GIS partial discharge optimal position matching method, system and industrial control computer

Abstract

The invention discloses a method and system for matching the optimal position of local discharge in a GIS. The method comprises the following steps: roughly locating a local discharge source based on time-domain full-waveform data of local discharge in the GIS to obtain a preliminary discharge intensity and a preliminary discharge position of the local discharge source; injecting the preliminary discharge intensity and the preliminary discharge position of the local discharge source into a simulator for simulation enhancement calculation to obtain a simulation enhancement result of the local discharge source, wherein the simulation enhancement result includes a simulation discharge position and a simulation discharge intensity; iteratively searching the measured discharge intensity calculated by using the time-domain full-waveform data collected in real time and the simulation enhancement result to obtain the actual discharge position of the local discharge source; when local discharge occurs in the GIS, the invention takes the discharge source calculated initially as the center, proposes a heuristic fast iterative search algorithm for optimal matching of measured values ​​in its neighborhood, and realizes real-time high-precision positioning of the discharge source.

Description

Technical Field

[0001] The present invention relates to the technical field of power equipment detection, and in particular to a GIS partial discharge optimal position matching method, system and industrial control computer. Background Art

[0002] With the widespread application of high-voltage gas insulated switchgear (GIS), insulation aging, loose components and other equipment failures have caused frequent equipment failures, which have seriously threatened the safe operation of substations at all levels. Since GIS equipment is a fully enclosed structure, when it fails, the affected area is extremely large and the maintenance cost is high. In GIS equipment, partial discharge is a common fault phenomenon that can cause equipment performance degradation, damage and even accidents. Accurately and quickly locating partial discharge signals inside GIS is crucial for equipment health status monitoring and fault diagnosis.

[0003] In the related technology, discharge detection is generally performed based on the high-frequency electromagnetic wave signal (Ultra High Frequency, UHF) inside the GIS. For example, the patent application document with publication number CN116662826A uses the KELM model to compare the measured UHF signal fingerprint with the local discharge positioning fingerprint library to locate the local discharge source; this scheme forms a fingerprint library by extracting features from local discharge signals of different locations and types in advance, and trains the KELM model, and then inputs the measured UHF signal fingerprint into the trained KELM model in order to locate the local discharge. However, this model relies on high-quality local discharge and does not perform an iterative optimization process. In the case of a complex operating environment for power equipment, it will affect the accuracy of positioning. The patent application document with publication number CN117706306A adopts a local discharge positioning technology based on electromagnetic time inversion to locate the local discharge source; this solution collects the ultra-high frequency electromagnetic wave pulse signal generated by the local discharge through an ultra-high frequency sensor, performs a time reversal operation on the pulse signal, and injects it into the GIS three-dimensional electromagnetic field model to obtain the distribution change of the internal electric field strength over time. Based on the distribution of the electric field strength, the local discharge position is determined. However, the electromagnetic wave is affected by various factors such as equipment structure, material properties, environmental conditions, etc. during the propagation process, and may encounter multi-path effects such as reflection and refraction. These changes may affect the signal convergence effect during the time inversion process. The patent application document with publication number CN114660448A adopts a fuzzy neural network optimized by a crowd search algorithm, performs pattern matching between the simulated fingerprint and the experimentally measured partial discharge fingerprint library, and obtains the corresponding spatial coordinates of the partial discharge source; by collecting the partial discharge simulation signal generated by the partial discharge, a partial discharge positioning simulation fingerprint library is constructed, the simulated fingerprint is normalized, and its characteristic parameters are extracted as the data set of training and test samples, and a matching model is constructed based on the fuzzy neural network theory. During the training process, the crowd search algorithm is used to correct the weights, and the training and test samples are used as input to train the constructed fuzzy neural network matching model. The FNN model has a complex structure and a large number of rules, which may lead to poor model interpretability and difficulty in understanding and maintenance. The performance of CSA and FNN depends on the selection and tuning of initial parameters. Improper parameter selection may cause the optimization process to converge slowly or fall into a local optimum.

[0004] In the doctoral dissertation "Research on Fault Location Method of Power Transformer Based on Multi-ant Colony Algorithm, Wei Luyuan, China University of Mining and Technology", an improved ant colony algorithm suitable for fault tree search was proposed. In the process of transformer fault detection, one detection method can detect several fault types, and one fault type can also use several detection methods. When using the ant colony algorithm to implement fault search, different detection methods are matched with multiple types of ant colonies. Ants select paths based on pheromone concentration and heuristic rules, and adjust the pheromone update strategy according to the structure and node status of the fault tree to achieve more accurate fault location. However, the fault tree itself is very complex, and the logical relationship and state transition between nodes require precise modeling and analysis, which increases the difficulty of algorithm implementation and debugging. The master's thesis "Research and Development of Intelligent Partial Discharge Inspection System, Li Ning, Hebei University of Technology" proposed to use an improved algorithm combining A* algorithm and ant colony algorithm to solve the path planning problem of substation partial discharge inspection. The partial discharge of the substation is located by an intelligent inspection car, and the ant colony algorithm is used to solve the path planning problem of the inspection car, which reduces the occurrence of repeated paths and enables the inspection car to detect more equipment and lines at a certain cost. By selecting RFID radio frequency technology as the main technical means, the signal loss model is analyzed, the least squares method is used to obtain the ranging function based on RSSI, and the three-sided positioning method is used to obtain the positioning information of the inspection car.

[0005] In summary, when the above-mentioned related technologies use UHF signals for discharge detection, the multiple reflections of UHF signals in different propagation paths in the complex topological structure of GIS will affect the positioning accuracy. The multipath effect will cause different time delays for the signal to reach the UHF sensor, thereby introducing positioning errors and making it impossible for the positioning system to accurately identify the location of the discharge source. Summary of the invention

[0006] The technical problem to be solved by the present invention is how to achieve high-precision positioning of partial discharge signals.

[0007] The present invention solves the above technical problems by the following technical means:

[0008] On the one hand, the present invention proposes a GIS partial discharge optimal position matching method, the method comprising:

[0009] Based on the full waveform data of GIS partial discharge in the time domain, the partial discharge source is roughly located to obtain the preliminary discharge intensity and preliminary discharge position of the partial discharge source;

[0010] Injecting the preliminary discharge intensity and preliminary discharge position of the local discharge source into the simulator for simulation enhancement calculation to obtain a simulation enhancement result of the local discharge source, wherein the simulation enhancement result includes a simulation discharge position and a simulation discharge intensity;

[0011] The measured discharge intensity calculated by the time domain full waveform data collected in real time and the simulation enhancement result are iteratively searched to obtain the actual discharge position of the partial discharge source.

