Optimal position matching method and system for GIS partial discharge, and industrial personal computer

Abstract

Disclosed in the present application are an optimal position matching method and system for GIS partial discharge. The method comprises: on the basis of time-domain full-waveform data of GIS partial discharge, performing coarse localization on a partial discharge source, so as to obtain a preliminary discharge intensity and preliminary discharge position of the partial discharge source; injecting the preliminary discharge intensity and preliminary discharge position of the partial discharge source into a simulator for enhanced simulation calculation, so as to obtain a simulation enhancement result of the partial discharge source, wherein the simulation enhancement result comprises a simulated discharge position and a simulated discharge intensity; and performing an iterative search on the simulation enhancement result and an actually measured discharge intensity calculated by means of the time-domain full-waveform data collected in real time, so as to obtain the actual discharge position of the partial discharge source. In the present application, when partial discharge occurs in a GIS, by using a preliminarily calculated discharge source as a center, a heuristic fast iterative search algorithm for optimal matching of measured values within a neighborhood thereof is provided, thereby realizing real-time high-precision localization of the discharge source.

Description

Optimal location matching method, system and industrial control computer for partial discharge in GIS

[0001] This application claims priority to Chinese Patent Application No. 202411626726.6, filed with the Chinese Patent Office on November 14, 2024, the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of power equipment testing technology, specifically to a method, system, and industrial control computer for optimal location matching of partial discharge in GIS. Background Technology

[0003] With the widespread application of gas-insulated switchgear (GIS), frequent equipment failures caused by insulation aging and component loosening have seriously threatened the safe operation of substations at all levels. Because GIS equipment has a fully enclosed structure, a fault can affect a large area and incur high maintenance costs. Partial discharge is a common fault phenomenon in GIS equipment, which can lead to performance degradation, damage, and even accidents. Accurately and quickly locating partial discharge signals within the GIS is crucial for equipment health monitoring and fault diagnosis.

[0004] In related technologies, discharge detection is generally based on the ultra-high frequency (UHF) electromagnetic wave signal inside the GIS. For example, the patent application document with publication number CN116662826A uses a KELM model to compare the measured UHF signal fingerprint with a partial discharge location fingerprint database to locate the partial discharge source. This scheme extracts features from partial discharge signals of different locations and types in advance to form a fingerprint database and trains a KELM model. Then, the measured UHF signal fingerprint is input into the trained KELM model to achieve partial discharge location. However, this model depends on high-quality partial discharge and does not have an iterative optimization process. In the case of complex operating environment of power equipment, it will affect the accuracy of the location. The patent application document with publication number CN117706306A uses partial discharge location technology based on electromagnetic time reversal to locate the source of partial discharge. This scheme uses an ultra-high frequency sensor to collect ultra-high frequency electromagnetic wave pulse signals generated by partial discharge, performs time reversal operation on the pulse signals, and injects them into a GIS three-dimensional electromagnetic field model to obtain the distribution change of internal electric field intensity over time. Based on the distribution of electric field intensity, the location of partial discharge is determined. However, electromagnetic waves are affected by various factors such as equipment structure, material properties, and environmental conditions during propagation, and may encounter multipath effects such as reflection and refraction. These changes may affect the signal convergence effect in the time reversal process. The patent application document with publication number CN114660448A employs a fuzzy neural network optimized based on a crowd search algorithm. It performs pattern matching between simulated fingerprints and an experimentally measured partial discharge fingerprint database to obtain the corresponding spatial coordinates of the partial discharge source. By collecting simulated partial discharge signals generated by partial discharge, a simulated fingerprint database for partial discharge localization is constructed. The simulated fingerprints are normalized, and their feature parameters are extracted as training and testing sample datasets. A matching model is constructed based on fuzzy neural network theory. During training, a crowd search algorithm is used to correct the weights. The training and testing samples are used as inputs 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. Inappropriate parameter selection may lead to slow convergence or getting trapped in local optima during the optimization process.

[0005] The paper "Research on Power Transformer Fault Location Method Based on Multi-Ant Colony Algorithm, Wei Luyuan, China University of Mining and Technology" proposes an improved ant colony algorithm suitable for fault tree search. In transformer fault detection, one detection method can detect several fault types, and a single fault type can be detected using several detection methods. When using the ant colony algorithm for fault search, different detection methods are mapped to multiple ant colony types. Ants select paths based on pheromone concentration and heuristic rules, and adjust the pheromone update strategy according to the fault tree structure and node states to achieve more accurate fault location. However, the fault tree itself is very complex; the logical relationships and state transitions between nodes require precise modeling and analysis, increasing the difficulty of algorithm implementation and debugging. The paper "Research and Development of Intelligent Partial Discharge Inspection System, Li Ning, Hebei University of Technology" proposes an improved algorithm combining A* algorithm and ant colony algorithm to solve the path planning problem of partial discharge inspection in substations. An intelligent inspection vehicle is used to locate partial discharges in the substation. The ant colony algorithm is used to solve the path planning problem of the inspection vehicle, reducing the occurrence of duplicate paths and enabling the inspection vehicle to detect more equipment and lines within a certain cost range. RFID radio frequency technology is selected as the main technical means. The signal loss model is analyzed, and the least squares method is used to obtain the RSSI-based ranging function. The trilateration method is then used to obtain the positioning information of the inspection vehicle.

[0006] In summary, when the aforementioned technologies utilize UHF signals for discharge detection, the multiple reflections of the UHF signal along different propagation paths within the complex topology of GIS can affect the positioning accuracy. The multipath effect leads to varying time delays in the signal reaching the UHF sensor, thereby introducing positioning errors and preventing the positioning system from accurately identifying the location of the discharge source. Summary of the Invention

[0007] The technical problem to be solved in this application is how to achieve high-precision positioning of partial discharge signals.

[0008] This application solves the above-mentioned technical problems through the following technical means:

[0009] On the one hand, this application proposes a method for optimal location matching of partial discharge in GIS, the method comprising:

[0010] Based on the time-domain full waveform data of partial discharge from GIS, the partial discharge source is coarsely located to obtain the initial discharge intensity and initial discharge location of the partial discharge source.

[0011] The initial discharge intensity and initial discharge location of the partial discharge source are injected into the simulator for simulation enhancement calculation to obtain the simulation enhancement result of the partial discharge source, which includes the simulated discharge location and simulated discharge intensity.

[0012] The actual discharge location of the partial discharge source is obtained by iteratively searching the measured discharge intensity calculated from the real-time collected time-domain full waveform data and the simulation enhancement results.

