An optimization control method and system for power grid operation and maintenance

Abstract

The present invention relates to the technical field of power operation and maintenance, and particularly to an optimized control method and system for power grid operation and maintenance. The method includes the following steps: obtaining transmission line current data, extracting current fluctuation characteristics and transient noise characteristics, and identifying current fluctuation patterns; combining historical operation and maintenance data, performing abnormal feature analysis on the current fluctuation patterns, and generating preliminary operation and maintenance results of the robot; performing electromagnetic spectrum analysis on the transmission line current data, constructing an electromagnetic coupling field simulation environment, and simulating the power grid potential field; comparing the simulated potential field with the operation and maintenance results, analyzing prediction deviations, and tracing the error sources during the operation and maintenance process; based on the error source data, optimizing the operation and maintenance results of the robot to achieve intelligent optimized control of power grid operation and maintenance; the present invention can achieve more reliable and safer optimized control of power grid operation and maintenance robots.

Description

Technical Field

[0001] The present invention relates to the technical field of power operation and maintenance, and particularly to an optimization control method and system for power grid operation and maintenance. Background Art

[0002] In the preliminary automation stage, power grid operation and maintenance robots mainly perform basic automation tasks, such as inspection and simple maintenance work, and use basic sensors (such as infrared and ultrasonic) for data collection and fault detection. However, the drawback of this stage is that the functions of the robots are single, the data processing ability is limited, and they lack intelligence, relying only on predetermined rules and scripts, with poor flexibility. With the development of technology, it has entered the integrated intelligent technology stage, where artificial intelligence and machine learning algorithms are introduced to improve data analysis and fault prediction capabilities. At the same time, a variety of sensors (such as optical, radar, and meteorological) are integrated to achieve multi-dimensional data collection, and wireless communication technology is used to achieve real-time data transmission and remote control. However, the problems of data quality and consistency still exist in this stage. In the multi-source data fusion, the data quality and consistency problems affect the accuracy of the analysis results, and the wireless communication is unstable in some environments, affecting real-time data transmission and control. At the same time, the computing resources mainly rely on the robots themselves, and the edge computing ability is limited, making it difficult to handle complex tasks. After entering the stage of edge computing and cloud computing collaboration, edge computing devices are deployed on the robots to achieve preliminary data processing and real-time analysis, and a cloud computing platform is used for complex data analysis and global optimization, with the edge and the cloud working together. In addition, the application of intelligent path planning algorithms has improved the robot's efficient navigation and task execution capabilities in complex environments. However, the communication delay between edge computing and cloud computing affects tasks with high real-time requirements, and data security and privacy protection are still a challenge during data transmission and storage. The integration complexity of the edge and cloud systems is relatively high, involving the coordination of multiple technologies and protocols. Summary of the Invention

[0003] Based on this, it is necessary to provide an optimization control method and system for power grid operation and maintenance to solve at least one of the above technical problems.

[0004] To achieve the above object, an optimization control method for power grid operation and maintenance includes the following steps:

[0005] Step S1: Obtain the current data of each section of the transmission line; extract the current fluctuation characteristics from the current data of each section of the transmission line to obtain the power grid current fluctuation characteristic data; perform noise oscillation analysis on the current data of each section of the transmission line to obtain the transient fluctuation noise data; perform fluctuation pattern recognition on the power grid current fluctuation characteristic data and the transient fluctuation noise data to generate the current fluctuation pattern;

[0006] Step S2: Obtain the historical operation and inspection database; perform current anomaly feature fusion analysis on the current fluctuation pattern based on the historical operation and inspection database to generate the robot operation and inspection results;

[0007] Step S3: Perform electromagnetic spectrum mapping on the current data of each section of the transmission line to obtain the power grid electromagnetic spectrum data; construct an electromagnetic coupling field for the power grid electromagnetic spectrum data to generate an electromagnetic field simulation environment;

[0008] Step S4: Simulate the power grid potential field for the electromagnetic field simulation environment to generate a simulated potential field; perform measured deviation analysis on the simulated potential field based on the robot operation and inspection results to obtain measured deviation data;

[0009] Step S5: Trace back the error sources of the predicted value deviation data to generate operation and inspection error source data; optimize the robot operation and inspection results based on the operation and inspection error source data to perform optimized control of the power grid operation and inspection robot.

[0010] The present invention extracts current fluctuation characteristics and transient noise characteristics from the current data of the transmission line, and identifies different current fluctuation patterns. This helps to more accurately characterize the current dynamic behavior of the power grid. Using historical operation and inspection data, combined with the current fluctuation pattern, perform current anomaly feature analysis to generate the operation and inspection diagnosis results of the robot. This is very helpful for fault prediction and preventive maintenance. Through electromagnetic spectrum mapping and electromagnetic field modeling, an electromagnetic environment simulation of the power grid is constructed. This provides a physical basis for subsequent potential field simulation. Based on the electromagnetic field simulation, simulate the potential field distribution of the power grid and perform deviation analysis with the measured data. This can help to discover power grid potential anomaly problems. Analyze the root causes of the errors occurring in the operation and inspection, and optimize the diagnosis results of the robot based on this. This is crucial for improving the fault diagnosis accuracy of the robot. It covers the entire process of the power grid operation and maintenance robot from data acquisition, feature extraction, fault diagnosis to optimized control.

[0011] In this specification, an optimized control system for power grid operation and inspection is also provided, which is used to execute the optimized control method for power grid operation and inspection as described above. The optimized control system for power grid operation and inspection includes:

[0012] A current fluctuation identification module, which is used to obtain the current data of each section of the transmission line; extract the current fluctuation characteristics from the current data of each section of the transmission line to obtain the power grid current fluctuation characteristic data; perform noise oscillation analysis on the current data of each section of the transmission line to obtain transient fluctuation noise data; identify the fluctuation pattern for the power grid current fluctuation characteristic data and the transient fluctuation noise data to generate a current fluctuation pattern;

[0013] An operation and inspection analysis module, which is used to obtain the historical operation and inspection database; perform current anomaly feature fusion analysis on the current fluctuation pattern based on the historical operation and inspection database to generate the robot operation and inspection results;

[0014] An electromagnetic field simulation module, which is used to perform electromagnetic spectrum mapping on the current data of each section of the transmission line to obtain power grid electromagnetic spectrum data; and construct an electromagnetic coupling field for the power grid electromagnetic spectrum data to generate an electromagnetic field simulation environment.

[0015] A deviation calculation module, which is used to simulate the power grid potential field for the electromagnetic field simulation environment to generate a simulated potential field; and perform measured deviation analysis on the simulated potential field based on the robot operation and inspection results to obtain measured deviation data.

[0016] A robot optimization module, which is used to trace back the error sources for the predicted value deviation data to generate operation and inspection error source data; and perform operation and inspection optimization on the robot operation and inspection results based on the operation and inspection error source data to execute the optimization control of the power grid operation and inspection robot.