[0012] Furthermore, the partial discharge source is roughly located based on the time domain full waveform data of the GIS partial discharge to obtain the preliminary discharge intensity and preliminary discharge position of the partial discharge source, including:

[0013] Calculate the initial discharge location of the PD source based on the full waveform data of GIS PD in the time domain , the formula is:

[0014]

[0015]

[0016] In the formula, Indicates The distance from the sensor to the local discharge source, It represents the propagation speed of electromagnetic waves in the current propagation medium. Indicates the partial discharge signal propagating to the reference sensor The time required, Indicates the partial discharge signal propagating to the reference sensor and propagates to the sensor The time difference, The coordinates of the UHF sensors are , ;

[0017] Calculate the initial discharge intensity of the local discharge source based on the full waveform data of GIS local discharge in the time domain .

[0018] Furthermore, the preliminary discharge intensity and preliminary discharge position of the local discharge source are injected into the simulator for simulation calculation to obtain the simulation result of the local discharge source, including:

[0019] Inject the initial discharge intensity and initial discharge position of the local discharge source into the simulator to simulate the local discharge phenomenon of GIS equipment;

[0020] The simulator is used to perform simulation enhancement calculations to obtain the simulated discharge position and simulated discharge intensity of the local discharge source.

[0021] Furthermore, the iterative search of the measured discharge intensity calculated by the time domain full waveform data collected in real time and the simulation enhancement result to obtain the actual discharge position of the partial discharge source includes:

[0022] According to the simulated discharge position and simulated discharge intensity at each moment, the state transition model corresponding to the process of the ant colony algorithm exploring the actual discharge position of the local discharge source with the initial discharge position of the local discharge source as the center is reconstructed using the Markov decision model;

[0023] Generate a heuristic metric using a heuristic learner based on a graph neural network, and transform the state transition model into a location exploration model that is influenced by the heuristic metric and requires graph traversal;

[0024] The position exploration model is iteratively searched using a neural-guided perturbation interleaved local search algorithm to obtain the actual discharge position of the local discharge source.

[0025] Furthermore, the state transition model corresponding to the process of reconstructing the ant colony algorithm to explore the actual discharge position of the local discharge source with the preliminary discharge position of the local discharge source as the center based on the simulated discharge position and the simulated discharge intensity at each moment, includes:

[0026] According to the simulated discharge position and simulated discharge intensity at each moment, the state space corresponding to the actual discharge position of the local discharge source is constructed in the simulator :

[0027]

[0028]

[0029] Where: Indicates that in the simulator The local discharge source node is explored at The corresponding status, Indicates The simulated discharge position at each moment, Indicates The simulated discharge intensity at each moment, It represents the initial discharge intensity of the local discharge source obtained based on the full waveform data of GIS local discharge in the time domain and initial discharge position As the initial state, represents the exploration cycle, ;

[0030] Constructing the action space corresponding to the actual discharge location of the PD source in the simulator :

[0031]

[0032] Where: Indicates The action of the moment Greedy strategy is the standard choice of action ;

[0033] The state transition model corresponding to the process of the reconstructed ant colony algorithm exploring the actual discharge position of the local discharge source with the initial discharge position of the local discharge source as the center is:

[0034]

[0035] Where: Indicates in action Next node To Node The probability of transfer, Indicates that in the simulator The local discharge source node is explored at The corresponding state, Represented by the node Transfer to Node The corresponding pheromone concentration, Represented by the node Transfer to Node The corresponding heuristic function is, Represented by the node Transfer to the included Nodes in The corresponding pheromone concentration, Represented by the node Transfer to the included Nodes in The corresponding heuristic function is, and Represent the pheromone concentration and the weight of the heuristic function, Represents the set of non-uniform grid local discharge source nodes that are allowed to be selected at the next moment.

[0036] Furthermore, by the node To Node The reward function for the transfer is , Indicates The measured discharge intensity is obtained by processing the full waveform in the time domain collected by the sensor at each moment. Indicates The partial discharge intensity obtained by simulation at each moment.

[0037] Furthermore, the heuristic metric is generated by using a heuristic learner based on a graph neural network, and the state transition model is converted into a location exploration model affected by the heuristic metric and requiring graph traversal, including:

[0038] The graph neural network The layer is The node and Edge features connecting nodes Mapping to heuristic metrics ;

[0039] The state transition model is transformed into a state transition model that is influenced by the heuristic metric and requires The location exploration model of step graph traversal is:

[0040]

[0041] Where: represents the location exploration model, Indicates that in the simulator The local discharge source node is explored at The corresponding status, Indicates that in the simulator The local discharge source node is explored at The corresponding status, represents the exploration cycle, .

[0042] Furthermore, the graph neural network The propagation characteristics of the layer are:

[0043]

[0044]

[0045] Where: Indicates Layer The characteristics of the nodes, Indicates Layer The characteristics of the nodes, Indicates Layer The characteristics of the nodes, Indicates The layer is The node and The edge features connecting the nodes are Indicates The layer is The node and The edge features connecting the nodes are , , , as well as Indicates The learnable parameters of the layer, represents the activation function, represents batch normalization processing, Indicates The neighborhood of each node is aggregated and calculated. Indicates The neighborhood of a node, express function, represents the Hadamard product, .

[0046] Furthermore, before using the heuristic learner based on the graph neural network to generate the heuristic metric and converting the state transition model into a location exploration model affected by the heuristic metric and requiring graph traversal, the method further includes:

[0047] The gradient strategy is used to train the heuristic learner based on the graph neural network. The objective function used in the training process is for:

[0048]

[0049] Where: represents the objective function corresponding to the actual discharge location of the PD source using the neural-guided perturbation interleaved local search algorithm, Indicates balance and Parameters, In the heuristic metric The expected value of the actual discharge position of the local discharge source under the influence of It represents the state space corresponding to the exploration of the actual discharge position of the partial discharge source;

[0050] Among them, the objective function The gradient of , the formula is:

[0051]

[0052] Where: represents the average target value for directly exploring the actual discharge location of the PD source, represents the average target value of the actual discharge location of the PD source explored using the local search algorithm with neural-guided perturbation interleaving, In the heuristic metric Explore the gradient of the actual discharge location of the partial discharge source under the influence;

[0053] When the maximum number of iterations is reached Or the maximum number of iterations was not reached But the objective function The training ends when Indicates the minimum threshold corresponding to the objective function.

