0169] Indicates the missing value filling result Adaptability;

[0170] μ n Indicates the power grid line data L n The mean of Indicates power grid line data The mean of

[0171] σ n Indicates the power grid line data L n The standard deviation of Indicates power grid line data The standard deviation of

[0172] ∈1,∈2 represent adjustment coefficients;

[0173] S14: Select the missing value filling result with the lowest fitness among the U missing value filling results as the optimal filling result obtained in the tth iteration

[0174] S15: Iterate the missing value filling results, where the missing value filling results The iteration formula is:

[0175]

[0176] in:

[0177] rand(0,1) represents a random number between 0 and 1;

[0178] c1, c2 represent control parameters;

[0179] exp(·) represents an exponential function with a natural constant as the base;

[0180] express The missing value filling result with the smallest fitness;

[0181] S16: Let t = t + 1, return to step S13, until the maximum number of iterations is reached, and select the missing value filling result with the lowest fitness among the U missing value filling results obtained in the final iteration, and use the selected missing value filling result as the power grid line data L n Missing value L n (q) fill value, get the power grid line data L n The preprocessing result of ′ n ;

[0182] Construct the target area power network topology based on the preprocessed power grid line data.

[0183] The method of constructing a target area power network topology based on the preprocessed power grid line data includes:

[0184] The target area power network topology is constructed based on the preprocessed power grid line data, where the construction process of the target area power network topology is as follows:

[0185] The transmission stations in the power network are used as nodes in the target area power network, and the power grid lines between the transmission stations are used as edges between nodes to form the target area power network topology:

[0186] G=(E,V)

[0187] E={e s |s∈[1,S]}

[0188]

[0189] in:

[0190] G represents the target area power network topology;

[0191] E represents the node set in the target area power network, e s represents the sth node in the target area power network, S represents the total number of nodes in the target area power network;

[0192] V represents the edge set in the target area power network, Represents node e s With e i The edge between

[0193] Calculate the weights of the edges in the target area power network topology, where the edges The transmission weight is:

[0194]

[0195] in:

[0196] Represents edge The transmission weight of

[0197] mean(·) means calculating the mean. express Corresponding pre-processed power grid line data;

[0198] Deploy power sensors in the target area power network topology.

[0199] The deploying of power sensors in the target area power network topology includes:

[0200] The power sensors are deployed at the locations of S nodes in the target area power network topology to form a power sensor set:

[0201] h={h s |s∈[1,S]}

[0202] in:

[0203] h represents the power sensor set, h s represents the sth node e deployed in the target area power network topology s Power sensors at the location.

[0204] S2: Construct a regional convergence access device deployment optimization objective function, where the regional convergence access device deployment optimization objective function takes the access device deployment location as the independent variable, and takes the access device coverage and coverage accuracy of the target area power network topology as the optimization target.

[0205] The step S2 constructs the regional convergence access device deployment optimization objective function, including:

[0206] The regional convergence access equipment deployment optimization objective function is constructed, where the regional convergence access equipment deployment optimization objective function takes the access equipment deployment location as the independent variable, and takes the access equipment's coverage of the target area power network topology and coverage accuracy as the optimization target. The access equipment is a regional data convergence access point with Beidou positioning function, which is responsible for aggregating and transmitting power sensor data within the coverage area. The constructed regional convergence access equipment deployment optimization objective function is:

[0207]

[0208] θ=(θ1,θ2,...,θ k ,...,θ K )

[0209] in:

[0210] F(·) represents the optimization objective function of regional aggregation access equipment deployment, θ represents the access equipment deployment location plan, θ1,θ2,...,θ k ,...,θ K represents the deployment positions of K access devices in the access device deployment position scheme θ, and F(θ) represents the objective function value of the access device deployment position scheme θ;

[0211] W s Represents node e s The transmission weight, Ω s Represents the target area power network topology with node e s is the set of edges with endpoints, c∈Ω s , c represents the set Ω s For any edge in , w(c) represents the transmission weight of edge c; Used to indicate the coverage accuracy of access equipment;

[0212] α(θ k ,e s ) means deployed in θ k The access device of node e s The coverage of max{α(θ k ,e s )|k∈[1,K]} represents the access device deployment location solution θ for node e s coverage;

[0213] R represents the aggregation radius of the access device;

[0214] dis(θ k ,e s ) represents θ k With node e s The distance between the locations.