[0017] The beneficial effects of the present invention are as follows: It covers the entire process from data acquisition, feature extraction, fault diagnosis to optimization control, and can better realize the intelligence and automation of power grid operation and maintenance. It makes full use of various data sources such as transmission line current data, historical operation and inspection data, and electromagnetic spectrum data, and realizes a three-dimensional and multi-angle analysis of the power grid state. Through the analysis of current fluctuation characteristics and the fusion of current anomaly characteristics, potential fault hazards of the power grid can be discovered in advance, providing a basis for preventive maintenance. Based on electromagnetic field simulation and potential field simulation, a power grid physical model is established, providing strong support for subsequent deviation analysis and optimization. Through the deviation analysis and error source tracing of the operation and inspection results, the optimization correction of the robot diagnosis results is realized, improving the diagnosis accuracy. By fusing multi-source data features, the current fluctuation pattern is identified, providing a reference for the intelligent judgment and decision-making of the power grid state. It makes full use of various data resources in power grid operation and maintenance, and through multi-step in-depth analysis and optimization, finally realizes the intelligent optimization control of the power grid operation and inspection robot, which is of great significance for improving the reliability and safety of the power grid. Description of the Drawings

[0018] Figure 1 It is a schematic diagram of the step flow of an optimization control method for power grid operation and inspection;

[0019] Figure 2 It is Figure 1 a schematic diagram of the detailed implementation step flow of step S2 in

[0020] Figure 3 It is Figure 1 a schematic diagram of the detailed implementation step flow of step S3 in

[0021] The realization of the purpose, functional characteristics and advantages of the present invention will be further described with reference to the embodiments and the accompanying drawings. Detailed Embodiments

[0022] The technical method of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0023] In addition, the accompanying drawings are only schematic diagrams of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor methods and / or microcontroller methods.

[0024] It should be understood that although terms such as "first" and "second" may be used here to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit can be called the second unit, and similarly the second unit can be called the first unit. The term "and / or" used here includes any and all combinations of one or more of the listed related items.

[0025] To achieve the above object, please refer to Figures 1 to 3 , an optimized control method for power grid operation and maintenance, including the following steps:

[0026] Step S1: Obtain the current data of each section of the transmission line; extract the current fluctuation characteristics from the current data of each section of the transmission line to obtain the power grid current fluctuation characteristic data; analyze the noise oscillation of the current data of each section of the transmission line to obtain the transient fluctuation noise data; identify the fluctuation pattern of the power grid current fluctuation characteristic data and the transient fluctuation noise data to generate the current fluctuation pattern;

[0027] Step S2: Obtain the historical operation and maintenance database; perform current anomaly feature fusion analysis on the current fluctuation pattern based on the historical operation and maintenance database to generate the robot operation and maintenance result;

[0028] Step S3: Perform electromagnetic spectrum mapping on the current data of each section of the transmission line to obtain the power grid electromagnetic spectrum data; construct an electromagnetic coupling field for the power grid electromagnetic spectrum data to generate an electromagnetic field simulation environment;

[0029] Step S4: Simulate the power grid potential field for the electromagnetic field simulation environment to generate a simulated potential field; perform measured deviation analysis on the simulated potential field based on the robot operation and maintenance result to obtain the measured deviation data;

[0030] Step S5: Trace back the error sources of the predicted value deviation data to generate operation and inspection error source data; optimize the operation and inspection results of the robot based on the operation and inspection error source data to perform optimized control of the power grid operation and inspection robot.

[0031] The present invention extracts current fluctuation features and transient noise features from transmission line current data and identifies different current fluctuation patterns. This helps to more accurately characterize the current dynamic behavior of the power grid. Using historical operation and inspection data, combined with current fluctuation patterns, analyze the abnormal current features to generate the operation and inspection diagnosis results of the robot. This is very helpful for fault prediction and preventive maintenance. Through electromagnetic spectrum mapping and electromagnetic field modeling, construct an electromagnetic environment simulation of the power grid. This provides a physical basis for subsequent potential field simulation. Based on the electromagnetic field simulation, simulate the potential field distribution of the power grid and perform deviation analysis with the measured data. This can help to discover potential abnormal problems in the power grid. Analyze the root causes of the errors occurring in the operation and inspection and optimize the diagnosis results of the robot based on this. This is crucial for improving the fault diagnosis accuracy of the robot. It covers the entire process of the power grid operation and maintenance robot from data acquisition, feature extraction, fault diagnosis to optimized control.

[0032] In the embodiment of the present invention, with reference to Figure 1 as described, it is a schematic diagram of the step flow of an optimized control method for power grid operation and inspection of the present invention. In this example, the optimized control method for power grid operation and inspection includes the following steps:

[0033] Step S1: Obtain the current data of each section of the transmission line; extract the current fluctuation features from the current data of each section of the transmission line to obtain the power grid current fluctuation feature data; perform noise oscillation analysis on the current data of each section of the transmission line to obtain the transient fluctuation noise data; perform fluctuation pattern recognition on the power grid current fluctuation feature data and the transient fluctuation noise data to generate a current fluctuation pattern.

[0034] In the embodiment of the present invention, the current data of each section of the transmission line is obtained by installing high-precision current sensors on each transmission line section to collect current data in real time. For the collected current data, current fluctuation features are extracted. Wavelet transform and Fourier transform signal processing techniques are used to extract the fluctuation features in the current data to obtain the power grid current fluctuation feature data. For noise oscillation analysis of the current data, signal denoising techniques such as empirical mode decomposition and adaptive noise cancellation are used to analyze the transient fluctuation noise in the data to obtain the transient fluctuation noise data. The power grid current fluctuation feature data and the transient fluctuation noise data are subjected to fluctuation pattern recognition. Machine learning algorithms such as support vector machines and neural networks are used to perform pattern recognition on the current fluctuation features and noise features to generate a current fluctuation pattern.

[0035] Step S2: Obtain the historical operation and inspection database; perform current anomaly feature fusion analysis on the current fluctuation pattern based on the historical operation and inspection database to generate the robot operation and inspection results;

[0036] In the embodiment of the present invention, the data in the historical operation and inspection database is obtained and cleaned and preprocessed using big data technology. Based on the historical operation and inspection database, current anomaly feature fusion analysis is performed on the current fluctuation pattern generated in the first step. Data mining techniques such as clustering analysis and association rule mining are used to identify the anomaly features in the current fluctuation pattern and generate the robot operation and inspection results. Through the comparative analysis of historical data and the current current fluctuation pattern, anomaly points and anomaly patterns are extracted, providing data support for subsequent electromagnetic spectrum mapping and electric potential field simulation.

[0037] Step S3: Perform electromagnetic spectrum mapping on the current data of each section of the transmission line to obtain grid electromagnetic spectrum data; construct an electromagnetic coupling field for the grid electromagnetic spectrum data to generate an electromagnetic field simulation environment;

[0038] In the embodiment of the present invention, electromagnetic spectrum mapping is performed on the current data of each section of the transmission line. Using electromagnetic simulation software such as COMSOL Multiphysics or Ansys HFSS, the current data is mapped into the electromagnetic spectrum to generate grid electromagnetic spectrum data. Based on the electromagnetic spectrum data, an electromagnetic coupling field is constructed for the entire grid, and the finite element analysis method is used to simulate the electromagnetic field of the grid to generate an electromagnetic field simulation environment. This simulation environment can accurately simulate the electromagnetic field distribution of each node in the grid, providing a basis for the simulation and error analysis of the electric potential field.