[0013] Furthermore, the partial discharge source is coarsely located using the time-domain full waveform data based on GIS partial discharge to obtain the initial discharge intensity and initial discharge location of the partial discharge source, including:

[0014] Based on the time-domain full waveform data of partial discharge from GIS, the initial discharge location of the partial discharge source is calculated. The formula is expressed as:

[0015] ;

[0016] ;

[0017] In the formula, Indicates the first The distance from the sensor to the local discharge power source This indicates the speed at which electromagnetic waves propagate in the current propagation medium. This indicates that the partial discharge signal has propagated to the reference sensor. The time required This indicates that the partial discharge signal has propagated to the reference sensor. and propagation to the sensor Time difference, the The coordinates of the UHF sensors are , ;

[0018] Based on the time-domain full waveform data of partial discharge from GIS, the initial discharge intensity E of the partial discharge source is calculated.

[0019] Furthermore, the step of injecting the initial discharge intensity and initial discharge location of the partial discharge source into the simulator for simulation calculation to obtain the simulation results of the partial discharge source includes:

[0020] The initial discharge intensity and initial discharge location of the partial discharge source are injected into the simulator to simulate the partial discharge phenomenon of GIS equipment.

[0021] The simulation enhancement calculation was performed using a simulator to obtain the simulated discharge location and simulated discharge intensity of the partial discharge source.

[0022] Further, the iterative search performed by comparing the measured discharge intensity calculated from the real-time acquired time-domain full waveform data with the simulation enhancement results to obtain the actual discharge location of the partial discharge source includes:

[0023] Based on the simulated discharge location and simulated discharge intensity at each moment, the state transition model corresponding to the process of the ant colony algorithm exploring the actual discharge location of the local discharge source with the initial discharge location of the local discharge source as the center is reconstructed using the Markov decision model.

[0024] Heuristic metrics are generated using a graph neural network-based heuristic learner, transforming the state transition model into a location exploration model that is influenced by heuristic metrics and requires graph traversal.

[0025] The location exploration model is iteratively searched using a neural-guided perturbation-interleaved local search algorithm to obtain the actual discharge location of the local discharge source.

[0026] 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 centered on the initial discharge position of the local discharge source based on the simulated discharge position and simulated discharge intensity at each moment using a Markov decision model includes:

[0027] Based on the simulated discharge location and simulated discharge intensity at each moment, a state space is constructed in the simulator to explore the actual discharge location of the partial discharge source. :

[0028] ;

[0029] ;

[0030] In the formula: Indicates the first in the simulator At that moment, the partial discharge node was discovered. The corresponding state, Indicates the first Simulated discharge position at each moment. Indicates the first The simulated discharge intensity at each moment. This represents the initial discharge intensity of the partial discharge source obtained from the time-domain full waveform data of partial discharge based on GIS. and the initial discharge location As the initial state Indicates the exploration period. ;

[0031] Constructing the action space corresponding to the actual discharge location of the partial discharge source in the simulator. :

[0032] ;

[0033] In the formula: Indicates the first The action at that moment, Greedy strategy selects actions based on standard criteria. ;

[0034] The state transition model corresponding to the process of the ant colony algorithm exploring the actual discharge location of the partial discharge source with the initial discharge location as the center is as follows:

[0035] ;

[0036] In the formula: Indicates the action The following nodes To the node The probability of transition, Indicates the first in the simulator At that moment, the partial discharge node was discovered. In the corresponding state, Indicates that by node Transfer to node The corresponding pheromone concentration, Indicates that by node Transfer to node The corresponding heuristic function, Indicates that by node Transfer to included Nodes in The corresponding pheromone concentration, Indicates that by node Transfer to included Nodes in The corresponding heuristic function, and These represent the pheromone concentration and the weights of the heuristic function, respectively. This represents the set of non-uniform mesh local discharge power nodes that can be selected in the next time step.

[0037] Furthermore, by nodes To the node The reward function for the transfer is , Indicates the first The measured discharge intensity is obtained by processing the time-domain full waveform acquired by the sensor at each moment. Indicates the first The partial discharge intensity obtained from the simulation at each time point.

[0038] Furthermore, the step of using a graph neural network-based heuristic learner to generate heuristic metrics, and converting the state transition model into a location exploration model influenced by heuristic metrics and requiring graph traversal, includes:

[0039] The graph neural network The layer is composed of the first The node and the first Edge features of nodes connected Mapping to Heuristic Metrics ;

[0040] The state transition model is transformed into a location exploration model that is influenced by heuristic metrics and requires T-step graph traversal:

[0041] ;

[0042] In the formula: This represents a location exploration model. Indicates the first in the simulator At that moment, the partial discharge node was discovered. The corresponding state, Indicates the first in the simulator At that moment, the partial discharge node was discovered. The corresponding state, Indicates the exploration period. .

[0043] Furthermore, the graph neural network of the first... The propagation characteristics of the layer are:

[0044] ;

[0045] ;

[0046] In the formula: Indicates the first The first layer The characteristics of each node Indicates the first The first layer The characteristics of each node Indicates the first The first layer The characteristics of each node Indicates the first The layer is composed of the first The node and the first Edge features of node connections Indicates the first The layer is composed of the first The node and the first Edge features of node connections , , , as well as Indicates the first Learnable parameters of the layer This represents the activation function. This indicates batch normalization processing. Indicates the first Aggregate calculations are performed on the neighborhoods of each node. Indicates the first The neighborhood of each node express function, It represents the Hadamah accumulation. .

[0047] Furthermore, before using a graph neural network-based heuristic learner to generate a heuristic metric and converting the state transition model into a location exploration model influenced by the heuristic metric and requiring graph traversal, the method further includes:

[0048] A gradient strategy is used to train a heuristic learner based on a graph neural network. The objective function used during training is... for:

[0049] ;

[0050] In the formula: This indicates that a neural-guided perturbation-interleaved local search algorithm is used to explore the objective function corresponding to the actual discharge location of the local discharge source. Indicates balance and The parameters, In heuristic measurement The expected value of the actual discharge location of the partial discharge source under the influence of [the following factors] This represents the state space corresponding to the actual discharge location of the partial discharge source;

[0051] Wherein, the objective function The gradient is The formula is expressed as:

[0052] ;

[0053] In the formula: This represents the average target value for directly exploring the actual discharge location of the partial discharge source. This represents the average target value used in a local search algorithm with neurally guided perturbation interleaving to explore the actual discharge location of the local discharge source. In heuristic measurement To explore the gradient of the actual discharge location of the partial discharge source under the influence of the influence;

[0054] When the maximum number of iterations is reached Or the maximum number of iterations has not been reached. But the objective function At that time, the training ended, among which, This represents the minimum threshold corresponding to the objective function.