[0215] S3: Taking maximizing the coverage and coverage accuracy of the access equipment to the target area power network topology as the optimization goal, the regional aggregation access equipment deployment optimization objective function is optimized and solved to obtain the optimal access equipment deployment location plan and deploy it.

[0216] In the step S3, the optimization objective is to maximize the coverage and coverage accuracy of the access equipment to the target area power network topology, and to optimize and solve the regional aggregation access equipment deployment optimization objective function, including:

[0217] Taking maximizing the coverage and coverage accuracy of the access equipment to the target area power network topology as the optimization goal, the regional convergence access equipment deployment optimization objective function is optimized and solved. The optimization solution process of the regional convergence access equipment deployment optimization objective function is as follows:

[0218] S31: Initialize and generate the positions of D dung beetles, wherein the initialization generation formula for the position of the dth dung beetle is:

[0219]

[0220] in:

[0221] represents the initialization position of the d-th dung beetle, d∈[1,D]; in the embodiment of the present invention, is a randomly generated position;

[0222] S32: Set the current iteration number of the dung beetle position to z, and the maximum iteration number to Z, then the zth iteration position of the d dung beetle is

[0223]

[0224] in:

[0225] The k-th dimension coordinate representing the iteration position corresponds to the deployment position of the k-th access device;

[0226] S33: Using the iteration position of the dung beetle as the input value of the regional convergence access device deployment optimization objective function to obtain the objective function value corresponding to the iteration position, where the iteration position The corresponding objective function value is

[0227] Sort the D groups of iteration positions obtained in the zth iteration in descending order of the objective function value, and evenly divide the sorting results to obtain 3 groups of iteration position sequences, where each group of iteration position sequences has D / 3 groups of iteration positions;

[0228] S34: using the iteration position in the first group of iteration position sequences as the position of the rolling ball dung beetle, and performing rolling ball dung beetle position iteration:

[0229]

[0230] in:

[0231] σ represents the deflection coefficient, and σ is set to 0.2;

[0232] It represents the iteration position with the minimum objective function value among the D group iteration positions obtained in the z-th iteration;

[0233] The iteration position in the second set of iteration position sequence is used as the position of the dancing dung beetle, and the dancing dung beetle position iteration is performed:

[0234]

[0235] in:

[0236] τ represents a random angle in (0,π / 2);

[0237] The iterative position in the third set of iterative position sequences is used as the position for breeding dung beetles, and the breeding dung beetle position is iterated:

[0238]

[0239] in:

[0240] It represents the iteration position with the largest objective function value among the D group iteration positions obtained in the z-th iteration;

[0241] S35: Let d=d+1, and return to step S33 until the maximum number of iterations is reached, and the dung beetle position with the largest objective function value after the maximum number of iterations is reached is used as the optimal access device deployment position solution, and the access device is deployed.

[0242] S4: Use the deployed access devices to aggregate the power sensor data of the target area, build a power fault identification model, identify the power fault on the power sensor data aggregated in the target area, identify the power sensor data with power faults, use the access devices to locate the power sensors that have identified the power faults, and obtain the location of the power fault area.

[0243] In the step S4, the power sensor data of the target area is aggregated by the deployed access device to build a power fault identification model, including:

[0244] The deployed access devices are used to aggregate the power sensor data in the target area and upload the data sensed by the power sensors to the access devices.s The power sensor data sensed is H s :

[0245] H s =(H s (1),H s (2),...,H s (a),...,H s (A)

[0246] in:

[0247] H s (1),H s (2),...,H s (a),...,H s (A) represents the power sensor h s The perceived node e s The power transmission sequence,

[0248] H s (a) represents node e s The amount of power transmitted at the ath transmission moment, A represents the number of transmission moments, a∈[1,A];

[0249] Constructing a power fault identification model, wherein the power fault identification model includes an input layer, a location association information extraction layer, a time sequence information extraction layer, an information fusion layer, and a fault identification layer;

[0250] The input layer is used to receive the power sensor data to be identified and the associated adjacent power sensor data, wherein there is an uninterrupted power grid line between the power sensor and the node where the adjacent power sensor is deployed, and the adjacent power sensor data is the data sensed by the adjacent power sensor. The position association information extraction layer is used to extract the position association information between the adjacent power sensor and the power sensor to be identified. The timing information extraction layer is used to extract the timing information of the power sensor data and the associated adjacent power sensor data respectively. The information fusion layer is used to fuse the position association information and the timing information to form a feature vector that characterizes the abnormal state of the power sensor data. The fault identification layer is used to convert the feature vector into a fault probability. If the fault probability is higher than a preset threshold, it indicates that there is a power fault in the power sensor data.