[0054] Furthermore, the local search algorithm using neural-guided perturbation interleaving is used to iteratively search the position exploration model to obtain the actual discharge position of the local discharge source, including:

[0055] Iteratively searching the location exploration model using a local search algorithm to obtain a local optimal solution;

[0056] Perform neural-guided perturbations on the local optimal solution obtained by the current iterative search, and obtain the optimal exploration scheme for the actual discharge position of the local discharge source of the current iterative search through the local search interleaved with neural-guided perturbations;

[0057] After all the local discharge source nodes are selected, the pheromone concentrations between the local discharge source nodes are updated;

[0058] When the iterative search reaches an iterative convergence condition, an optimal exploration scheme for the actual discharge position of the partial discharge source is determined, and matching of the actual discharge position of the partial discharge source is achieved based on the optimal exploration scheme.

[0059] Furthermore, the local search algorithm is used to iteratively search the location exploration model to obtain a local optimal solution. The formula is:

[0060]

[0061]

[0062] Where: represents the objective function, represents the local search operator, represents the state space corresponding to the exploration of the actual discharge position of the PD source, Indicates that the local search for the discharge position is performed under the local search operator. Indicates The measured discharge intensity is obtained by processing the full waveform in the time domain collected by the sensor at each moment. Indicates The local discharge intensity obtained by simulation at each moment is represents the exploration cycle, .

[0063] Furthermore, the neural-guided perturbation is performed on the local optimal solution obtained by the current iterative search, and the optimal exploration scheme for the actual discharge position of the local discharge source of the current iterative search is obtained through the neural-guided perturbation staggered local search, which is expressed as follows:

[0064]

[0065] Where: represents the optimal exploration scheme of the actual discharge position of the local discharge source in the current iteration obtained by performing neural-guided perturbation on the local optimal solution, is the local optimal solution; represents the number of perturbation moves, Represents a heuristic metric.

[0066] Furthermore, the updating formula of the pheromone concentration is expressed as:

[0067]

[0068] Where: represents the update coefficient, Indicates The pheromone concentration during the exploration cycle, Indicates The pheromone concentration during the exploration cycle, Indicates Exploration cycle to The change in pheromone concentration during an exploration cycle.

[0069] Furthermore, the simulator is a GIS three-dimensional simulation module, and the simulator is divided into non-uniform grids.

[0070] Furthermore, after performing iterative search on the measured discharge intensity calculated by the time domain full waveform data collected in real time and the simulation enhancement result to obtain the actual discharge position of the partial discharge source, the method further includes:

[0071] Based on the actual discharge position of the partial discharge source, an early warning is given.

[0072] In addition, the present invention also proposes a GIS partial discharge optimal position matching system, the system comprising:

[0073] The coarse positioning module is used to coarsely locate the local discharge source based on the time-domain full waveform data of the GIS local discharge, and obtain the preliminary discharge intensity and preliminary discharge position of the local discharge source;

[0074] A simulation module, used for injecting the preliminary discharge intensity and preliminary discharge position of the local discharge source into the simulator for simulation enhancement calculation to obtain a simulation enhancement result of the local discharge source, wherein the simulation enhancement result includes a simulation discharge position and a simulation discharge intensity;

[0075] The iterative search module is used to iteratively search the measured discharge intensity calculated by the time domain full waveform data collected in real time and the simulation enhancement result to obtain the actual discharge position of the partial discharge source.

[0076] In addition, the present invention also proposes an industrial computer, which includes a memory and a processor; wherein the processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to implement the GIS partial discharge optimal position matching method as described above.

[0077] In addition, the present invention also proposes a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the GIS partial discharge optimal position matching method as described above is implemented.

[0078] In addition, the present invention also proposes a multi-structure adaptive GIS field-electric fusion real-time status perception and early warning system, including a signal synchronization acquisition and measurement embedded device and the industrial control computer as described above, the industrial control computer is connected to the signal synchronization acquisition and measurement embedded device, the signal synchronization acquisition and measurement embedded device is used to collect the time domain full waveform data of GIS local discharge, and the industrial control computer is used to search for the actual discharge position of the GIS local discharge source.

[0079] The advantages of the present invention are:

[0080] (1) The present invention firstly roughly locates the local discharge source based on the time-domain full-waveform data of the local discharge of GIS, obtains the preliminary discharge intensity and preliminary discharge position of the local discharge source, injects the preliminary discharge intensity and preliminary discharge position of the local discharge source into the simulator, simulates the local discharge phenomenon of the GIS equipment and performs simulation calculation, obtains the simulation results of the local discharge position and the change of the local discharge intensity, and then compares the difference between the measured discharge intensity obtained by the time-domain full-waveform data of the local discharge of GIS and the simulation result based on the analysis of the simulation calculation results, and continuously iterates and optimizes until the difference between the measured value and the simulation value is minimized, thereby achieving the optimal position matching of the local discharge of GIS.

[0081] (2) The present invention takes the preliminarily calculated local discharge source as the center, and proposes a heuristic fast iterative search algorithm for the optimal match of the measured values ​​in its neighborhood. The fast iterative search problem of the optimal matching position of the discharge source is input into the graph neural network to obtain its heuristic metric as a substitute for the expert design heuristic, and based on this, the ant colony algorithm is iterated. Under the framework of the ant colony algorithm, local search of the constructed solution will obtain a better solution. The deep ant colony algorithm uses the graph neural network to generate the heuristic metric, reducing the need for expert knowledge. In combination with the probabilistic local search, the local search method of neural-guided perturbation interleaving will achieve better performance, so as to ensure efficient solution and realize the fast iterative search of the optimal matching position of the discharge source.

[0082] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 It is a flow chart of a method for matching the optimal position of GIS partial discharge proposed in one embodiment of the present invention;

[0084] Figure 2 It is a principle block diagram of GIS partial discharge optimal position matching based on deep ant colony algorithm in one embodiment of the present invention;

[0085] Figure 3 It is a schematic diagram of the optimal location search process of GIS partial discharge based on the deep ant colony algorithm in one embodiment of the present invention;

[0086] Figure 4 It is a structural schematic diagram of a GIS partial discharge optimal position matching system proposed in one embodiment of the present invention. DETAILED DESCRIPTION

[0087] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in combination with the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0088] like Figure 1 As shown, the first embodiment of the present invention proposes a method for matching the optimal position of GIS partial discharge, and the method comprises the following steps:

[0089] S10, based on the time domain full waveform data of GIS local discharge, the local discharge source is roughly located to obtain the preliminary discharge intensity and preliminary discharge position of the local discharge source;

[0090] It should be noted that this embodiment can use UHF sensors or other data to collect partial discharge signals of GIS equipment to obtain UHF time domain full waveform data when the GIS equipment is running. Then, methods such as time difference of arrival (TDOA) are used to roughly locate the UHF time domain full waveform data to obtain the preliminary discharge intensity and preliminary discharge position of the partial discharge source.