[0055] Furthermore, the iterative search of the location exploration model using the neural-guided perturbation interleaving local search algorithm to obtain the actual discharge location of the local discharge source includes:

[0056] A local search algorithm is used to iteratively search the location exploration model to obtain a local optimum.

[0057] The local optimum solution obtained in the current iteration is subjected to neural-guided perturbation. The optimal exploration scheme for the actual discharge position of the local discharge source in the current iteration is obtained through the interleaved local search of neural-guided perturbation.

[0058] After all local discharge power nodes are selected, update the pheromone concentration between each local discharge power node;

[0059] When the iterative search reaches the iterative convergence condition, the optimal exploration scheme for the actual discharge location of the partial discharge source is determined, and the actual discharge location of the partial discharge source is matched based on the optimal exploration scheme.

[0060] Furthermore, the local search algorithm is used to iteratively search the location exploration model to obtain a local optimum. The formula is expressed as:

[0061] ;

[0062] ;

[0063] In the formula: Describe the objective function. This represents a local search operator. This represents the state space corresponding to the actual discharge location of the partial discharge source. This indicates that a local search is performed on the discharge location under the local search operator. Indicates the first The measured discharge intensity is obtained by processing the time-domain full waveform acquired by the sensor at each moment. Indicates the first The partial discharge intensity obtained from the simulation at each time point Indicates the exploration period. .

[0064] Furthermore, the process of performing neural-guided perturbations on the local optimal solution obtained in the current iteration, and obtaining the optimal exploration scheme for the actual discharge location of the local discharge source in the current iteration through interleaved neural-guided perturbations in the local search, is expressed by the formula:

[0065] ;

[0066] In the formula: This represents the optimal exploration scheme for the actual discharge position of the local discharge source in the current iteration, obtained by neural-guided perturbation of the local optimal solution. This is a locally optimal solution; Indicates the number of moves in the disturbance. This represents a heuristic metric.

[0067] Furthermore, the formula for updating the pheromone concentration is expressed as:

[0068] ;

[0069] In the formula: Indicates the update coefficients. Indicates the first Pheromon concentration within an exploration cycle Indicates the first Pheromon concentration within an exploration cycle Indicates the first The first exploration cycle to the first The change in pheromone concentration during an exploration cycle.

[0070] Furthermore, the simulator is a GIS 3D simulation module, and the simulator is divided into a non-uniform grid.

[0071] Furthermore, after iteratively searching the measured discharge intensity calculated from the real-time acquired time-domain full waveform data with the simulation enhancement results to obtain the actual discharge location of the partial discharge source, the method further includes:

[0072] An early warning is issued based on the actual discharge location of the partial discharge source.

[0073] Furthermore, this application also proposes an optimal location matching system for partial discharge in GIS, the system comprising:

[0074] The coarse positioning module is used to coarsely locate the partial discharge source based on the time-domain full waveform data of GIS partial discharge, and obtain the initial discharge intensity and initial discharge location of the partial discharge source.

[0075] The simulation module is used to inject the initial discharge intensity and initial discharge location of the partial discharge power source into the simulator for simulation enhancement calculation, and obtain the simulation enhancement result of the partial discharge power source, which includes the simulated discharge location and simulated discharge intensity.

[0076] The iterative search module is used to iteratively search the measured discharge intensity calculated from the real-time acquired time-domain full waveform data with the simulation enhancement results to obtain the actual discharge location of the partial discharge source.

[0077] Furthermore, this application also proposes an industrial control computer, which includes a memory and a processor; wherein the processor reads executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the GIS partial discharge optimal location matching method as described above.

[0078] In addition, this application also proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the GIS partial discharge optimal location matching method as described above.

[0079] Furthermore, this application also proposes a multi-structure adapted GIS field-electric fusion real-time status perception and early warning system, including a signal synchronous acquisition and measurement embedded device and an industrial control computer as described above. The industrial control computer is connected to the signal synchronous acquisition and measurement embedded device, which is used to acquire time-domain full waveform data of GIS partial discharge, and the industrial control computer is used to search for the actual discharge location of the GIS partial discharge power source.

[0080] (1) This application firstly performs coarse localization of the partial discharge source based on the time-domain full waveform data of GIS partial discharge, and obtains the initial discharge intensity and initial discharge position of the partial discharge source. The initial discharge intensity and initial discharge position of the partial discharge source are injected into the simulator to simulate the partial discharge phenomenon of GIS equipment and perform simulation calculations to obtain the simulation results of the changes in partial discharge position and partial discharge intensity. Then, based on the analysis of the simulation calculation results, the difference between the measured discharge intensity obtained by the time-domain full waveform data of GIS partial discharge and the simulation results is compared, and iterative optimization is continuously performed until the difference between the measured value and the simulation value is minimized, thereby achieving the optimal position matching of GIS partial discharge.

[0081] (2) This application takes the preliminarily calculated local discharge source as the center and proposes a heuristic fast iterative search algorithm for the optimal matching of the measurement value 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, which is used as a substitute for the expert design heuristic. Based on this, the ant colony algorithm is iterated. Under the framework of the ant colony algorithm, the 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. Combined with probabilistic local search, the local search method with neural guidance perturbation will achieve better performance to ensure efficient solution and realize the fast iterative search of the optimal matching position of the discharge source. Attached Figure Description

[0082] Figure 1 is a flowchart illustrating an embodiment of a GIS partial discharge optimal location matching method proposed in this application;

[0083] Figure 2 is a block diagram of the principle of optimal location matching of partial discharge in GIS based on deep ant colony algorithm in one embodiment of this application;

[0084] Figure 3 is a schematic diagram of the GIS partial discharge optimal location search process based on deep ant colony algorithm in one embodiment of this application;

[0085] Figure 4 is a schematic diagram of the structure of a GIS partial discharge optimal location matching system proposed in an embodiment of this application. Detailed Implementation

[0086] As shown in Figure 1, the first embodiment of this application proposes a method for optimal location matching of partial discharge in GIS, the method comprising the following steps:

[0087] S10. Based on the time-domain full waveform data of partial discharge from GIS, the partial discharge source is coarsely located to obtain the initial discharge intensity and initial discharge location of the partial discharge source.