[0039] Step S4: Perform grid electric potential field simulation on the electromagnetic field simulation environment to generate a simulated electric potential field; perform measured deviation analysis on the simulated electric potential field based on the robot operation and inspection results to obtain measured deviation data;

[0040] In the embodiment of the present invention, grid electric potential field simulation is performed on the electromagnetic field simulation environment. Using numerical calculation methods such as the finite difference method or the boundary element method, the electric potential distribution in the electromagnetic field is simulated to generate a simulated electric potential field. Based on the robot operation and inspection results generated in the second step, measured deviation analysis is performed on the simulated electric potential field. The simulation results are compared with the actual measurement data to calculate the measured deviation data of the simulated electric potential field. Through deviation analysis, the error sources and improvement directions in the simulation model are identified, providing data support for subsequent error backtracking and operation and inspection optimization.

[0041] Step S5: Perform error source backtracking on the predicted value deviation data to generate operation and inspection error source data; perform operation and inspection optimization on the robot operation and inspection results based on the operation and inspection error source data to perform optimized control of the grid operation and inspection robot.

[0042] In the embodiments of the present invention, for the predicted value deviation data, error source backtracking is carried out, and causal analysis and traceability technology are adopted to identify the main sources and propagation paths of errors, and operation and inspection error source data is generated. Based on the operation and inspection error source data, the operation and inspection results of the robot are optimized. Optimization algorithms such as genetic algorithms or particle swarm optimization are used to adjust the operation and inspection strategies and parameters to reduce errors and improve the accuracy of operation and inspection. Through the cyclic iteration of error source backtracking and operation and inspection optimization, the control method of the power grid operation and inspection robot is continuously improved to ensure its efficient and accurate operation in a complex power grid environment.

[0043] Preferably, step S1 includes the following steps:

[0044] Step S11: The power grid transmission lines are monitored for current along the line by the power grid operation and inspection robot to obtain the current data of each section of the transmission line;

[0045] Step S12: Perform a fast Fourier transform on the current data of each section of the transmission line to generate power grid current frequency domain data;

[0046] Step S13: Extract the current fluctuation characteristics from the power grid current frequency domain data to obtain the power grid current fluctuation characteristic data;

[0047] Step S14: Extract the high-frequency noise from the current data of each section of the transmission line to generate current high-frequency noise data;

[0048] Step S15: Analyze the noise oscillation of the current high-frequency noise data to obtain transient fluctuation noise data;

[0049] Step S16: Identify the fluctuation patterns of the power grid current fluctuation characteristic data and the transient fluctuation noise data to generate the current fluctuation pattern.

[0050] The advantages of the present invention are as follows: By using the power grid operation and inspection robot to monitor the current of the transmission line in real time, detailed current data of each line segment can be obtained, laying a foundation for subsequent frequency domain analysis and feature extraction. Converting the time-domain current data to the frequency domain can reveal the spectral characteristics of the current fluctuation, providing a basis for identifying abnormal current states. Analyzing the frequency domain current data can extract key characteristic parameters reflecting the stability of the power grid operation, providing an important basis for fault prediction and abnormal diagnosis. Stripping the high-frequency noise and transient components from the original current signal helps to reduce interference and improve the accuracy and reliability of feature extraction. Fusing the current fluctuation characteristics and transient noise characteristics can more accurately describe the current operation state of the power grid, providing support for formulating subsequent optimization control strategies. Making full use of the power grid real-time monitoring data, through time-frequency domain analysis, feature extraction and pattern recognition technologies, valuable data support and analysis basis are provided for the fault prediction, abnormal diagnosis and intelligent optimization control of the power grid operation and inspection robot.

[0051] In the embodiments of the present invention, the grid operation and maintenance robot is used to monitor the current along the grid transmission line. High-precision current sensors installed on each section of the transmission line are used to collect the current data of each section of the transmission line in real time. These sensors transmit the data to the central monitoring system through wireless communication technology to ensure the integrity and real-time nature of the data. During the inspection process, the robot scans and records the current data along the line to obtain the current data of each section of the transmission line. Each time the collected data needs to be preliminarily verified and validated to ensure the accuracy and reliability of the data. The fast Fourier transform (FFT) is performed on the current data of each section of the transmission line to convert the time-domain current data into frequency-domain data. FFT is an efficient algorithm that decomposes the time-domain signal into different frequency components to generate the frequency-domain data of the grid current. During the calculation process, the sampling frequency and signal length are considered to ensure the resolution and accuracy of the frequency-domain data. After the calculation is completed, the frequency-domain data is used to further analyze the frequency characteristics in the current signal, providing a data basis for the subsequent steps. The current fluctuation characteristics of the grid current frequency-domain data are extracted using spectrum analysis and signal processing techniques to identify the main frequency components and fluctuation characteristics in the current signal. By analyzing the peaks, bandwidths, and harmonic components in the frequency-domain data, the current fluctuation characteristics are extracted to obtain the grid current fluctuation characteristic data. During the extraction process, the data needs to be smoothed and denoised to improve the accuracy and robustness of the characteristic extraction. These fluctuation characteristic data are used to analyze the change patterns and abnormal conditions of the current signal. The high-frequency noise of the current data of each section of the transmission line is extracted using band-pass filters or wavelet transform signal processing techniques to separate the high-frequency noise components in the current signal. The design of the filter needs to consider the noise frequency range and signal characteristics to ensure that the extracted high-frequency noise data is representative and accurate. The generated current high-frequency noise data is used for subsequent noise analysis and feature extraction. During the extraction process, the filter parameters need to be adjusted and optimized to ensure the filtering effect and calculation efficiency. The noise oscillation of the current high-frequency noise data is analyzed using empirical mode decomposition (EMD) and Hilbert transform signal processing techniques to analyze the transient oscillation characteristics in the noise signal. By decomposing the high-frequency noise signal, its intrinsic mode functions (IMFs) are extracted, and the instantaneous frequency and amplitude of each IMF are calculated to obtain the transient fluctuation noise data. During the analysis process, the mode mixing and endpoint effects during the decomposition process need to be processed to improve the accuracy and stability of the analysis results. These transient fluctuation noise data are used to analyze the time-varying characteristics and oscillation patterns of the noise signal. The fluctuation patterns of the grid current fluctuation characteristic data and the transient fluctuation noise data are identified using support vector machine (SVM) and deep neural network (DNN) machine learning algorithms to identify the fluctuation patterns in the current signal. By constructing a training set and a test set, the characteristic data is classified and identified using supervised learning methods to generate the current fluctuation patterns. During the identification process, the model parameters need to be optimized and adjusted to improve the classification accuracy and the generalization ability of the model.The generated current fluctuation pattern is used to analyze anomalies in the current signal and provide early warning of faults, providing a reference basis for subsequent operation inspection and maintenance.

[0052] Preferably, step S2 includes the following steps:

[0053] Step S21: Obtain historical operation inspection data; construct a timestamp index for the historical operation inspection data to generate a historical operation inspection database;

[0054] Step S22: Perform matching retrieval on the historical operation inspection database based on the current fluctuation pattern to obtain historical matching current fluctuation data;

[0055] Step S23: Extract matching characteristic parameters from the historical operation inspection database according to the historical matching current fluctuation data to obtain a matching characteristic matrix;

[0056] Step S24: Perform abnormal feature fusion mapping on the matching characteristic matrix to generate a robot operation inspection result.