[0091] S20, injecting the preliminary discharge intensity and preliminary discharge position of the local discharge source into the simulator for simulation enhancement calculation to obtain a simulation enhancement result of the local discharge source, wherein the simulation enhancement result includes a simulation discharge position and a simulation discharge intensity;

[0092] It should be noted that by injecting the preliminary discharge intensity and preliminary discharge position of the partial discharge source into the simulator of the GIS for simulation calculation, the simulation results of the partial discharge source, including the changing partial discharge position and partial discharge intensity, can be obtained.

[0093] S30, iteratively searching the measured discharge intensity calculated by the time domain full waveform data collected in real time and the simulation enhancement result to obtain the actual discharge position of the partial discharge source.

[0094] This embodiment compares the measured discharge intensity obtained through the time domain full waveform data of GIS partial discharge with the simulation results, analyzes the difference between the measured discharge intensity and the simulation results, and continuously iterates and optimizes until the difference between the measured value and the simulation value is minimized, thereby achieving the optimal position matching of GIS partial discharge.

[0095] As a further preferred technical solution, the step S10: coarsely locating the local discharge source based on the time-domain full waveform data of the GIS local discharge to obtain the preliminary discharge intensity and preliminary discharge position of the local discharge source includes the following steps:

[0096] S11. Calculate the initial discharge position of the local discharge source based on the full waveform data of GIS local discharge in the time domain , the formula is:

[0097]

[0098]

[0099] In the formula, Indicates The distance from the sensor to the local discharge source, It represents the propagation speed of electromagnetic waves in the current propagation medium. Indicates the partial discharge signal propagating to the reference sensor The time required, Indicates the partial discharge signal propagating to the reference sensor and propagates to the sensor The time difference, The coordinates of the UHF sensors are , ;

[0100] S12. Calculate the initial discharge intensity of the local discharge source based on the full waveform data of GIS local discharge in the time domain .

[0101] It should be noted that, in this embodiment, the local discharge intensity is obtained by applying a specific relationship between the UHF signal and the maximum value of the actual discharge amount.

[0102] As a further preferred technical solution, the step S20: injecting the preliminary discharge intensity and preliminary discharge position of the local discharge source into the simulator for simulation calculation to obtain the simulation result of the local discharge source, comprises the following steps:

[0103] S21, injecting the initial discharge intensity and initial discharge position of the local discharge source into the simulator to simulate the local discharge phenomenon of the GIS equipment;

[0104] S22. Perform simulation enhancement calculation using a simulator to obtain a simulated discharge position and a simulated discharge intensity of the local discharge source.

[0105] It should be noted that, in this embodiment, the preliminary discharge intensity and preliminary discharge position of the local discharge source are injected into the simulator of the GIS, and then the simulation software is used to perform simulation calculations to obtain the simulated discharge position and simulated discharge intensity of the local discharge source.

[0106] Preferably, the GIS simulator used in this embodiment is divided into non-uniform grids of different thicknesses, and the preliminary discharge position and preliminary discharge intensity of the local discharge source are injected into the corresponding grid cells of the non-uniform grid. The non-uniform grid simulator is used to enhance the coarse grid simulation data to obtain enhanced fine grid data, and the simulation of the local discharge node will be more refined, making the positioning more accurate. It should be understood that the GIS simulator in this embodiment can also use a uniform grid, and this embodiment does not specifically limit this.

[0107] As a further preferred technical solution, the step S30: iteratively searching the measured discharge intensity calculated by the time domain full waveform data collected in real time and the simulation enhancement result to obtain the actual discharge position of the partial discharge source includes the following steps:

[0108] S31, according to the simulated discharge position and simulated discharge intensity at each moment, using the Markov decision model to reconstruct the state transition model corresponding to the process of the ant colony algorithm exploring the actual discharge position of the local discharge source with the initial discharge position of the local discharge source as the center;

[0109] It should be noted that, in this embodiment, the initial discharge position is taken as the center in the non-uniform grid of the simulator, and based on the simulated discharge position and the simulated discharge intensity at each moment, a state space and an action space corresponding to the process of exploring the actual discharge position of the local discharge source with the initial discharge position as the center is constructed in the non-uniform grid, thereby constructing a state transition model corresponding to the process of exploring the actual discharge position of the local discharge source with the initial discharge position of the local discharge source as the center.

[0110] S32, using a heuristic learner based on a graph neural network to generate a heuristic metric, and converting the state transition model into a location exploration model that is affected by the heuristic metric and requires graph traversal;

[0111] S33. Use a neural-guided perturbation interleaved local search algorithm to iteratively search the position exploration model to obtain an actual discharge position of the local discharge source.

[0112] like Figure 2 As shown, in this embodiment, the heuristic rules in the ant colony algorithm are automatically designed and enhanced in a data-driven manner to more effectively solve the rapid iterative search for the optimal matching position of the discharge source. The deep ant colony algorithm uses a graph neural network to generate heuristic metrics, reducing the need for expert knowledge, and combines probabilistic local search to achieve better performance to ensure efficient solution.

[0113] As a further preferred technical solution, the step S31: based on the simulated discharge position and the simulated discharge intensity at each moment, using the Markov decision model to reconstruct the state transition model corresponding to the process of the ant colony algorithm exploring the actual discharge position of the local discharge source with the preliminary discharge position of the local discharge source as the center, specifically includes:

[0114] S311, constructing a state space corresponding to the actual discharge position of the local discharge source in the simulator according to the simulated discharge position and simulated discharge intensity at each moment :

[0115]

[0116]

[0117] Where: Indicates that in the simulator The local discharge source node is explored at The corresponding status, Indicates The simulated discharge position at each moment, Indicates The simulated discharge intensity at each moment, It represents the initial discharge intensity of the local discharge source obtained based on the full waveform data of GIS local discharge in the time domain and initial discharge position As the initial state, represents the exploration cycle, ;

[0118] S312, constructing the action space corresponding to the actual discharge position of the local discharge source in the simulator :

[0119]

[0120] Where: Indicates The action of the moment Greedy strategy is the standard choice of action ;

[0121] S313, the state transition model corresponding to the process of reconstructing the ant colony algorithm to explore the actual discharge position of the local discharge source with the initial discharge position of the local discharge source as the center is:

[0122]

[0123] Where: Indicates in action The discharge node To the discharge node The probability of transfer, Indicates that in the simulator The local discharge source node is explored at The corresponding state, Represented by the node Transfer to Node The corresponding pheromone concentration, Represented by the node Transfer to Node The corresponding heuristic function is, Represented by the node Transfer to the included Nodes in The corresponding pheromone concentration, Represented by the node Transfer to the included Nodes in The corresponding heuristic function is, and Represent the pheromone concentration and the weight of the heuristic function, Represents the set of non-uniform grid local discharge source nodes that are allowed to be selected at the next moment.