[0251] Power fault identification is performed on the power sensor data gathered in the target area to identify the power sensor data with power faults.

[0252] The method of using the power fault identification model to identify power faults in the power sensor data collected in the target area, and identifying the power sensor data with power faults, includes:

[0253] The power fault identification model is used to identify power faults in the power sensor data collected in the target area, where the power sensor data H s The power fault identification process is as follows:

[0254] S41: The input layer receives the power sensor data H to be identified s and associated proximity power sensor data Where B represents the power sensor data H s The total number of nearby power sensor data, Indicates power sensor data H s The bth neighboring power sensor data, b∈[1,B];

[0255] S42: The location association information extraction layer extracts the neighboring power sensor data corresponding to the neighboring power sensor and the power sensor h s The location association information between the adjacent power sensor data Corresponding nearby power sensor With power sensor s The location association information between them is:

[0256]

[0257] in:

[0258] Indicates proximity to power sensors With power sensor s The location association information between them;

[0259] Indicates proximity to power sensors With power sensor s the distance between deployed positions;

[0260] Indicates proximity to power sensors With power sensor s The transmission weight of the power grid lines between the corresponding nodes;

[0261] S43: The timing information extraction layer extracts timing information of the power sensor data and the associated adjacent power sensor data respectively, wherein the power sensor data H s The result of extracting the timing information is:

[0262]

[0263] in:

[0264] W 1represents the time series information convolution matrix, * represents the convolution operator;

[0265] F2(H s ) represents the power sensor data H s The time series information extraction results;

[0266] S44: The information fusion layer fuses the position association information and the time sequence information to form a feature vector representing the abnormal state of the power sensor data:

[0267]

[0268] in:

[0269] F3(H s ) represents the power sensor data H s The feature vector of abnormal state;

[0270] S45: The fault identification layer transforms the feature vector F s (H s ) is converted into the failure probability P(H s ), if the failure probability P(H s ) is higher than the preset threshold, it means that the power sensor data H s There is a power failure, where the failure probability P(H s ) is:

[0271]

[0272] in:

[0273] W 2 Represents the probability mapping convolution matrix;

[0274] The access device locates the power sensor that has identified the power fault according to the transmission direction of the power sensor data, and uses the node location of the power sensor as the power fault area location.

[0275] Embodiment 2:

[0276] like Figure 2 , is a schematic diagram of the structure of an electronic device for implementing a method for accurately identifying power fault locations based on Beidou technology provided by an embodiment of the present invention.

[0277] The electronic device 1 may include a processor 10 , a memory 11 , a communication interface 13 , and a bus, and may also include a computer program stored in the memory 11 and executable on the processor 10 , such as a program 12 .

[0278] The memory 11 includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a mobile hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. The memory 11 may be an internal storage unit of the electronic device 1 in some embodiments, such as a mobile hard disk of the electronic device 1. The memory 11 may also be an external storage device of the electronic device 1 in other embodiments, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 1. Further, the memory 11 may also include both an internal storage unit of the electronic device 1 and an external storage device. The memory 11 may be used not only to store application software and various types of data installed in the electronic device 1, such as the code of the program 12, etc., but also to temporarily store data that has been output or is to be output.

[0279] The processor 10 may be composed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or a plurality of integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips, etc. The processor 10 is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect the various components of the entire electronic device, and executes or executes the program or module stored in the memory 11 (a program 12 for realizing accurate identification of power fault locations based on Beidou technology, etc.), and calls the data stored in the memory 11 to execute various functions of the electronic device 1 and process data.

[0280] The communication interface 13 may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices, and to achieve connection and communication between internal components of the electronic device.