[0088] This embodiment can use a UHF sensor or other data acquisition GIS equipment to obtain partial discharge signals and acquire UHF time-domain full waveform data during GIS equipment operation. Then, methods such as Time Difference of Arrival (TDOA) are used to coarsely locate the UHF time-domain full waveform data to obtain the initial discharge intensity and initial discharge location of the partial discharge source.

[0089] S20. The initial discharge intensity and initial discharge location of the partial discharge source are injected into the simulator for simulation enhancement calculation to obtain the simulation enhancement result of the partial discharge source. The simulation enhancement result includes the simulated discharge location and the simulated discharge intensity.

[0090] By injecting the initial discharge intensity and initial discharge location of the partial discharge source into the GIS simulator for simulation calculation, the simulation results of the partial discharge source, including the changing partial discharge location and partial discharge intensity, can be obtained.

[0091] S30. The measured discharge intensity calculated from the real-time collected time-domain full waveform data is iteratively searched with the simulation enhancement results to obtain the actual discharge location of the partial discharge source.

[0092] This embodiment compares the measured discharge intensity obtained from 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 location matching of GIS partial discharge.

[0093] As a technical solution, step S10: coarsely locating the partial discharge source based on the time-domain full waveform data of GIS partial discharge to obtain the initial discharge intensity and initial discharge location of the partial discharge source, including the following steps:

[0094] S11. Based on the time-domain full waveform data of partial discharge from GIS, calculate the initial discharge location of the partial discharge source. The formula is expressed as:

[0095] ;

[0096] ;

[0097] In the formula, Indicates the first The distance from the sensor to the local discharge power source This indicates the speed at which electromagnetic waves propagate in the current propagation medium. This indicates that the partial discharge signal has propagated to the reference sensor. The time required This indicates that the partial discharge signal has propagated to the reference sensor. and propagation to the sensor Time difference, the The coordinates of the UHF sensors are , .

[0098] S12. Calculate the initial discharge intensity E of the partial discharge source based on the time-domain full waveform data of GIS partial discharge.

[0099] This embodiment obtains the partial discharge intensity by applying a specific relationship between the UHF signal and the maximum value of the actual discharge.

[0100] As a technical solution, step S20, which involves injecting the initial discharge intensity and initial discharge location of the partial discharge power source into a simulator for simulation calculation to obtain the simulation results of the partial discharge power source, includes the following steps:

[0101] S21. The initial discharge intensity and initial discharge location of the partial discharge power source are injected into the simulator to simulate the partial discharge phenomenon of the GIS equipment.

[0102] S22. Use a simulator to perform simulation enhancement calculations to obtain the simulated discharge location and simulated discharge intensity of the partial discharge source.

[0103] In this embodiment, the initial discharge intensity and initial discharge location of the partial discharge source are injected into the GIS simulator, and then simulation calculations are performed using simulation software to obtain the simulated discharge location and simulated discharge intensity of the partial discharge source.

[0104] The GIS simulator used in this embodiment employs a non-uniform grid with varying coarseness. The initial discharge location and intensity of the local discharge source are injected into the corresponding grid cells of the non-uniform grid. The non-uniform grid simulator enhances the coarse grid simulation data to obtain enhanced fine grid data, resulting in more refined simulation of the local discharge nodes and more accurate positioning. It should be understood that the GIS simulator in this embodiment can also use a uniform grid; this embodiment does not specifically limit its use.

[0105] As a technical solution, step S30 involves iteratively searching the measured discharge intensity calculated from the real-time acquired time-domain full waveform data with the simulation enhancement results to obtain the actual discharge location of the partial discharge source, including the following steps:

[0106] S31. Based on the simulated discharge position and simulated discharge intensity at each moment, reconstruct the state transition model corresponding to the process of the ant colony algorithm exploring the actual discharge position of the partial discharge source with the initial discharge position of the partial discharge source as the center using the Markov decision model.

[0107] In this embodiment, the initial discharge position is taken as the center in the non-uniform grid of the simulator. Based on the simulated discharge position and simulated discharge intensity at each moment, the state space and action space corresponding to the exploration of the actual discharge position of the partial discharge source in the non-uniform grid with the initial discharge position as the center are constructed. Thus, the state transition model corresponding to the process of exploring the actual discharge position of the partial discharge source with the initial discharge position as the center is constructed.

[0108] S32. Use a graph neural network-based heuristic learner to generate a heuristic metric, and convert the state transition model into a location exploration model that is affected by the heuristic metric and requires graph traversal.

[0109] S33. The location exploration model is iteratively searched using a neural-guided perturbation interleaved local search algorithm to obtain the actual discharge location of the local discharge source.

[0110] As shown in Figure 2, this embodiment uses a data-driven approach to automatically design and enhance the heuristic rules in the ant colony algorithm, thereby more effectively solving the rapid iterative search for the optimal matching position of the discharge source. The deep ant colony algorithm uses graph neural networks to generate heuristic metrics, reducing the need for expert knowledge, and combines probabilistic local search to achieve better performance and ensure efficient solution.

[0111] As a technical solution, step S31: Based on the simulated discharge position and simulated discharge intensity at each moment, a Markov decision model 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. Specifically, this includes:

[0112] S311. Based on the simulated discharge position and simulated discharge intensity at each moment, construct the state space corresponding to the exploration of the actual discharge position of the partial discharge source in the simulator. :

[0113] ;

[0114] ;

[0115] In the formula: Indicates the first in the simulator At that moment, the partial discharge node was discovered. The corresponding state, Indicates the first Simulated discharge position at each moment. Indicates the first The simulated discharge intensity at each moment. This represents the initial discharge intensity of the partial discharge source obtained from the time-domain full waveform data of partial discharge based on GIS. and the initial discharge location As the initial state Indicates the exploration period. ;

[0116] S312. Construct the action space corresponding to the actual discharge location of the partial discharge source in the simulator. :

[0117] ;

[0118] In the formula: Indicates the first The action at that moment, Greedy strategy selects actions based on standard criteria. .

[0119] S313. The state transition model corresponding to the process of the ant colony algorithm exploring the actual discharge position of the partial discharge source with the initial discharge position of the partial discharge source as the center is as follows:

[0120] ;

[0121] In the formula: Indicates the action The following nodes To the node The probability of transition, Indicates the first in the simulator At that moment, the partial discharge node was discovered. In the corresponding state, Indicates that by node Transfer to node The corresponding pheromone concentration, Indicates that by node Transfer to node The corresponding heuristic function, Indicates that by node Transfer to included Nodes in The corresponding pheromone concentration, Indicates that by node Transfer to included Nodes in The corresponding heuristic function, and These represent the pheromone concentration and the weights of the heuristic function, respectively. This represents the set of non-uniform mesh local discharge power nodes that can be selected in the next time step.