[0057] The advantages of the present invention are as follows: By establishing a historical operation inspection database through timestamp indexing, it provides basic data support for subsequent data analysis and pattern matching. Based on the current fluctuation pattern obtained from real-time monitoring, it quickly retrieves the historical current fluctuation data that matches it in the historical database, laying a foundation for feature extraction and anomaly analysis. For the historical matching current fluctuation data, relevant characteristic parameters are extracted to form a characteristic matrix, providing a key basis for anomaly identification and fault prediction. The extracted characteristic parameters are subjected to multi-dimensional fusion analysis to generate a more comprehensive and accurate robot operation inspection result, providing a basis for optimizing control decisions. It makes full use of the historical operation inspection data of the power grid and provides a solid data support and analysis framework for the power grid operation inspection robot to realize fault prediction, status monitoring, and intelligent optimization control through pattern matching, feature extraction, and abnormal fusion analysis techniques.

[0058] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0059] Step S21: Obtain historical operation inspection data; construct a timestamp index for the historical operation inspection data to generate a historical operation inspection database;

[0060] In an embodiment of the present invention, historical operation and maintenance data is obtained by retrieving historical records stored in the power grid operation and maintenance system, including current data of each section of the transmission line, inspection records of inspection robots, and fault reports. A timestamp index is constructed for these data, and each record is sorted and indexed in chronological order for quick retrieval and analysis. Using a database management system (such as an SQL or NoSQL database), the indexed data is stored as a historical operation and maintenance database. The construction of the database needs to ensure the integrity and consistency of the data and perform appropriate compression and optimization to improve retrieval efficiency and storage utilization.

[0061] Step S22: Perform a matching search on the historical operation and maintenance database based on the current fluctuation pattern to obtain historical matching current fluctuation data;

[0062] In an embodiment of the present invention, a matching search is performed on the historical operation and maintenance database based on the current fluctuation pattern. Using a pattern matching algorithm (such as the dynamic time warping algorithm DTW or similarity measure), records similar to the current current fluctuation pattern are searched in the historical operation and maintenance database. During the matching process, time alignment and scale changes of the data need to be considered to ensure the accuracy and reliability of the matching results. After the matching search is completed, historical matching current fluctuation data is obtained, which contains all data points similar to the current current fluctuation pattern in the historical records.

[0063] Step S23: Extract matching feature parameters from the historical operation and maintenance database according to the historical matching current fluctuation data to obtain a matching feature matrix;

[0064] In an embodiment of the present invention, matching feature parameters are extracted from the historical operation and maintenance database according to the historical matching current fluctuation data. Using a feature extraction algorithm (such as principal component analysis PCA or feature selection method), key feature parameters are extracted from the matched historical data. These feature parameters include the amplitude, frequency, and phase of the current fluctuation. By calculating and analyzing these parameters, a matching feature matrix is constructed. The matching feature matrix is a multi-dimensional data structure that contains the values of all matching feature parameters for further analysis and processing.

[0065] Step S24: Perform abnormal feature fusion mapping on the matching feature matrix to generate the operation and maintenance results of the robot.

[0066] In the embodiments of the present invention, an abnormal feature fusion mapping is performed on the matching feature matrix, and a fusion algorithm (such as weighted fusion or decision tree fusion) is used to comprehensively analyze multiple feature parameters to identify abnormal features among them. The abnormal features include current fluctuations exceeding the normal range and sudden current changes. Through the fusion mapping, the abnormal features are marked and classified to generate the operation and inspection results of the robot. The operation and inspection results include all the identified abnormal features and their corresponding time and location information, which are used to guide the next operations and decisions of the operation and inspection robot. The mapping process needs to combine the statistical analysis of historical data and machine learning algorithms to ensure the accuracy and reliability of the identification results.

[0067] Preferably, step S24 includes the following steps:

[0068] Step S241: Perform complex event integration on the matching feature matrix to obtain a complex feature vector;

[0069] Step S242: Perform event network mapping on the complex feature vector to generate a complex feature event network;

[0070] Step S243: Analyze the evolution law of the complex feature event network to obtain a dynamic complex event graph;

[0071] Step S244: Infer the power grid operation state from the dynamic complex event graph to generate the operation and inspection results of the robot.

[0072] The advantages of the present invention are as follows: comprehensively integrating and analyzing the multi-dimensional feature parameters in the matching feature matrix to generate a more complex feature vector, which can more comprehensively describe the abnormal state of the power grid. Converting the complex feature vector into the form of an event network can intuitively display the mutual relationships and influences among various abnormal events, providing a basis for subsequent dynamic evolution analysis. Based on the constructed event network, through time series analysis methods, the dynamic evolution law of power grid abnormal events can be mined, providing an important basis for power grid state prediction and control decision-making. Comprehensively using the analysis results of the dynamic complex event graph to comprehensively infer and evaluate the current operation state of the power grid, providing reliable support for the generation of the operation and inspection results of the robot. Giving full play to the advantages of event network analysis and dynamic evolution law mining technologies, through in-depth analysis of power grid abnormal features, the real-time operation state of the power grid can be inferred more accurately, providing important support for the intelligent fault warning and optimized control decision-making of power grid operation and inspection robots.

[0073] In the embodiments of the present invention, the matching feature matrix in the historical operation and inspection database is analyzed. By using advanced data processing technologies, such as multi-dimensional feature extraction algorithms, various matching features in the matrix are highly integrated. Through clustering algorithms, relevant features are classified into complex events. Using hierarchical clustering or density clustering methods, features with high similarity are aggregated into a group. With complex event integration technology, these aggregated features are further processed to generate a multi-dimensional complex feature vector. This complex feature vector contains the integrated information of various features in the matching feature matrix. Applying advanced network mapping technology, the complex feature vector is mapped into an event network. By using methods of graph theory and network analysis, a relationship model of nodes and edges is established, where nodes represent each complex feature and edges represent the relationships and mutual influences between features. Using shortest path algorithms such as Dijkstra algorithm or Floyd-Warshall algorithm, the relevance and path length between features are analyzed to construct a complex feature event network. This event network reflects the dynamic relationships and influence mechanisms between various complex features. Using network evolution analysis technology, in-depth analysis of the event network is carried out. Through time series analysis and evolution models (such as SIS model, SIR model) methods, the changing rules of network nodes and edges are studied. Using dynamic network analysis technology, the evolution process of feature events in the time dimension is observed and recorded. Combining with the actual operation of the power grid, a dynamic complex event graph is generated, which intuitively shows the evolution trajectories and dynamic changes of each feature event, providing a basis for subsequent state inference. Using a power grid state inference algorithm, in-depth analysis of the dynamic graph is carried out. By using machine learning algorithms, such as decision trees, support vector machines (SVM), combined with historical operation and inspection data and the actual operation of the power grid, the states of each node in the dynamic graph are predicted and inferred. The operation state and potential risks of the power grid at different time nodes are analyzed to identify abnormal situations. According to the inference results, a detailed robot operation and inspection report is generated, including the operation state of the power grid, potential fault points, and recommended maintenance measures.