[0281] The bus may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0282] Figure 2 Only an electronic device with components is shown, and those skilled in the art will understand that Figure 2 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0283] For example, although not shown, the electronic device 1 may also include a power source (such as a battery) for supplying power to each component. Preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include any components such as one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, etc. The electronic device 1 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0284] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visual user interface.

[0285] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0286] The program 12 stored in the memory 11 in the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can achieve:

[0287] Collecting power grid line data and filling missing values ​​according to context to obtain preprocessed power grid line data, constructing a target area power network topology based on the preprocessed power grid line data, and deploying power sensors in the target area power network topology;

[0288] Construct the regional aggregation access equipment deployment optimization objective function;

[0289] Taking maximizing the coverage and coverage accuracy of the access equipment to the target area power network topology as the optimization goal, the regional aggregation access equipment deployment optimization objective function is optimized and solved, and the optimal access equipment deployment location plan is obtained and deployed;

[0290] The deployed access devices are used to aggregate the power sensor data of the target area, and a power fault identification model is constructed. Power faults are identified on the power sensor data aggregated in the target area, and power sensor data with power faults are identified. The access devices are used to locate the power sensors that have identified power faults, and the location of the power fault area is obtained.

[0291] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to Figure 1 to Figure 2 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0292] It should be noted that the serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. And the terms "including", "comprising" or any other variants thereof in this article are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "including a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0293] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0294] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for accurately identifying the location of power faults based on Beidou technology, characterized in that: The method comprises: S1: Collect power grid line data and fill in missing values ​​to obtain preprocessed power grid line data, build the target area power network topology based on the preprocessed power grid line data, and deploy power sensors in the target area power network topology; S2: Constructing the regional convergence access equipment deployment optimization objective function, wherein the regional convergence access equipment deployment optimization objective function takes the access equipment deployment location as the independent variable, and takes the access equipment's coverage of the target area power network topology and coverage accuracy as the optimization target; S3: Taking maximizing the coverage and coverage accuracy of the access equipment to the target area power network topology as the optimization goal, the regional aggregation access equipment deployment optimization objective function is optimized and solved to obtain the optimal access equipment deployment location plan and deploy it; S4: Use the deployed access devices to aggregate the power sensor data of the target area, build a power fault identification model, identify the power fault on the power sensor data aggregated in the target area, identify the power sensor data with power faults, use the access devices to locate the power sensors that have identified the power faults, and obtain the location of the power fault area.

2. The method for accurately identifying the location of power faults based on Beidou technology as claimed in claim 1, characterized in that: In the step S1, the power grid line data is collected and missing values ​​are filled to obtain preprocessed power grid line data, including: Collect power grid line data, where the collected power grid line data is in the form of: {L n |n∈[1,N]} L n =(L n (1),L n (2),...,L n (m),...,L n (M)) in: L n represents the collected grid line data of the nth grid line, N represents the total number of grid lines; the grid line is the transmission line between transmission stations; L n (1),L n (2),...,L n (m),...,L n (M) represents the current value collected by the nth power line at M collection moments, L n (m) represents the current value of the nth power line at the mth acquisition moment, m∈[1,M]; Fill in missing values ​​in the power grid line data according to the context, where the power grid line data L n Missing value L n The filling process of (q) is: S11: Initialize and generate U missing value filling results, where the u-th missing value filling result generated by initialization is: S12: Set the current iteration number of the missing value filling result to t, the initial value of t is 0, and the maximum value is Max, then the missing value filling result The result of the tth iteration is S13: Fill the missing value filling result into the power grid line data L n The corresponding power grid line data is obtained, where the missing values ​​are filled in. The power grid line data is Calculate power grid line data With the grid line data L n The error between Fitness S14: Select the missing value filling result with the lowest fitness among the U missing value filling results as the optimal filling result obtained in the tth iteration S15: Iterate the missing value filling results; S16: Let t = t + 1, return to step S13, until the maximum number of iterations is reached, and select the missing value filling result with the lowest fitness among the U missing value filling results obtained in the final iteration, and use the selected missing value filling result as the power grid line data L n Missing value L n (q) fill value, get the power grid line data L n The preprocessing result L ′ n , q∈[1,N]; Construct the target area power network topology based on the preprocessed power grid line data.