[0122] By node To the node The reward function for the transfer is , Indicates the first The measured discharge intensity is obtained by processing the time-domain full waveform acquired by the sensor at each moment. Indicates the first The partial discharge intensity obtained from the simulation at each time point.

[0123] This embodiment utilizes Markov models to formalize and model the state transition process in the ant colony algorithm. This modeling helps to understand and analyze the ant colony algorithm's search process in the solution space, including key steps such as ant path selection and pheromone updates.

[0124] As a technical solution, step S32: using a graph neural network-based heuristic learner to generate a heuristic metric, and converting the state transition model into a location exploration model that is influenced by the heuristic metric and requires graph traversal, includes the following steps:

[0125] S321. Using a multilayer perceptron to extract the graph neural network's first... The layer is composed of the first The node and the first Edge features of nodes connected Mapping to Heuristic Metrics .

[0126] S322. The state transition model is converted into a location exploration model that is influenced by heuristic metrics and requires T-step graph traversal.

[0127] ;

[0128] In the formula: This represents a location exploration model. Indicates the first in the simulator At that moment, the partial discharge node was discovered. The corresponding state, Indicates the first in the simulator At that moment, the partial discharge node was discovered. The corresponding state, Indicates the exploration period. .

[0129] This embodiment constructs a heuristic learner based on graph neural networks to heuristically parameterize the exploration of local discharge sources in GIS. By using graph neural networks to generate heuristic metrics and combining them with probabilistic local search, better performance is achieved to ensure efficient solution.

[0130] The propagation characteristics of the l-th layer of the graph neural network are as follows:

[0131] ;

[0132] ;

[0133] In the formula: Indicates the first The first layer The characteristics of each node Indicates the first The first layer The characteristics of each node Indicates the first The first layer The characteristics of each node Indicates the first The layer is composed of the first The node and the first Edge features of node connections Indicates the first The layer is composed of the first The node and the first Edge features of node connections , , , as well as Indicates the first Learnable parameters of the layer This represents the activation function. This indicates batch normalization processing. Indicates the first Aggregate calculations are performed on the neighborhoods of each node. Indicates the first The neighborhood of each node express function, It represents the Hadamah accumulation. .

[0134] When exploring the location of partial discharges, the ant colony algorithm constructs a graph by traversing the discharge nodes, and the nodes of the graph neural network correspond to the discharge nodes.

[0135] As a technical solution, before step S32: generating a heuristic metric using a graph neural network-based heuristic learner and converting the state transition model into a location exploration model influenced by the heuristic metric and requiring graph traversal, the method further includes the following steps:

[0136] A gradient strategy is used to train a heuristic learner based on a graph neural network. The objective function used during training is... for:

[0137] ;

[0138] In the formula: This indicates that a neural-guided perturbation-interleaved local search algorithm is used to explore the objective function corresponding to the actual discharge location of the local discharge source. Indicates balance and The parameters, In heuristic measurement The expected value of the actual discharge location of the partial discharge source under the influence of [the following factors] This represents the state space corresponding to the actual discharge location of the partial discharge source. This represents the objective function.

[0139] Wherein, the objective function The gradient is The formula is expressed as:

[0140] ;

[0141] In the formula: This represents the average target value for directly exploring the actual discharge location of the partial discharge source. This represents the average target value used in a local search algorithm with neurally guided perturbation interleaving to explore the actual discharge location of the local discharge source. In heuristic measurement The gradient of the actual discharge location of the partial discharge source was explored under the influence of the influence.

[0142] When the maximum number of iterations is reached Or the maximum number of iterations has not been reached. But the objective function At that time, the training ended, among which, This represents the minimum threshold corresponding to the objective function.

[0143] As a technical solution, step S33: using a neural-guided perturbation interleaved local search algorithm to iteratively search the location exploration model to obtain the actual discharge location of the local discharge source, specifically includes the following steps:

[0144] S331. The location exploration model is iteratively searched using a local search algorithm to obtain a local optimum.

[0145] Local optimal solution The formula is expressed as:

[0146] ;

[0147] ;

[0148] In the formula: Describe the objective function. This represents a local search operator. This represents the state space corresponding to the actual discharge location of the partial discharge source. This indicates that a local search is performed on the discharge location under the local search operator. Indicates the first The measured discharge intensity is obtained by processing the time-domain full waveform acquired by the sensor at each moment. Indicates the first The partial discharge intensity obtained from the simulation at each time point Indicates the exploration period. .

[0149] S332. The local optimal solution obtained in the current iteration is subjected to neural-guided perturbation. The optimal exploration scheme for the actual discharge position of the local discharge source in the current iteration is obtained through the interleaved local search of neural-guided perturbation.

[0150] The optimal exploration scheme for the actual discharge location of the partial discharge source obtained in the previous iteration is expressed by the following formula:

[0151] ;

[0152] In the formula: This represents the optimal exploration scheme for the actual discharge position of the local discharge source in the current iteration, obtained by neural-guided perturbation of the local optimal solution. This is a locally optimal solution; Indicates the number of moves in the disturbance. This represents a heuristic metric.

[0153] The optimal exploration scheme for the actual discharge location of the local discharge source obtained in the current iteration, based on the perturbation, is then subjected to a local search to obtain the optimal solution for the current iteration. Local optimal solution obtained by local search :

[0154] ;

[0155] go through The optimal exploration scheme for GIS local discharge sources was obtained through a series of iterations of neural-guided perturbation-interleaved local search. :

[0156] ;

[0157] In the formula: This represents the set of parameters or variables that make the function achieve its minimum value.

[0158] The location exploration model is iteratively searched using a local search algorithm to obtain a local optimum. Local optima may occur. In this embodiment, a heuristic metric obtained from training a graph neural network is used to perturb the local optimum. The heuristic metric can help the algorithm quickly locate and escape local optima, guide the search process toward the global optimum, and guide the search process to explore the solution space more effectively.

[0159] S333. After all local discharge power nodes are selected, update the pheromone concentration between each local discharge power node.