[0074] Preferably, step S3 includes the following steps:

[0075] Step S31: Perform corresponding power grid topology association on the current data of each section of the transmission line to generate corresponding power grid topology associated current data;

[0076] Step S32: Perform electromagnetic spectrum mapping analysis on the corresponding power grid topology associated current data to obtain power grid electromagnetic spectrum data;

[0077] Step S33: Perform spectral parameter space quantization on the power grid electromagnetic spectrum data to obtain power grid electromagnetic field parameters;

[0078] Step S34: Perform electromagnetic field coupling drive simulation on the power grid electromagnetic field parameters to generate an electromagnetic field simulation environment.

[0079] The advantages of the present invention are as follows: By associating and mapping the current data of transmission lines with the power grid topology structure, the coupling relationship between each line can be more accurately reflected, laying a foundation for subsequent electromagnetic field analysis. Using spectrum analysis technology to perform spectrum mapping on the associated current data can obtain the electromagnetic spectrum characteristics of the entire power grid, providing a basis for the quantification of electromagnetic field parameters. Converting the electromagnetic spectrum data into quantifiable electromagnetic field parameters, such as electric field strength and magnetic field strength, provides operable inputs for subsequent electromagnetic field coupling analysis. Based on the quantified electromagnetic field parameters, an electromagnetic field simulation environment of the power grid is constructed by using electromagnetic field coupling simulation technology, providing simulation support for the operation and optimal control of robots in the actual power grid. By making full use of the power grid topology structure and current data, through electromagnetic spectrum analysis, parameter quantification, and coupling simulation technology means, a digital twin model of the electromagnetic field environment is constructed, providing important simulation support for the state perception, fault diagnosis, and optimal decision-making of power grid operation and maintenance robots in the power grid.

[0080] As an example of the present invention, refer to Figure 3 shown, in this example, step S3 includes:

[0081] Step S31: Perform corresponding power grid topology association on the current data of each section of the transmission line to generate corresponding power grid topology associated current data;

[0082] In the embodiment of the present invention, the current data of each section of the transmission line is analyzed by using the power grid topology structure diagram. First, obtain the topology structure information of the power grid, including the connection relationship, node position, and line path of each transmission line section. Then, match the current data of each section of the transmission line with its corresponding topological position to form a topological association relationship. Through the topological association algorithm, calculate the specific position and association degree of each section of the transmission line in the entire power grid topology, and generate a current data set containing topological relationships. This data set not only contains the current information of each section of the transmission line but also reflects its topological position and connection relationship in the entire power grid, thus providing basic data for subsequent spectrum analysis.

[0083] Step S32: Perform electromagnetic spectrum mapping analysis on the corresponding power grid topology associated current data to obtain power grid electromagnetic spectrum data;

[0084] In the embodiment of the present invention, it is processed by using electromagnetic spectrum analysis technology. First, use the fast Fourier transform (FFT) to convert the current data in the time domain into frequency domain data and extract the spectrum characteristics in the current signal. Apply the spectrum mapping algorithm to map the frequency domain data to the electromagnetic spectrum space to generate the electromagnetic spectrum data corresponding to each section of the transmission line. Identify the frequency components in the current signal and their distribution in the power grid. Finally, obtain a data set containing the overall electromagnetic spectrum information of the power grid, providing a basis for subsequent electromagnetic field parameter quantification.

[0085] Step S33: Perform spectral parameter space quantization on the power grid electromagnetic spectrum data to obtain power grid electromagnetic field parameters;

[0086] In the embodiment of the present invention, detailed spectral parameter analysis and quantization processing are performed on the electromagnetic spectrum data. First, key spectral parameters such as frequency, amplitude, and phase are selected, and parameter extraction is performed on the electromagnetic spectrum data of each section of the transmission line. Then, spatial quantization technology is applied to map these parameters to the electromagnetic field parameter space to form an electromagnetic field parameter set. Through statistical analysis and feature extraction, the specific parameter values of each section of the transmission line in the electromagnetic field, such as electric field strength and magnetic field strength, are determined.

[0087] Step S34: Perform electromagnetic field coupling drive simulation on the power grid electromagnetic field parameters to generate an electromagnetic field simulation environment.

[0088] In the embodiment of the present invention, electromagnetic field simulation software is used to perform electromagnetic field simulation on the entire power grid. The electromagnetic field parameters are input into the simulation software, and the simulation boundary conditions and initial conditions are set. Applying electromagnetic field coupling drive technology, through numerical calculation methods such as the finite element method (FEM) or the finite difference method (FDM), global simulation of the electromagnetic field is performed. During the simulation process, the electromagnetic coupling effects of each section of the transmission line are simulated, and the distribution and changes of the electromagnetic field are calculated. Finally, a detailed electromagnetic field simulation environment is generated, including the electromagnetic field distribution data of the entire power grid, providing a basis for further analysis and optimization of the power grid operation state.

[0089] Preferably, step S4 includes the following steps:

[0090] Step S41: Infer the power grid field state distribution from the electromagnetic field simulation environment to obtain power grid field state distribution data;

[0091] Step S42: Perform electric potential field simulation on the power grid field state distribution data to generate a simulated electric potential field;

[0092] Step S43: Perform collaborative prediction of node electric potential on the simulated electric potential field to obtain predicted values of node electric potential distribution;

[0093] Step S44: Compare the predicted values of node electric potential distribution based on the robot operation and inspection results to obtain measured deviation data.

[0094] The advantages of the present invention are as follows: By analyzing the electromagnetic field simulation environment, the field state distribution of each node in the power grid can be inferred, providing basic data for subsequent potential field simulation. Based on the field state distribution data, the simulated potential field of the power grid is constructed using potential field simulation technology, providing a reliable data source for predicting the node potential distribution. By utilizing the potential coupling relationship between nodes, the node potential distribution of the entire power grid is predicted and analyzed, providing a reference basis for comparing the measured value with the predicted value. By comparing the predicted value of the node potential distribution with the measured results of the robot, the deviation between the actual operation state of the power grid and the simulation model can be accurately analyzed, providing an important basis for subsequent optimal control decisions. The electromagnetic field simulation model of the power grid is fully utilized. Through the prediction and deviation analysis of the power grid field state, potential field, and node potential distribution, real-time monitoring and feedback of the overall operation state of the power grid are provided for the power grid operation and maintenance robot, thus providing an important reference basis for optimal control decisions. These steps help to improve the intelligence level and control accuracy of the power grid operation and maintenance robot.