3. A method for accurately identifying power fault locations based on Beidou technology as described in claim 2, characterized in that: The method of constructing a target area power network topology based on the preprocessed power grid line data includes: The target area power network topology is constructed based on the preprocessed power grid line data, where the construction process of the target area power network topology is as follows: The transmission stations in the power network are used as nodes in the target area power network, and the power grid lines between the transmission stations are used as edges between nodes to form the target area power network topology: G=(E,V) E={e s |s∈[1,S]} in: G represents the target area power network topology; E represents the node set in the target area power network, e s represents the sth node in the target area power network, S represents the total number of nodes in the target area power network; V represents the edge set in the target area power network, Represents node e s With e i The edge between Calculate the weights of the edges in the target area power network topology, where the edges The transmission weight is: in: Represents edge The transmission weight of mean(·) means calculating the mean. express Corresponding pre-processed power grid line data; Deploy power sensors in the target area power network topology.

4. The method for accurately identifying the location of power faults based on Beidou technology as claimed in claim 3 is characterized in that: The deploying of power sensors in the target area power network topology includes: The power sensors are deployed at the locations of S nodes in the target area power network topology to form a power sensor set: h={h s |s∈[1,S]} in: h represents the power sensor set, h s represents the sth node e deployed in the target area power network topology s Power sensors at the location.

5. The method for accurately identifying the location of power faults based on Beidou technology as claimed in claim 1, characterized in that: The step S2 constructs the regional convergence access device deployment optimization objective function, including: The regional convergence access equipment deployment optimization objective function is constructed, where the regional convergence access equipment deployment optimization objective function takes the access equipment deployment location as the independent variable, and takes the access equipment's coverage of the target area power network topology and coverage accuracy as the optimization target. The access equipment is a regional data convergence access point with Beidou positioning function, which is responsible for aggregating and transmitting power sensor data within the coverage area. The constructed regional convergence access equipment deployment optimization objective function is: θ=(θ1,θ2,...,θ k ,...,θ K ) in: F(·) represents the optimization objective function of regional aggregation access equipment deployment, θ represents the access equipment deployment location plan, θ1,θ2,...,θ k ,...,θ K represents the deployment positions of K access devices in the access device deployment position scheme θ, and F(θ) represents the objective function value of the access device deployment position scheme θ; W s Represents node e s The transmission weight, Ω s Represents the target area power network topology with node e s is the set of edges with endpoints, c∈Ω s , c represents the set Ω s For any edge in , w(c) represents the transmission weight of edge c; Used to indicate the coverage accuracy of access equipment; α(θ k ,e s ) means deployed in θ k The access device of node e s The coverage of max{α(θ k ,e s )|k∈[1,K]} represents the access device deployment location solution θ for node e s coverage; R represents the aggregation radius of the access device; dis(θ k ,e s ) represents θ k With node e s The distance between the locations.

6. A method for accurately identifying power fault locations based on Beidou technology as claimed in claim 5, characterized in that: In the step S3, the optimization objective is to maximize the coverage and coverage accuracy of the access equipment to the target area power network topology, and to optimize and solve the regional aggregation access equipment deployment optimization objective function, including: Taking maximizing the coverage and coverage accuracy of the access equipment to the target area power network topology as the optimization goal, the regional convergence access equipment deployment optimization objective function is optimized and solved. The optimization solution process of the regional convergence access equipment deployment optimization objective function is as follows: S31: Initialize and generate the positions of D dung beetles, wherein the initialization generation formula for the position of the dth dung beetle is: in: represents the initial position of the d-th dung beetle, d∈[1,D]; S32: Set the current iteration number of the dung beetle position to z, and the maximum iteration number to Z, then the zth iteration position of the d dung beetle is in: The k-th dimension coordinate representing the iteration position corresponds to the deployment position of the k-th access device; S33: Using the iteration position of the dung beetle as the input value of the regional convergence access device deployment optimization objective function to obtain the objective function value corresponding to the iteration position, where the iteration position The corresponding objective function value is Sort the D groups of iteration positions obtained in the zth iteration in descending order of the objective function value, and evenly divide the sorting results to obtain 3 groups of iteration position sequences, where each group of iteration position sequences has D / 3 groups of iteration positions; S34: using the iteration position in the first group of iteration position sequences as the position of the rolling ball dung beetle, and performing rolling ball dung beetle position iteration; The iterative positions in the second group of iterative position sequences are used as the positions of the dancing dung beetles, and the positions of the dancing dung beetles are iterated; The iterative positions in the third group of iterative position sequences are used as positions for breeding dung beetles, and the iterative positions for breeding dung beetles are performed; S35: Let d=d+1, and return to step S33 until the maximum number of iterations is reached, and the dung beetle position with the largest objective function value after the maximum number of iterations is reached is used as the optimal access device deployment position solution, and the access device is deployed.