[0160] The formula for updating pheromone concentration is expressed as:

[0161] ;

[0162] In the formula: Indicates the update coefficients. Indicates the first Pheromon concentration within an exploration cycle Indicates the first Pheromon concentration within an exploration cycle Indicates the first The first exploration cycle to the first The change in pheromone concentration during an exploration cycle.

[0163] S333. When the iterative search reaches the iterative convergence condition, determine the optimal exploration scheme for the actual discharge location of the partial discharge power source, and achieve the matching of the actual discharge location of the partial discharge power source based on the optimal exploration scheme, wherein the iterative convergence condition can be a set number of iterations.

[0164] In this embodiment, the ant colony algorithm simulates the search process of ants in a graph to find possible locations of partial discharge sources. Graph neural networks can effectively propagate partial discharge signal information through message passing and feature aggregation, and combine the search results from the ant colony algorithm to infer and locate the source. The parallel computing power and efficient feature learning of graph neural networks enable them to quickly process large-scale graph data, making them suitable for real-time or near-real-time partial discharge localization tasks. As a heuristic optimization method, the ant colony algorithm, combined with the feature learning capabilities of graph neural networks, can intelligently guide the search process and quickly discover possible discharge source locations. This method effectively avoids getting trapped in local optima and improves global search capabilities, enabling fast, real-time, and high-precision localization.

[0165] As shown in Figure 3, this embodiment can determine the optimal exploration scheme by taking the initial discharge position as the center and the radius of the spherical area for partial discharge positioning as the radius of the GIS cross section, and then match the actual discharge position of the partial discharge source according to the path of the optimal exploration scheme, thus enabling rapid positioning of the partial discharge source.

[0166] As a technical solution, in step S30: after iteratively searching the measured discharge intensity calculated from the real-time acquired time-domain full waveform data with the simulation enhancement result to obtain the actual discharge location of the partial discharge source, the method further includes the following steps:

[0167] An early warning is issued based on the actual discharge location of the partial discharge source.

[0168] Based on the optimal location matching method for partial discharge in GIS proposed in this embodiment, rapid early warning of GIS faults can be achieved, fault problems can be detected in a timely and accurate manner, and accidents can be avoided.

[0169] Furthermore, as shown in Figure 4, the second embodiment of this application proposes a GIS partial discharge optimal location matching system. The system includes: a coarse positioning module 10, used to coarsely locate the partial discharge source based on the time-domain full waveform data of GIS partial discharge, and obtain the initial discharge intensity and initial discharge location of the partial discharge source; a simulation module 20, used to inject the initial discharge intensity and initial discharge location of the partial discharge source into a simulator for simulation enhancement calculation, and obtain the simulation enhancement result of the partial discharge source, the simulation enhancement result including the simulation discharge location and simulation discharge intensity; and an iterative search module 30, used to iteratively search the measured discharge intensity calculated by the real-time collected time-domain full waveform data with the simulation enhancement result to obtain the actual discharge location of the partial discharge source.

[0170] As a technical solution, the simulation module 20 includes: an initial position injection unit, used to inject the initial discharge intensity and initial discharge position of the partial discharge power source into the GIS simulator to simulate the partial discharge phenomenon of the GIS equipment; and a simulation calculation unit, used to perform simulation enhancement calculations using the simulator to obtain the simulated discharge position and simulated discharge intensity of the partial discharge power source.

[0171] As a technical solution, the iterative search module 30 includes: a state transition model construction unit, used to reconstruct the 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 using a Markov decision model based on the simulated discharge position and simulated discharge intensity at each moment; a heuristic space parameterization unit, used to generate heuristic metrics using a heuristic learner based on graph neural networks, and convert the state transition model into a position exploration model that is affected by heuristic metrics and requires graph traversal; and an iterative search unit, used to iteratively search the position exploration model using a local search algorithm with neural-guided perturbation interleaving to obtain the actual discharge position of the local discharge source.

[0172] As a technical solution, the state transition model construction unit is specifically used to: construct the state space corresponding to the actual discharge location of the partial discharge source in the simulator, based on the simulated discharge location and simulated discharge intensity at each moment. :

[0173] ;

[0174] ;

[0175] In the formula: Indicates the first in the simulator At that moment, the partial discharge node was discovered. The corresponding state, Indicates the first Simulated discharge position at each moment. Indicates the first The simulated discharge intensity at each moment. This represents the initial discharge intensity of the partial discharge source obtained from the time-domain full waveform data of partial discharge based on GIS. and the initial discharge location As the initial state Indicates the exploration period. ;

[0176] Constructing the action space corresponding to the actual discharge location of the partial discharge source in the simulator. :

[0177] ;

[0178] In the formula: Indicates the first The action at that moment, Greedy strategy selects actions based on standard criteria. ;

[0179] The state transition model corresponding to the process of the ant colony algorithm exploring the actual discharge location of the partial discharge source with the initial discharge location as the center is as follows:

[0180] ;

[0181] In the formula: Indicates the action The following nodes To the node The probability of transition, Indicates the first in the simulator At that moment, the partial discharge node was discovered. In the corresponding state, Indicates that by node Transfer to node The corresponding pheromone concentration, Indicates that by node Transfer to node The corresponding heuristic function, Indicates that by node Transfer to included Nodes in The corresponding pheromone concentration, Indicates that by node Transfer to included Nodes in The corresponding heuristic function, and These represent the pheromone concentration and the weights of the heuristic function, respectively. This represents the set of non-uniform mesh local discharge power nodes that can be selected in the next time step.

[0182] As a technical solution, the heuristic spatial parameterization unit is specifically used for: […]. The layer is composed of the first The node and the first Edge features of nodes connected Mapping to Heuristic Metrics ;

[0183] The state transition model is transformed into one that is influenced by heuristic metrics and requires... The step-by-step traversal location exploration model is as follows:

[0184] ;

[0185] In the formula: This represents a location exploration model. Indicates the first in the simulator At that moment, the partial discharge node was discovered. The corresponding state, Indicates the first in the simulator At that moment, the partial discharge node was discovered. The corresponding state, Indicates the exploration period. .

[0186] As a technical solution, the iterative search unit is specifically used for: iteratively searching the location exploration model using a local search algorithm to obtain a local optimum; performing neural-guided perturbation on the local optimum obtained in the current iteration search, and obtaining the optimal exploration scheme for the actual discharge location of the local discharge power source in the current iteration through interleaved local searches of neural-guided perturbation; updating the pheromone concentration among the local discharge power source nodes after all local discharge power source nodes have been selected; determining the optimal exploration scheme for the actual discharge location of the local discharge power source when the iterative search reaches the iterative convergence condition, and achieving matching of the actual discharge location of the local discharge power source based on the optimal exploration scheme.