[0124] Furthermore, by the node To Node The reward function for the transfer is , Indicates The measured discharge intensity is obtained by processing the full waveform in the time domain collected by the sensor at each moment. Indicates The partial discharge intensity obtained by simulation at each moment.

[0125] It should be noted that this embodiment uses the Markov model to formalize and model the state transition process in the ant colony algorithm. This modeling can help understand and analyze the search process of the ant colony algorithm in the solution space, including key steps such as ant path selection and pheromone update.

[0126] As a further preferred technical solution, the step S32: using a heuristic learner based on a graph neural network to generate a heuristic metric, converting the state transition model into a location exploration model affected by the heuristic metric and requiring graph traversal, includes the following steps:

[0127] S321, use the multi-layer perceptron to extract the graph neural network The layer is The node and Edge features connecting nodes and mapped to the heuristic metric ;

[0128] S322, converting the state transition model into a state transition model that is affected by heuristic metrics and requires The location exploration model of step graph traversal is:

[0129]

[0130] Where: represents the location exploration model, Indicates that in the simulator The local discharge source node is explored at The corresponding status, Indicates that in the simulator The local discharge source node is explored at The corresponding status, represents the exploration cycle, .

[0131] It should be noted that this embodiment constructs a heuristic learner based on graph neural network, parameterizes the heuristic space for exploring GIS local discharge sources, generates heuristic metrics by using graph neural network, and combines probabilistic local search to achieve better performance to ensure efficient solution.

[0132] Furthermore, the graph neural network The propagation characteristics of the layer are:

[0133]

[0134]

[0135] Where: Indicates Layer The characteristics of the nodes, Indicates Layer The characteristics of the nodes, Indicates Layer The characteristics of the nodes, Indicates The layer is The node and The edge features connecting the nodes are Indicates The layer is The node and The edge features connecting the nodes are , , , as well as Indicates The learnable parameters of the layer, represents the activation function, represents batch normalization processing, Indicates The neighborhood of each node is aggregated and calculated. Indicates The neighborhood of a node, express function, represents the Hadamard product, .

[0136] It should be noted that when the ant colony algorithm explores the location of local discharge, it constructs a graph by traversing the discharge nodes, and the nodes of the graph neural network correspond to the discharge nodes.

[0137] As a further preferred technical solution, in step S32: before generating a heuristic metric using a heuristic learner based on a graph neural network and converting the state transition model into a location exploration model affected by the heuristic metric and requiring graph traversal, the method further includes the following steps:

[0138] The gradient strategy is used to train the heuristic learner based on the graph neural network. The objective function used in the training process is for:

[0139]

[0140] Where: represents the objective function corresponding to the actual discharge location of the PD source using the neural-guided perturbation interleaved local search algorithm, Indicates balance and Parameters, In the heuristic metric The expected value of the actual discharge position of the local discharge source under the influence of represents the state space corresponding to the exploration of the actual discharge position of the PD source, represents the objective function;

[0141] Among them, the objective function The gradient of , the formula is:

[0142]

[0143] Where: represents the average target value for directly exploring the actual discharge location of the PD source, represents the average target value of the actual discharge location of the PD source explored using the local search algorithm with neural-guided perturbation interleaving, In the heuristic metric Explore the gradient of the actual discharge location of the partial discharge source under the influence;

[0144] When the maximum number of iterations is reached Or the maximum number of iterations was not reached But the objective function The training ends when Indicates the minimum threshold corresponding to the objective function.

[0145] As a further preferred technical solution, the step S33: using a neural-guided perturbation interleaved local search algorithm to iteratively search the position exploration model to obtain the actual discharge position of the local discharge source specifically includes the following steps:

[0146] S331, using a local search algorithm to iteratively search the location exploration model to obtain a local optimal solution;

[0147] Specifically, the local optimal solution The formula is expressed as:

[0148]

[0149]

[0150] Where: represents the objective function, represents the local search operator, represents the state space corresponding to the exploration of the actual discharge position of the PD source, Indicates that the local search for the discharge position is performed under the local search operator. Indicates The measured discharge intensity is obtained by processing the full waveform in the time domain collected by the sensor at each moment. Indicates The local discharge intensity obtained by simulation at each moment is represents the exploration cycle, .

[0151] S332, performing neural-guided perturbation on the local optimal solution obtained by the current iterative search, and obtaining the optimal exploration scheme for the actual discharge position of the local discharge source of the current iterative search through the neural-guided perturbation staggered local search;

[0152] Specifically, the optimal exploration scheme of the actual discharge position of the local discharge source in the previous iteration is obtained, and the formula is expressed as:

[0153]

[0154] Where: represents the optimal exploration scheme of the actual discharge position of the local discharge source in the current iteration obtained by performing neural-guided perturbation on the local optimal solution, is the local optimal solution; represents the number of perturbation moves, Represents a heuristic metric.

[0155] It should be noted that the optimal exploration scheme for the actual discharge position of the local discharge source in the current iteration obtained by the disturbance is then locally searched to obtain the optimal exploration scheme for the actual discharge position of the local discharge source in the current iteration. Local search to obtain the local optimal solution :

[0156]

[0157] go through The optimal exploration scheme of GIS local discharge sources obtained by local search with neural-guided perturbation interleaving :

[0158]

[0159] Where: Represents a set of parameters or variables that minimizes the function.

[0160] It should be noted that when a local search algorithm is used to iteratively search the position exploration model to obtain a local optimal solution, a local optimal situation may occur. This embodiment perturbs the local optimal solution by utilizing the heuristic metric obtained by graph neural network training. The heuristic metric can help the algorithm quickly locate and jump out of the local optimal solution, guide the search process to approach the global optimal solution, and guide the search process to explore the solution space more effectively.

[0161] S333, after all local discharge source nodes are selected, updating the pheromone concentration between each local discharge source node;

[0162] Specifically, the update formula of pheromone concentration is expressed as:

[0163]

[0164] Where: represents the update coefficient, Indicates The pheromone concentration during the exploration cycle, Indicates The pheromone concentration during the exploration cycle, Indicates Exploration cycle to The change in pheromone concentration during an exploration cycle.

[0165] S333. When the iterative search reaches an iterative convergence condition, determine an optimal exploration scheme for the actual discharge position of the partial discharge source, and achieve matching of the actual discharge position of the partial discharge source based on the optimal exploration scheme, wherein the iterative convergence condition may be a set number of iterations.