7. A method for accurately identifying power fault locations based on Beidou technology as claimed in claim 6, characterized in that: In the step S4, the power sensor data of the target area is aggregated by the deployed access device to build a power fault identification model, including: The deployed access devices are used to aggregate the power sensor data in the target area and upload the data sensed by the power sensors to the access devices. s The power sensor data sensed is H s : H s =(H s (1),H s (2),...,H s (a),...,H s (A)) in: H s (1),H s (2),...,H s (a),...,H s (A) represents the power sensor h s The perceived node e s The power transmission sequence, H s (a) represents node e s The amount of power transmitted at the ath transmission moment, A represents the number of transmission moments, a∈[1,A]; Constructing a power fault identification model, wherein the power fault identification model includes an input layer, a location association information extraction layer, a time sequence information extraction layer, an information fusion layer, and a fault identification layer; The input layer is used to receive the power sensor data to be identified and the associated adjacent power sensor data, wherein there is an uninterrupted power grid line between the power sensor and the node where the adjacent power sensor is deployed, and the adjacent power sensor data is the data sensed by the adjacent power sensor. The position association information extraction layer is used to extract the position association information between the adjacent power sensor and the power sensor to be identified. The timing information extraction layer is used to extract the timing information of the power sensor data and the associated adjacent power sensor data respectively. The information fusion layer is used to fuse the position association information and the timing information to form a feature vector that characterizes the abnormal state of the power sensor data. The fault identification layer is used to convert the feature vector into a fault probability. If the fault probability is higher than a preset threshold, it indicates that there is a power fault in the power sensor data. Power fault identification is performed on the power sensor data gathered in the target area to identify the power sensor data with power faults.

8. The method for accurately identifying the location of power faults based on Beidou technology as claimed in claim 7, characterized in that: The method of using the power fault identification model to identify power faults in the power sensor data collected in the target area, and identifying the power sensor data with power faults, includes: The power fault identification model is used to identify power faults in the power sensor data collected in the target area, where the power sensor data H s The power fault identification process is as follows: S41: The input layer receives the power sensor data H to be identified s and associated proximity power sensor data Where B represents the power sensor data H s The total number of nearby power sensor data, Indicates power sensor data H s The bth neighboring power sensor data, b∈[1,B]; S42: The location association information extraction layer extracts the neighboring power sensor data corresponding to the neighboring power sensor and the power sensor h s The location association information between the adjacent power sensor data Corresponding nearby power sensor With power sensor s The location association information between them is: in: Indicates proximity to power sensors With power sensor s The location association information between them; Indicates proximity to power sensors With power sensor s the distance between deployed positions; Indicates proximity to power sensors With power sensor s The transmission weight of the power grid lines between the corresponding nodes; S43: The timing information extraction layer extracts timing information of the power sensor data and the associated adjacent power sensor data respectively, wherein the power sensor data H s The result of extracting the timing information is: in: W 1 represents the time series information convolution matrix, * represents the convolution operator; F2(H s ) represents the power sensor data H s The time series information extraction results; S44: The information fusion layer fuses the position association information and the time sequence information to form a feature vector representing the abnormal state of the power sensor data: in: F3(H s ) represents the power sensor data H s The feature vector of abnormal state; S45: The fault identification layer transforms the feature vector F3(H s ) is converted into the failure probability P(H s ), if the failure probability P(H s ) is higher than the preset threshold, it means that the power sensor data H s There is a power failure, where the failure probability P(H s ) is: in: W 2 Represents the probability mapping convolution matrix; The access device locates the power sensor that has identified the power fault according to the transmission direction of the power sensor data, and uses the node location of the power sensor as the power fault area location.

Images