[0095] In the embodiments of the present invention, an electromagnetic field simulation environment is utilized to infer the field state distribution of the power grid. First, the initial conditions and boundary conditions of the electromagnetic field simulation are set, and the simulation program is run to globally simulate the electromagnetic field of the entire power grid. Through numerical calculation methods, such as the finite element method (FEM) or the finite difference method (FDM), the electromagnetic field distribution and variation of each node in the power grid are simulated. The simulation results are analyzed, key parameters are extracted, such as the electric field intensity, magnetic field intensity, and field state distribution, to generate power grid field state distribution data. Using the power grid field state distribution data, a potential field simulation is carried out. First, based on the field state distribution data, a potential field model is established, and the initial conditions and boundary conditions of the simulation are set. Then, numerical calculation methods, such as the finite difference method (FDM) or the finite element method (FEM), are used to simulate and calculate the potential distribution in the power grid. During the simulation process, the potential changes and distribution characteristics of each node are focused on to ensure the accuracy and reliability of the simulation results. A detailed simulated potential field is generated, including the potential distribution data of the entire power grid. Using the simulated potential field data, the potentials of each node are collaboratively predicted. A collaborative prediction algorithm, such as a neural network or a support vector machine (SVM), is selected to predict the potentials of each node in the simulated potential field. The collaborative prediction algorithm takes into account the mutual correlation and influence between each node and can accurately predict the potential distribution of each node. Through iterative calculation and optimization, the predicted values of the potential distribution of each node are generated. This data set reflects the expected distribution of the potentials of each node in the simulation environment. Using the results of robot operation and inspection, the predicted values of the node potential distribution are compared with the actual measurements. First, the node potential data recorded by the robot during actual operation and inspection are obtained and compared with the predicted values for analysis. Through difference calculation, the deviation between the predicted value and the measured value of the potential of each node is determined. Statistical analysis methods, such as the mean square error (MSE) or the absolute error (AE), are applied to quantify the degree of deviation and distribution characteristics. Finally, the measured deviation data are generated, which reflects the difference between the simulation prediction and the actual measurement, providing a basis for further optimization and adjustment of the power grid operation state.

[0096] Preferably, step S43 includes the following steps:

[0097] Step S431: Cut the simulated potential field into independent potential field nodes to obtain an independent node potential field;

[0098] Step S432: Predict the connection of collaborative relationships for the independent node potential field to generate a collaborative relationship matrix;

[0099] Step S433: Simulate the potential propagation for the collaborative relationship matrix to generate simulated potential propagation data;

[0100] Step S434: Based on the simulated potential propagation data, perform collaborative prediction of the potentials of each node for the independent node potential field to obtain the predicted values of the node potential distribution.

[0101] The advantages of the present invention are that the simulated electric potential field is divided into multiple independent node electric potential fields, which is beneficial to subsequent more refined collaborative relationship analysis for individual nodes. By analyzing the independent node electric potential fields, the electric potential collaborative relationships between nodes are predicted and organized into a collaborative relationship matrix, providing input for the simulation of electric potential propagation. Based on the collaborative relationship matrix, the electric potential propagation simulation technology is used to predict the electric potential propagation process in the entire power grid, providing data support for the final node electric potential prediction. The results of the electric potential propagation simulation are fed back into the independent node electric potential fields, and through the electric potential coupling relationship between nodes, the electric potential distribution of each node is collaboratively predicted to obtain a more accurate predicted value of the node electric potential distribution. The characteristics of the simulated electric potential field are fully utilized, and through means such as node cutting, collaborative relationship analysis, and electric potential propagation simulation technology, the refined prediction of the node electric potential distribution in the power grid is achieved. This not only helps to improve the accuracy of the prediction but also provides a more reliable data reference for subsequent optimal control decisions. The entire process reflects the intelligent capabilities of the power grid operation and maintenance robot in power grid state perception and analysis.

[0102] In the embodiments of the present invention, independent cutting operations are performed on each node in the simulated potential field to separately analyze the potential field of each node. The specific operations include dividing the simulated potential field into multiple sub-regions, and each sub-region only contains one node and the potential data around it. In this way, an independent node potential field can be generated to ensure that the potential information of each node is independent and not affected by other nodes. This process can be achieved through image processing techniques or spatial segmentation algorithms, such as region growing algorithms or K-means clustering algorithms, so as to ensure the independence and accuracy of the node potential field. Predict the co-relationship connection of the independent node potential field. First, establish an initial relationship network containing all nodes, and then apply a co-prediction algorithm, such as a prediction model based on a graph neural network (GNN), to predict the relationship between the potential field of each node and other nodes. During the prediction process, consider the physical connection and electromagnetic coupling effect between nodes to generate a co-relationship matrix. This matrix reflects the strength of the co-relationship between each node and provides basic data for subsequent potential propagation simulation. After obtaining the co-relationship matrix, perform potential propagation simulation. The specific operations include constructing a potential propagation model based on the co-relationship matrix and setting the initial potential distribution conditions. Then, use numerical simulation methods, such as the finite difference method (FDM) or the random walk algorithm, to simulate the propagation process of the potential in the network. Through the simulation, detailed potential propagation data can be generated, reflecting the potential changes of each node at different time steps. Use the simulated potential propagation data to perform co-prediction on the independent node potential field. The specific operations include applying a co-prediction algorithm, such as a prediction model based on a Bayesian network or a long short-term memory network (LSTM), to predict the potential of each node. During the prediction process, combine the simulated potential propagation data and the co-relationship matrix, and comprehensively consider the mutual influence and co-effect between each node. Through iterative optimization, generate the predicted value of the node potential distribution, reflecting the potential distribution of each node at future moments. This predicted value will be used for the monitoring and control of the actual power grid operation status, providing a basis for improving the stability and reliability of the power grid operation.

[0103] Preferably, step S5 includes the following steps:

[0104] Step S51: Classify the measured deviation data by error type to obtain an error classification data set;

[0105] Step S52: Trace back the error sources of the error classification data set to generate operation and inspection error source data;

[0106] Step S53: Perform robot operation and inspection simulation on the operation and inspection error source data to obtain robot measurement misalignment data;

[0107] Step S54: Optimize the operation and inspection results of the robot based on the robot measurement misalignment data to obtain optimized operation and inspection results, and transmit the optimized operation and inspection results back to the grid robot to perform optimized control of the grid operation and inspection robot.

[0108] The advantages of the present invention are as follows: classifying the measured deviation data according to different error types, establishing a complete error classification data set, and providing basic data for subsequent error source tracing. By analyzing the error classification data set, determining the specific error sources causing the measurement deviation, generating operation and inspection error source data, and providing a reference for robot operation and inspection simulation. Based on the operation and inspection error source data, simulating the actual measurement process of the grid operation and inspection robot to obtain robot measurement misalignment data, and providing a basis for optimized control decisions. Using the robot measurement misalignment data to optimize the original operation and inspection results, obtaining more accurate and reliable optimized operation and inspection results, and feeding them back to the grid operation and inspection robot to perform optimized control. Making full use of the comparative analysis of the measured data and the simulation model, through technical means such as error classification, error source tracing, and robot operation and inspection simulation, realizing the optimization of the measurement results of the grid operation and inspection robot, and improving the accuracy and reliability of the robot's optimized control. This helps to further improve the intelligent level of the grid operation and inspection robot in grid state perception, analysis, and control, thereby providing a more powerful guarantee for the safe and stable operation of the grid.