[0166] The ant colony algorithm in this embodiment simulates the search process of ants in a graph to find the possible location of the local discharge source. The graph neural network can effectively propagate the information of the local discharge signal through message passing and feature aggregation, and infer and locate the location in combination with the search results of the ant colony algorithm. The parallel computing capability and efficient feature learning of the graph neural network enable it to quickly process large-scale graph data, which is suitable for real-time or near-real-time local discharge location tasks. As a heuristic optimization method, the ant colony algorithm, combined with the feature learning ability of the graph neural network, can intelligently guide the search process and quickly discover the possible location of the discharge source. This method can effectively avoid falling into the local optimal solution and improve the global search capability, and can achieve fast real-time high-precision positioning.

[0167] It should be noted that if Figure 3As shown, in this embodiment, the initial discharge position can be used as the center of the circle and the radius of the GIS cross section can be used as the radius of the spherical area for local discharge positioning to determine the optimal exploration scheme, and the actual discharge position of the local discharge source can be matched according to the optimal exploration scheme path, so that the local discharge source position can be quickly positioned.

[0168] As a further preferred technical solution, in step S30: after iteratively searching the measured discharge intensity calculated by the time domain full waveform data collected in real time and the simulation enhancement result to obtain the actual discharge position of the partial discharge source, the method further comprises the following steps:

[0169] Based on the actual discharge position of the partial discharge source, an early warning is given.

[0170] It should be noted that the GIS partial discharge optimal position matching method proposed in this embodiment can achieve rapid early warning of GIS faults, timely and accurately discover fault problems, and avoid accidents.

[0171] In addition, if Figure 4 As shown, the second embodiment of the present invention proposes a GIS partial discharge optimal position matching system, the system comprising:

[0172] A coarse positioning module 10 is used to coarsely locate the local discharge source based on the time domain full waveform data of the GIS local discharge, and obtain the preliminary discharge intensity and preliminary discharge position of the local discharge source;

[0173] The simulation module 20 is used to inject the preliminary discharge intensity and preliminary discharge position of the local discharge source into the simulator to perform simulation enhancement calculation to obtain the simulation enhancement result of the local discharge source, wherein the simulation enhancement result includes the simulation discharge position and the simulation discharge intensity;

[0174] The iterative search module 30 is used to iteratively search the measured discharge intensity calculated by the time domain full waveform data collected in real time and the simulation enhancement result to obtain the actual discharge position of the partial discharge source.

[0175] As a further preferred technical solution, the simulation module 20 includes:

[0176] The initial position injection unit is used to inject the initial discharge intensity and initial discharge position of the local discharge source into the GIS simulator to simulate the local discharge phenomenon of the GIS equipment;

[0177] The simulation calculation unit is used to perform simulation enhancement calculation using a simulator to obtain a simulated discharge position and a simulated discharge intensity of a local discharge source.

[0178] As a further preferred technical solution, the iterative search module 30 includes:

[0179] A state transition model construction unit is used to reconstruct the state transition model corresponding to the process of the ant colony algorithm exploring the actual discharge position of the local discharge source with the initial discharge position of the local discharge source as the center according to the simulated discharge position and the simulated discharge intensity at each moment by using the Markov decision model;

[0180] An heuristic space parameterization unit, used for generating a heuristic metric using a heuristic learner based on a graph neural network, and converting the state transition model into a position exploration model that is affected by the heuristic metric and requires graph traversal;

[0181] The iterative search unit is used to iteratively search the position exploration model using a neural-guided perturbation interleaved local search algorithm to obtain an actual discharge position of a local discharge source.

[0182] As a further preferred technical solution, the state transition model building unit is specifically used to:

[0183] According to the simulated discharge position and simulated discharge intensity at each moment, the state space corresponding to the actual discharge position of the local discharge source is constructed in the simulator :

[0184]

[0185]

[0186] Where: Indicates that in the simulator The local discharge source node is explored at The corresponding status, Indicates The simulated discharge position at each moment, Indicates The simulated discharge intensity at each moment, It represents the initial discharge intensity of the local discharge source obtained based on the full waveform data of GIS local discharge in the time domain and initial discharge position As the initial state, represents the exploration cycle, ;

[0187] Constructing the action space corresponding to the actual discharge location of the PD source in the simulator :

[0188]

[0189] Where: Indicates The action of the moment Greedy strategy is the standard choice of action ;

[0190] The state transition model corresponding to the process of the reconstructed ant colony algorithm exploring the actual discharge position of the local discharge source with the initial discharge position of the local discharge source as the center is:

[0191]

[0192] Where: Indicates in action The discharge node To the discharge node The probability of transfer, Indicates that in the simulator The local discharge source node is explored at The corresponding state, Represented by the node Transfer to Node The corresponding pheromone concentration, Represented by the node Transfer to Node The corresponding heuristic function is, Represented by the node Transfer to the included Nodes in The corresponding pheromone concentration, Represented by the node Transfer to the included Nodes in The corresponding heuristic function is, and Represent the pheromone concentration and the weight of the heuristic function, Represents the set of non-uniform grid local discharge source nodes that are allowed to be selected at the next moment.

[0193] As a further preferred technical solution, the heuristic space parameterization unit is specifically used for:

[0194] The graph neural network The layer is The node and Edge features connecting nodes Mapping to heuristic metrics ;

[0195] The state transition model is transformed into a state transition model that is influenced by the heuristic metric and requires The location exploration model of step graph traversal is:

[0196]

[0197] Where: represents the location exploration model, Indicates that in the simulator The local discharge source node is explored at The corresponding status, Indicates that in the simulator The local discharge source node is explored at The corresponding status, represents the exploration cycle, .

[0198] As a further preferred technical solution, the iterative search unit is specifically used to:

[0199] Iteratively searching the location exploration model using a local search algorithm to obtain a local optimal solution;

[0200] Perform neural-guided perturbations on the local optimal solution obtained by the current iterative search, and obtain the optimal exploration scheme for the actual discharge position of the local discharge source of the current iterative search through the local search interleaved with neural-guided perturbations;

[0201] After all the local discharge source nodes are selected, the pheromone concentrations between the local discharge source nodes are updated;

[0202] When the iterative search reaches an iterative convergence condition, an optimal exploration scheme for the actual discharge position of the partial discharge source is determined, and matching of the actual discharge position of the partial discharge source is achieved based on the optimal exploration scheme.

[0203] As a further preferred technical solution, the GIS partial discharge optimal position matching system also includes a training module for training a heuristic learner based on a graph neural network using a gradient strategy.

[0204] This embodiment is based on the deep ant colony algorithm. When partial discharge occurs in GIS, a heuristic fast iterative search algorithm is proposed for the optimal matching of measurement values ​​in the neighborhood of the initially calculated discharge source, so as to realize real-time high-precision positioning of the discharge source.

[0205] It should be noted that other embodiments or implementation methods of the GIS partial discharge optimal position matching system of the present invention can refer to the above-mentioned method embodiments and will not be described in detail here.