[0187] As a technical solution, the GIS partial discharge optimal location matching system also includes a training module for training a graph neural network-based heuristic learner using a gradient strategy.

[0188] This embodiment is based on the deep ant colony algorithm. In the case of partial discharge in GIS, a heuristic fast iterative search algorithm is proposed to achieve the optimal matching of the measured values ​​in the neighborhood of the initially calculated discharge source, so as to realize the real-time high-precision positioning of the discharge source.

[0189] Other embodiments or implementation methods of the GIS partial discharge optimal location matching system described in this application can refer to the above-described method embodiments, and will not be repeated here.

[0190] Furthermore, the third embodiment of this application also proposes an industrial control computer, which includes a memory and a processor; wherein the processor reads executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the GIS partial discharge optimal location matching method as described in the first embodiment above.

[0191] Furthermore, the fourth embodiment of this application also proposes a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the GIS partial discharge optimal location matching method as described in the first embodiment above.

[0192] The industrial control computer or computer-readable storage medium provided above can be used to execute the GIS partial discharge optimal location matching method provided in the above embodiments, and has corresponding functions and beneficial effects.

[0193] Furthermore, the fifth embodiment of this application also proposes a multi-structure adapted GIS field-electric fusion real-time status perception and early warning system, including a signal synchronous acquisition and measurement embedded device and an industrial control computer as described above. The industrial control computer is connected to the signal synchronous acquisition and measurement embedded device, which is used to acquire time-domain full waveform data of GIS partial discharge, and the industrial control computer is used to search for the actual discharge location of the GIS partial discharge power source.

[0194] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0195] The various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0196] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0197] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for optimal location matching of partial discharge in GIS, comprising: Based on the time-domain full waveform data of partial discharge from GIS, the partial discharge source is coarsely located to obtain the initial discharge intensity and initial discharge location of the partial discharge source. The initial discharge intensity and initial discharge location of the partial discharge source are injected into the simulator for simulation enhancement calculation to obtain the simulation enhancement result of the partial discharge source, which includes the simulated discharge location and simulated discharge intensity. The actual discharge location of the partial discharge source is obtained by iteratively searching the measured discharge intensity calculated from the real-time collected time-domain full waveform data and the simulation enhancement results.

2. The GIS partial discharge optimal position matching method of claim 1, wherein, The time-domain full waveform data of partial discharge based on GIS is used to coarsely locate the partial discharge source, obtaining the initial discharge intensity and initial discharge location of the partial discharge source, including: Based on the GIS partial discharge time domain full waveform data, the preliminary discharge position of the partial discharge source is calculated The formula is expressed as: ； ； In the formulae, represents the 1 the distance of the sensor to the partial discharge source, denotes the propagation speed of the electromagnetic wave in the current propagation medium, representing partial discharge signal propagation to reference sensor time needed, representing partial discharge signal propagation to reference sensor and propagate to the sensor time difference, the first The coordinates of the UHF sensor are ， ； Based on the time-domain full waveform data of partial discharge from GIS, the initial discharge intensity E of the partial discharge source is calculated.

3. The GIS partial discharge optimal position matching method of claim 1, wherein, The process of injecting the initial discharge intensity and initial discharge location of the partial discharge source into the simulator for simulation calculation to obtain the simulation results of the partial discharge source includes: The initial discharge intensity and initial discharge location of the partial discharge source are injected into the simulator to simulate the partial discharge phenomenon of GIS equipment. The simulation enhancement calculation was performed using a simulator to obtain the simulated discharge location and simulated discharge intensity of the partial discharge source.

4. The GIS partial discharge optimal position matching method of claim 1, wherein, The step of iteratively searching the measured discharge intensity calculated from the real-time acquired time-domain full waveform data with the simulation enhancement results to obtain the actual discharge location of the partial discharge source includes: Based on the simulated discharge location and simulated discharge intensity at each moment, the state transition model corresponding to the process of the ant colony algorithm exploring the actual discharge location of the local discharge source with the initial discharge location of the local discharge source as the center is reconstructed using the Markov decision model. Heuristic metrics are generated using a graph neural network-based heuristic learner, transforming the state transition model into a location exploration model that is influenced by heuristic metrics and requires graph traversal. The location exploration model is iteratively searched using a neural-guided perturbation-interleaved local search algorithm to obtain the actual discharge location of the local discharge source.

5. The GIS partial discharge optimal position matching method of claim 4, wherein, 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 centered on the initial discharge position of the local discharge source, based on the simulated discharge position and simulated discharge intensity at each moment using a Markov decision model, includes: According to the simulation discharge position and the simulation discharge intensity at each time, a state space corresponding to the exploration of the actual discharge position of the partial discharge source in the simulator is constructed ： ； ； In the formulae: represents the first momentary exploration to partial discharge source node corresponding state, represents the 1 a simulated discharge location at a moment in time, represents the 1 a simulated discharge intensity at a moment, represents the preliminary discharge intensity of the partial discharge source obtained based on the GIS partial discharge time-domain full waveform data and the preliminary discharge position As an initial state, denotes an exploration period, ； Action space corresponding to the exploration of the actual discharge location of a partial discharge source built in a simulator ： ； In the formulae: represents the 1 at a time, to A greedy strategy is used to select the action as standard ； The state transition model corresponding to the process of the ant colony algorithm exploring the actual discharge location of the partial discharge source with the initial discharge location as the center is as follows: ； In the formulae: indicates in the action Down by the node To node the probability of metastasis, represents the first momentary exploration to partial discharge source node The corresponding state, represents a node Transfer to node corresponding pheromone concentration, represents a node Transfer to node a corresponding heuristic function, represents a node transferred to a container comprising node in the network corresponding pheromone concentration, represents a node transferred to a container comprising node in the network a corresponding heuristic function, With respectively denote the pheromone concentration and the weight of the heuristic function, This represents the set of non-uniform mesh local discharge power nodes that can be selected in the next time step.

6. The GIS partial discharge optimal position matching method of claim 5, wherein, The reward function of the transition from node to node is , where is the measured discharge intensity at the th time point obtained by time-domain full waveform processing of the sensor, is the simulated partial discharge intensity at the th time point.