[0109] In the embodiments of the present invention, the measured deviation data is classified in detail according to error types. This process includes using advanced data analysis and classification algorithms, such as Support Vector Machine (SVM), Random Forest, or deep learning models (such as Convolutional Neural Network), to classify the measured deviation data. By extracting features and preprocessing the data set, the accuracy and integrity of the data are ensured. The goal of classification is to clearly distinguish different types of deviations (such as measurement errors, environmental interferences, errors caused by equipment failures), so that subsequent error source analysis and robot operation and maintenance simulation can accurately identify and process different types of errors. Using the error classification data set, backtracking analysis of error sources is carried out. The main purpose of this step is to determine the specific sources and causes of each error type by deeply analyzing the error classification data set. Advanced data mining techniques and statistical analysis methods, such as causal relationship analysis, Bayesian network, or regression analysis, are used to identify and analyze the root causes of errors. For example, for measurement errors, factors such as sensor accuracy, environmental changes, and noise during data acquisition are involved. Through backtracking analysis, detailed operation and maintenance error source data are generated, providing accurate input data for subsequent robot operation and maintenance simulation. Using the operation and maintenance error source data, robot operation and maintenance simulation is carried out. This includes simulating various error situations encountered by the robot during actual operation and evaluating the impact of these errors on measurement results. Using a simulation platform or model, such as Discrete Event Simulation (DES) or Agent-Based Model (ABM), to simulate the operation and maintenance behavior of the robot in the power grid environment. Through simulation, robot measurement misalignment data can be obtained, that is, the difference between the measurement results of the robot in the real scenario and the ideal situation. Using the robot measurement misalignment data, the operation and maintenance results of the robot are optimized. This process includes analyzing and correcting the errors caused by measurement misalignment to improve the measurement accuracy and precision of the robot. By applying data correction algorithms or real-time adjustment strategies, the optimized parameters obtained in the simulation are applied to the actual power grid operation and maintenance tasks. The optimized operation and maintenance results will be fed back to the power grid robot through the control system to achieve real-time optimized control and operation and maintenance management.

[0110] Preferably, step S52 includes the following steps:

[0111] Step S521: Stitch error trees for the error classification data set to obtain an operation and maintenance error tree;

[0112] Step S522: Statistically analyze the proportion of each category in the error classification data set to generate error category proportion data;

[0113] Step S523: Quantify the error contribution degree of the operation and maintenance error tree based on the error category proportion data to obtain error contribution degree data;

[0114] Step S524: Backtrack the error sources of the error classification dataset according to the error contribution degree data to generate operation and inspection error source data.

[0115] The advantages of the present invention are as follows: by associating and splicing various types of errors in the error classification dataset, a complete operation and inspection error tree is formed, clearly showing the hierarchical relationship between various error types, which is beneficial to the subsequent quantification of error contribution degree. Statistical analysis of the error classification dataset is carried out to obtain the proportion of different error categories in the overall error, providing basic data for the quantification of error contribution degree. Combining the error tree and the error category proportion data, a quantitative analysis of the contribution degree of various errors in the operation and inspection process is carried out to obtain detailed error contribution degree data, providing a basis for accurate error source backtracking. Based on the error contribution degree data, in-depth analysis is carried out on each monitoring index in the error classification dataset, accurately tracing the key error sources that cause measurement deviation, and generating operation and inspection error source data. Through in-depth exploration and analysis of the error generation mechanism in the power grid operation and inspection process, by means of multi-dimensional technical means such as error tree construction, error proportion statistics, and error contribution degree quantification, the accurate positioning of key error sources is realized. This not only helps to improve the accuracy of robot measurement results, but also provides a more reliable basis for subsequent optimization control decisions. The whole process fully demonstrates the intelligent level of the power grid operation and inspection robot in fault diagnosis and optimization analysis.

[0116] In the embodiments of the present invention, the error classification dataset is used for splicing and constructing an error tree. An error tree is a hierarchical structure used to display and understand different error types and their relationships. This process involves using data mining and graph theory techniques, such as decision tree algorithms or graph algorithms, to analyze and integrate the error classification data. By connecting various errors according to their sources and impact degrees, a complete operation and maintenance error tree is formed. For example, using the decision tree algorithm for step-by-step segmentation and classification based on features, different error types and subtypes are hierarchically organized for subsequent quantification of error contribution degrees and backtracking analysis of error sources. Statistical analysis of the proportion of each category is performed on the already classified error dataset. By calculating the proportion and distribution of each error category in the overall error dataset, error category proportion data is obtained. Statistical methods and data analysis tools, such as data visualization tools or statistical software, are used to count the quantity and calculate the proportion of each error type. These data will help evaluate the importance and impact degree of various error types in the overall error, providing a basis for the next step of quantifying error contribution degrees. The quantification analysis of error contribution degrees is carried out using the operation and maintenance error tree and error category proportion data. This process involves evaluating the contribution degree of each error type to the final error result according to its proportion in the overall. Through mathematical modeling and statistical analysis, the influence weight or contribution ratio of each error type is calculated and the results are mapped to each node of the operation and maintenance error tree. These error contribution degree data will reveal which error types have a greater impact on the final operation and maintenance results. According to the error contribution degree data, backtracking analysis of the error sources is performed on the error classification dataset. The goal of this step is to determine the specific reasons and sources for each error type. Using statistical inference, causal relationship analysis or model verification methods, the root causes of various error types are deeply analyzed and identified. Through backtracking analysis, detailed operation and maintenance error source data is generated, providing actionable suggestions and strategies for subsequent optimization and control of operation and maintenance results. The specific operations include using data loading tools (such as SQL queries, data import scripts) to import the error contribution degree data generated in step S523 into the analysis platform. Through a standardized data interface, the error contribution degree data is associated with the error classification dataset to ensure data consistency and accuracy. Causal inference algorithms (such as Bayesian networks, structural equation models) are used to model the causal relationships of the error classification dataset. Causal inference algorithms can identify the causal relationships between data and reveal the potential sources of various errors. First, a causal relationship model is established to preliminarily identify the potential causes of each error type. Then, through parameter estimation and hypothesis testing, the model is optimized and verified to ensure the accuracy of the causal relationship. Based on the causal relationship model, statistical analysis methods (such as regression analysis, path analysis) are used to identify the specific sources of each error type. By analyzing the variables in the error classification dataset, it is determined which factors have a significant impact on a specific error type.For each type of error, generate a detailed error source report, recording its specific causes and degrees of impact. Conduct error source verification experiments. Select representative error cases and design and execute experimental verification schemes. Through comparative analysis of experimental data, verify whether the identified error sources conform to the actual situation. If deviations are found, adjust the causal relationship model and re-identify the error sources until the verification results are consistent with the model predictions. Integrate the results of error source identification and verification to generate operation and maintenance error source data. The operation and maintenance error source data should include detailed source descriptions, influencing factors and their contribution degrees, and verification experiment result information for each type of error. Save the error source data in a standard data storage format (such as JSON, XML) for subsequent analysis and application.

[0117] In this specification, an optimization control system for power grid operation and maintenance is also provided, which is used to execute an optimization control method for power grid operation and maintenance as described above. The optimization control system for power grid operation and maintenance includes:

[0118] A current fluctuation identification module, which is used to obtain the current data of each section of the transmission line; extract the current fluctuation characteristics from the current data of each section of the transmission line to obtain the power grid current fluctuation characteristic data; analyze the noise oscillation of the current data of each section of the transmission line to obtain the transient fluctuation noise data; identify the fluctuation mode for the power grid current fluctuation characteristic data and the transient fluctuation noise data to generate a current fluctuation mode;

[0119] An operation and maintenance analysis module, which is used to obtain the historical operation and maintenance database; perform current anomaly feature fusion analysis on the current fluctuation mode based on the historical operation and maintenance database to generate the robot operation and maintenance results;

[0120] An electromagnetic field simulation module, which is used to perform electromagnetic spectrum mapping on the current data of each section of the transmission line to obtain the power grid electromagnetic spectrum data; construct an electromagnetic coupling field for the power grid electromagnetic spectrum data to generate an electromagnetic field simulation environment;

[0121] A deviation calculation module, which is used to simulate the power grid potential field for the electromagnetic field simulation environment to generate a simulated potential field; perform measured deviation analysis on the simulated potential field based on the robot operation and maintenance results to obtain measured deviation data;

[0122] A robot optimization module, which is used to trace back the error sources for the predicted value deviation data to generate operation and maintenance error source data; perform operation and maintenance optimization on the robot operation and maintenance results based on the operation and maintenance error source data to execute the optimization control of the power grid operation and maintenance robot.