[0206] In addition, the third embodiment of the present invention also proposes an industrial computer, which includes a memory and a processor; wherein the processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to implement the GIS partial discharge optimal position matching method as described in the first embodiment above.

[0207] In addition, the fourth embodiment of the present invention further proposes a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the GIS partial discharge optimal position matching method described in the first embodiment above is implemented.

[0208] The industrial computer or computer-readable storage medium provided above can be used to execute the GIS partial discharge optimal position matching method provided in the above embodiment, and has corresponding functions and beneficial effects.

[0209] In addition, the fifth embodiment of the present invention also proposes a multi-structure adaptive GIS field-electric fusion real-time status perception and early warning system, including a signal synchronization acquisition and measurement embedded device and an industrial computer as described above, the industrial computer is connected to the signal synchronization acquisition and measurement embedded device, the signal synchronization acquisition and measurement embedded device is used to collect the time domain full waveform data of GIS local discharge, and the industrial computer is used to search for the actual discharge position of the GIS local discharge source.

[0210] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0211] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0212] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0213] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0214] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A GIS partial discharge optimal position matching method, characterized in that: The method comprises: Based on the full waveform data of GIS partial discharge in the time domain, the partial discharge source is roughly located to obtain the preliminary discharge intensity and preliminary discharge position of the partial discharge source; Injecting the preliminary discharge intensity and preliminary discharge position of the local discharge source into the simulator for simulation enhancement calculation to obtain a simulation enhancement result of the local discharge source, wherein the simulation enhancement result includes a simulation discharge position and a simulation discharge intensity; An iterative search is performed between the measured discharge intensity calculated by the time domain full waveform data collected in real time and the simulation enhancement result to obtain the actual discharge position of the partial discharge source, including: According to the simulated discharge position and simulated discharge intensity at each moment, the state transition model corresponding to the process of the ant colony algorithm exploring the actual discharge position of the local discharge source with the initial discharge position of the local discharge source as the center is reconstructed using the Markov decision model; Generate a heuristic metric using a heuristic learner based on a graph neural network, and transform the state transition model into a location exploration model that is influenced by the heuristic metric and requires graph traversal; The position exploration model is iteratively searched using a neural-guided perturbation interleaved local search algorithm to obtain the actual discharge position of the local discharge source.

2. The GIS partial discharge optimal position matching method according to claim 1, characterized in that: The method of roughly locating the local discharge source based on the time domain full waveform data of the local discharge of the GIS to obtain the preliminary discharge intensity and preliminary discharge position of the local discharge source includes: Based on the full waveform data of GIS partial discharge in the time domain, the initial discharge position P = (x, y, z) of the partial discharge source is calculated. The formula is expressed as follows: r n =v(t+τ 1n ) In the formula, r n represents the distance from the nth sensor to the local discharge source, v represents the propagation speed of the electromagnetic wave in the current propagation medium, t represents the time required for the local discharge signal to propagate to the reference sensor S1, τ 1n It indicates that the partial discharge signal propagates to the reference sensor S1 and propagates to the sensor S n The time difference, the coordinates of the nth UHF sensor are (x n ,y n ,z n ), n=1,2,…,N; Based on the full waveform data of GIS partial discharge in the time domain, the initial discharge intensity E of the partial discharge source is calculated.

3. The GIS partial discharge optimal position matching method according to claim 1, characterized in that: The preliminary discharge intensity and preliminary discharge position of the local discharge source are injected into the simulator for simulation calculation to obtain the simulation result of the local discharge source, including: Inject the initial discharge intensity and initial discharge position of the local discharge source into the simulator to simulate the local discharge phenomenon of GIS equipment; The simulator is used to perform simulation enhancement calculations to obtain the simulated discharge position and simulated discharge intensity of the local discharge source.

4. The GIS partial discharge optimal position matching method according to claim 1, characterized in that: The state transition model corresponding to the process of reconstructing the ant colony algorithm based on the simulated discharge position and the simulated discharge intensity at each moment and centering on the preliminary discharge position of the local discharge source and exploring the actual discharge position of the local discharge source by the Markov decision model includes: According to the simulated discharge position and simulated discharge intensity at each moment, the state space corresponding to the actual discharge position of the local discharge source is constructed in the simulator Where: It indicates the state corresponding to the local discharge source node i explored in the simulator at the tth moment, p t represents the simulated discharge position at the tth moment, E t represents the simulated discharge intensity at the tth moment, It means that the initial discharge intensity E and the initial discharge position P of the local discharge source obtained based on the time domain full waveform data of GIS local discharge are used as the initial state, T represents the exploration period, t=1,2,…,T; Constructing the action space corresponding to the actual discharge location of the PD source in the simulator Where: Represents the action at the tth moment, and selects the action based on the ε-greedy strategy The state transition model corresponding to the process of the reconstructed ant colony algorithm exploring the actual discharge position of the local discharge source with the initial discharge position of the local discharge source as the center is: Where: Indicates in action The probability of transferring from discharge node i to discharge node j is: Indicates the state corresponding to the local discharge source node j explored in the simulator at the t+1th time, τ ij represents the pheromone concentration corresponding to the transfer from node i to node j, η ij represents the heuristic function corresponding to the transfer from node i to node j, τ im Indicates that node i is transferred to the node contained in allowed S The pheromone concentration corresponding to the node m in , η im Indicates that node i is transferred to the node contained in allowed S The heuristic function corresponding to the node m in , α and β represent the weight of pheromone concentration and heuristic function respectively, and allowed S Represents the set of non-uniform grid local discharge source nodes that are allowed to be selected at the next moment.

5. The GIS partial discharge optimal position matching method according to claim 4, characterized in that: The reward function for transferring from node i to node j is E′ t It represents the measured discharge intensity obtained by processing the full waveform in the time domain collected by the sensor at the tth moment, E t Represents the local discharge intensity obtained by simulation at the tth moment.

6. The GIS partial discharge optimal position matching method according to claim 1, characterized in that: The heuristic metric is generated by using a heuristic learner based on a graph neural network, and the state transition model is converted into a location exploration model affected by the heuristic metric and requiring graph traversal, including: The edge features connecting the i-th node and the j-th node in the l-th layer of the graph neural network Mapping to the heuristic metric η θ ; The state transition model is converted into a location exploration model that is affected by heuristic metrics and requires T-step graph traversal: Where: represents the location exploration model, It means that the state corresponding to the local discharge source node i is explored in the simulator at the tth moment. It indicates that the state corresponding to the local discharge source node i is explored at the t+1th time in the simulator, T represents the exploration period, t=1,2,…,T.