7. The GIS partial discharge optimal position matching method of claim 4, wherein, The step of using a graph neural network-based heuristic learner to generate heuristic metrics, transforming the state transition model into a location exploration model influenced by heuristic metrics and requiring graph traversal, includes: The graph neural network is trained layer of the first material a node and a first Edge features of a node connection Mapping to heuristic metrics ； The state transition model is transformed into a location exploration model that is influenced by heuristic metrics and requires T-step graph traversal: ； In the formulae: position exploration model, represents the first momentary exploration to partial discharge source node corresponding state, represents the first momentary exploration to partial discharge source node corresponding state, denotes an exploration period, 。 8. The GIS partial discharge optimal position matching method of claim 4, wherein, the graph neural network is configured to propagation properties of the layers are: ； ； In the formulae: represents the 1 layer of the first characteristics of the nodes, represents the 1 layer of the first characteristics of the nodes, represents the 1 layer of the first characteristics of the nodes, represents the 1 layer of the first material a node and a first edge features of a node connection, represents the 1 layer of the first material a node and a first edge features of a node connection, 、 、 、 and represents the 1 learnable parameters of the layer, denotes an activation function, denotes a batch normalization process, represents a group of the formula (A) or (B) aggregating calculations by the neighborhood of a node, represents the 1 a neighborhood of a node, indicates function, denotes a Hadamard product, 。 9. The GIS partial discharge optimal location matching method as described in claim 8, further comprising, before generating a heuristic metric using a graph neural network-based heuristic learner and converting the state transition model into a location exploration model influenced by the heuristic metric and requiring graph traversal: The gradient strategy is adopted to train the heuristic learner based on the graph neural network, and the target function adopted in the training process for: ； In the formulae: representing a local search algorithm using neural-guided perturbation interleaving to explore the objective function corresponding to the actual discharge location of the partial discharge source, represents equilibrium With parameters of the user, representing the heuristic metric to explore the expected value of the actual discharge location of the partial discharge source, represents a state space corresponding to the case of exploring the actual discharge location of the partial discharge source, Represent the objective function; wherein the objective function the gradient of the function f is The formula is expressed as: ； In the formulae: representing an average target value for directly exploring the actual discharge location of the partial discharge source, representing the average target value of the actual discharge location of the partial discharge source explored by the local search algorithm using the neural-guided perturbation interleaving, representing the heuristic metric To explore the gradient of the actual discharge location of the partial discharge source under the influence of the influence; when the maximum number of iterations is reached or the maximum number of iterations is not reached But the objective function At this time, the training ends, wherein, This represents the minimum threshold corresponding to the objective function.

10. The GIS partial discharge optimal position matching method of claim 4, wherein, The local search algorithm using neurally guided perturbation interleaving is used to iteratively search the location exploration model to obtain the actual discharge location of the local discharge source, including: A local search algorithm is used to iteratively search the location exploration model to obtain a local optimum. The local optimum solution obtained in the current iteration is subjected to neural-guided perturbation. The optimal exploration scheme for the actual discharge position of the local discharge source in the current iteration is obtained through the interleaved local search of neural-guided perturbation. After all local discharge power nodes are selected, update the pheromone concentration between each local discharge power node; When the iterative search reaches the iterative convergence condition, the optimal exploration scheme for the actual discharge location of the partial discharge source is determined, and the actual discharge location of the partial discharge source is matched based on the optimal exploration scheme.

11. The GIS partial discharge optimal position matching method of claim 10, wherein, The local search algorithm is used to iteratively search the position exploration model to obtain a local optimal solution The formula is represented as: ； ； In the formulae: representing an objective function, denotes a local search operator, represents a state space corresponding to the case of exploring the actual discharge location of the partial discharge source, denotes a local search of the discharge location under a local search operator, represents the 1 The measured discharge intensity obtained by processing the time-domain full waveform collected by the sensor at the moment, represents the 1 the partial discharge intensity obtained from the time-domain simulation, denotes an exploration period, 。 12. The GIS partial discharge optimal position matching method of claim 10, wherein, The process involves neurally guided perturbation of the local optimum obtained in the current iteration, and then using interleaved neurally guided perturbations to obtain the optimal exploration scheme for the actual discharge location of the local discharge source in the current iteration. The formula is as follows: ； In the formulae: represents an optimal exploration scheme of the current iteration local discharge source actual discharge position obtained by performing neural guided perturbation on the local optimal solution, for local optimal solution; the number of moves representing the perturbation, This represents a heuristic metric.

13. The GIS partial discharge optimal position matching method of claim 10, wherein, The formula for updating the pheromone concentration is expressed as: ； In the formulae: representing an update coefficient, represents the 1 pheromone concentration within an exploration period, represents the 1 pheromone concentration within an exploration period, represents the 1 one exploration cycle to the first The change in pheromone concentration during an exploration cycle.

14. The GIS partial discharge optimal position matching method of claim 1, wherein, The simulator is a GIS 3D simulation module, and the simulator is divided into a non-uniform grid.

15. The GIS partial discharge optimal position matching method of claim 1, wherein, After iteratively searching the measured discharge intensity calculated from the real-time acquired time-domain full waveform data with the simulation enhancement results to obtain the actual discharge location of the partial discharge source, the method further includes: An early warning is issued based on the actual discharge location of the partial discharge source.

16. A GIS partial discharge optimal location matching system, comprising: The coarse positioning module is used to coarsely locate the partial discharge source based on the time-domain full waveform data of GIS partial discharge, and obtain the initial discharge intensity and initial discharge location of the partial discharge source. The simulation module is used to inject the initial discharge intensity and initial discharge location of the partial discharge power source into the simulator for simulation enhancement calculation, and obtain the simulation enhancement result of the partial discharge power source, which includes the simulated discharge location and simulated discharge intensity. The iterative search module is used to iteratively search the measured discharge intensity calculated from the real-time acquired time-domain full waveform data with the simulation enhancement results to obtain the actual discharge location of the partial discharge source.

17. An industrial computer comprising a memory, a processor; wherein, The processor runs a program corresponding to the executable program code stored in the memory to implement the method as described in any one of claims 1-15.

18. A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method as described in any one of claims 1-15.

19. A GIS multi-structure adaptive field-electric fusion real-time state sensing and early warning system, comprising a signal synchronous acquisition and measurement embedded device and the industrial computer of claim 17, wherein the industrial computer is connected with the signal synchronous acquisition and measurement embedded device, the signal synchronous acquisition and measurement embedded device is used for acquiring time-domain full waveform data of GIS partial discharge, and the industrial computer is used for searching an actual discharge position of a GIS partial discharge source.

Images