7. The GIS partial discharge optimal position matching method according to claim 1, characterized in that: The propagation characteristics of the lth layer of the graph neural network are: Where: represents the feature of the i-th node in the l-th layer, represents the feature of the jth node in the lth layer, represents the feature of the i-th node in the l+1th layer, represents the edge feature connecting the i-th node and the j-th node in the l-th layer, represents the edge feature connecting the i-th node and the j-th node in the l+1th layer, U l 、V l , P l , Q l and R l represents the learnable parameters of layer l, represents the activation function, BN represents batch normalization, Indicates the aggregation calculation of the neighborhood of the i-th node. represents the neighborhood of the i-th node, σ represents the Sigmoid function, ⊙ represents the Hadamard product, and l = 1, 2,…, L.

8. The GIS partial discharge optimal position matching method according to claim 7, characterized in that: Before using the heuristic learner based on the graph neural network to generate the heuristic metric and converting the state transition model into a location exploration model affected by the heuristic metric and requiring graph traversal, the method further includes: The gradient strategy is used to train the heuristic learner based on the graph neural network. The objective function used in the training process is for: Where: represents the objective function corresponding to the actual discharge position of the PD source using the local search algorithm guided by neural perturbations, and W represents the balance and Parameters, In the heuristic metric η θ The expected value of the actual discharge position of the local discharge source under the influence of represents the state space corresponding to the exploration of the actual discharge position of the PD source, and f(·) represents the objective function; Among them, the objective function The gradient of The formula is: Where: represents the average target value for directly exploring the actual discharge location of the PD source, represents the average target value of the actual discharge location of the PD source explored using the local search algorithm with neural-guided perturbation interleaving, In the heuristic metric η θ Explore the gradient of the actual discharge location of the partial discharge source under the influence; When the maximum number of iterations is reached Or the maximum number of iterations was not reached But the objective function The training ends when θ represents the minimum threshold value corresponding to the objective function.

9. The GIS partial discharge optimal position matching method according to claim 1, characterized in that: The local search algorithm using neural-guided perturbation interleaving is used to iteratively search the position exploration model to obtain the actual discharge position of the local discharge source, including: Iteratively searching the location exploration model using a local search algorithm to obtain a local optimal solution; Perform neural-guided perturbations on the local optimal solution obtained by the current iterative search, and obtain the optimal exploration scheme for the actual discharge position of the local discharge source of the current iterative search through the local search interleaved with neural-guided perturbations; After all the local discharge source nodes are selected, the pheromone concentrations between the local discharge source nodes are updated; When the iterative search reaches an iterative convergence condition, an optimal exploration scheme for the actual discharge position of the partial discharge source is determined, and matching of the actual discharge position of the partial discharge source is achieved based on the optimal exploration scheme.

10. The GIS partial discharge optimal position matching method according to claim 9, characterized in that: The local search algorithm is used to iteratively search the location exploration model to obtain a local optimal solution. The formula is expressed as: Where: f(·) represents the objective function, LS represents the local search operator, represents the state space corresponding to the actual discharge position of the local discharge source, LS() represents the local search for the discharge position under the local search operator, and E′ t It represents the measured discharge intensity obtained by processing the full waveform in the time domain collected by the sensor at the tth moment, E t represents the local discharge intensity obtained by simulation at the tth moment, T represents the exploration period, t=1,2,…,T.

11. The GIS partial discharge optimal position matching method according to claim 9, characterized in that: The neural-guided perturbation is performed on the local optimal solution obtained by the current iterative search, and the optimal exploration scheme for the actual discharge position of the local discharge source of the current iterative search is obtained through the neural-guided perturbation staggered local search, which is expressed as follows: Where: represents the optimal exploration scheme of the actual discharge position of the local discharge source in the current iteration obtained by performing neural-guided perturbation on the local optimal solution, is the local optimal solution; T p represents the number of perturbation moves, η θ Represents a heuristic metric.

12. The GIS partial discharge optimal position matching method according to claim 9, characterized in that: The updating formula of the pheromone concentration is expressed as: t ij (k+1)=rt ij (k)+Δτ ij (k,k+1) Where: ρ represents the update coefficient, τ ij (T) represents the pheromone concentration in the kth exploration cycle, τ ij (k+1) represents the pheromone concentration in the k+1th exploration cycle, Δτ ij (k, k+1) represents the change in pheromone concentration from the kth exploration cycle to the k+1th exploration cycle.

13. The GIS partial discharge optimal position matching method according to claim 1, characterized in that: The simulator is a GIS three-dimensional simulation module, and the simulator is divided into non-uniform grids.

14. The GIS partial discharge optimal position matching method according to claim 1, characterized in that: After the actual discharge position of the partial discharge source is obtained by iteratively searching the measured discharge intensity calculated by the time domain full waveform data collected in real time and the simulation enhancement result, the method further includes: Based on the actual discharge position of the partial discharge source, an early warning is given.

15. A GIS partial discharge optimal position matching system, characterized in that: The system comprises: The coarse positioning module is used to coarsely locate the local discharge source based on the time-domain full waveform data of the GIS local discharge, and obtain the preliminary discharge intensity and preliminary discharge position of the local discharge source; A simulation module, used for injecting the preliminary discharge intensity and preliminary discharge position of the local discharge source into the simulator for simulation enhancement calculation to obtain a simulation enhancement result of the local discharge source, wherein the simulation enhancement result includes a simulation discharge position and a simulation discharge intensity; An iterative search module, used for iteratively searching the measured discharge intensity calculated by the time domain full waveform data collected in real time and the simulation enhancement result to obtain the actual discharge position of the partial discharge source; The iterative search module comprises: A state transition model construction unit is used to reconstruct the state transition model corresponding to the process of the ant colony algorithm exploring the actual discharge position of the local discharge source with the initial discharge position of the local discharge source as the center according to the simulated discharge position and the simulated discharge intensity at each moment by using the Markov decision model; An heuristic space parameterization unit, used for generating a heuristic metric using a heuristic learner based on a graph neural network, and converting the state transition model into a position exploration model that is affected by the heuristic metric and requires graph traversal; The iterative search unit is used to iteratively search the position exploration model using a neural-guided perturbation interleaved local search algorithm to obtain an actual discharge position of a local discharge source.

16. An industrial computer, characterized in that: The industrial computer includes a memory and a processor; wherein the processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to implement the method as described in any one of claims 1-14.

17. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 14 is implemented.

18. A multi-structure adaptive GIS field-electricity fusion real-time status perception and early warning system, characterized in that: It comprises a signal synchronous acquisition and measurement embedded device and an industrial computer as described in claim 16, wherein the industrial computer is connected to the signal synchronous acquisition and measurement embedded device, the signal synchronous acquisition and measurement embedded device is used to acquire the time domain full waveform data of GIS local discharge, and the industrial computer is used to search for the actual discharge position of the GIS local discharge source.

Images