[0123] The advantages of the present invention lie in covering the entire process from data acquisition, feature extraction, fault diagnosis to optimization control, which can better realize the intellectualization and automation of power grid operation and maintenance. By making full use of various data sources such as transmission line current data, historical operation and maintenance data, and electromagnetic spectrum data, a three-dimensional and multi-angle analysis of the power grid state is realized. Through the analysis of current fluctuation characteristics and the fusion of current anomaly characteristics, potential fault hazards in the power grid can be discovered in advance, providing a basis for preventive maintenance. Based on electromagnetic field simulation and electric potential field simulation, a physical model of the power grid is established, providing strong support for subsequent deviation analysis and optimization. Through the deviation analysis of operation and maintenance results and the traceability of error sources, the optimization and correction of the robot diagnosis results are realized, improving the diagnosis accuracy. By fusing multi-source data features, the current fluctuation pattern is identified, providing a reference for the intelligent judgment and decision-making of the power grid state. By making full use of various data resources in power grid operation and maintenance and through multi-step in-depth analysis and optimization, the intelligent optimization control of the power grid operation and maintenance robot is finally realized, which is of great significance for improving the reliability and safety of the power grid.

[0124] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the application document are intended to be encompassed within the present invention.

[0125] The above description is only the specific implementation manners of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An optimization control method for power grid operation and inspection, characterized in that: The following steps are involved: Step S1: obtaining current data of each section of the transmission line; Extract the current fluctuation characteristics of the current data of each section of the transmission line to obtain the current fluctuation characteristic data of the power grid; Perform noise oscillation analysis on the current data of each section of the transmission line to obtain transient fluctuation noise data; Perform fluctuation pattern recognition on the grid current fluctuation characteristic data and transient fluctuation noise data to generate a current fluctuation pattern; Step S2: Acquire a historical operation and inspection database; perform current abnormality feature fusion analysis on the current fluctuation pattern based on the historical operation and inspection database to generate a robot operation and inspection result; Step S3: Perform electromagnetic spectrum mapping on the current data of each section of the transmission line to obtain electromagnetic spectrum data of the power grid; construct an electromagnetic coupling field on the electromagnetic spectrum data of the power grid to generate an electromagnetic field simulation environment; Step S4: simulating the electric potential field of the power grid in the electromagnetic field simulation environment to generate a simulated electric potential field; performing measured deviation analysis on the simulated electric potential field based on the robot operation and inspection results to obtain measured deviation data; Step S5: perform error source backtracking on the measured deviation data to generate operation and inspection error source data; perform operation and inspection optimization on the robot operation and inspection results based on the operation and inspection error source data to perform optimized control of the power grid operation and inspection robot.

2. The optimization control method for power grid operation and inspection according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: using a power grid operation and inspection robot to monitor the current along the power grid transmission line to obtain current data of each section of the transmission line; Step S12: performing fast Fourier transform on the current data of each section of the transmission line to generate grid current frequency domain data; Step S13: extracting current fluctuation characteristics from the grid current frequency domain data to obtain grid current fluctuation characteristic data; Step S14: extracting high-frequency noise from the current data of each section of the transmission line to generate current high-frequency noise data; Step S15: performing noise oscillation analysis on the current high-frequency noise data to obtain transient fluctuation noise data; Step S16: performing fluctuation pattern recognition on the grid current fluctuation characteristic data and transient fluctuation noise data to generate a current fluctuation pattern.

3. The optimization control method for power grid operation and inspection according to claim 2, characterized in that: Step S2 includes the following steps: Step S21: Acquire historical operation and inspection data; construct a timestamp index for the historical operation and inspection data to generate a historical operation and inspection database; Step S22: performing a matching search on a historical operation and inspection database based on the current fluctuation pattern to obtain historical matching current fluctuation data; Step S23: extract matching feature parameters from the historical operation and inspection database according to the historical matching current fluctuation data to obtain a matching feature matrix; Step S24: Perform abnormal feature fusion mapping on the matching feature matrix to generate the robot operation inspection result.

4. The optimization control method for power grid operation and inspection according to claim 3, characterized in that: Step S24 includes the following steps: Step S241: performing complex event integration on the matching feature matrix to obtain a complex feature vector; Step S242: performing event network mapping on the complex feature vector to generate a complex feature event network; Step S243: Analyze the evolution law of the complex feature event network to obtain a dynamic complex event graph; Step S244: Inferring the power grid operation status of the dynamic complex event graph and generating robot operation and inspection results.

5. The optimization control method for power grid operation and inspection according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: Correlate the current data of each section of the transmission line with the corresponding power grid topology to generate current data associated with the corresponding power grid topology; Step S32: performing electromagnetic spectrum mapping analysis on the current data associated with the corresponding power grid topology to obtain electromagnetic spectrum data of the power grid; Step S33: performing spectrum parameter space quantization on the power grid electromagnetic spectrum data to obtain the power grid electromagnetic field parameters; Step S34: performing electromagnetic field coupling drive simulation on the electromagnetic field parameters of the power grid to generate an electromagnetic field simulation environment.

6. The optimization control method for power grid operation and inspection according to claim 4, characterized in that: Step S4 includes the following steps: Step S41: inferring the power grid state distribution in the electromagnetic field simulation environment to obtain power grid state distribution data; Step S42: performing electric potential field simulation on the power grid state distribution data to generate a simulated electric potential field; Step S43: performing node potential collaborative prediction on the simulated potential field to obtain node potential distribution prediction values; Step S44: Based on the robot operation and inspection results, the predicted value of the node potential distribution is compared with the potential distribution to obtain the measured deviation data.

7. The optimization control method for power grid operation and inspection according to claim 6, characterized in that: Step S43 includes the following steps: Step S431: performing independent potential field node cutting on the simulated potential field to obtain independent node potential fields; Step S432: performing collaborative relationship connection prediction on the electric potential field of the independent nodes to generate a collaborative relationship matrix; Step S433: performing electric potential propagation simulation on the synergy relationship matrix to generate simulated electric potential propagation data; Step S434: Based on the simulated potential propagation data, the potential of each node is collaboratively predicted for the independent node potential field to obtain the predicted value of the node potential distribution.

8. The optimization control method for power grid operation and inspection according to claim 1, characterized in that: Step S5 includes the following steps: Step S51: classify the measured deviation data by error type to obtain an error classification data set; Step S52: performing error source backtracking on the error classification data set to generate operation and inspection error source data; Step S53: performing robot operation and inspection simulation on the operation and inspection error source data to obtain robot measurement inaccuracy data; Step S54: Optimize the robot operation and inspection results based on the robot measurement inaccuracy data to obtain optimized operation and inspection results, and transmit the optimized operation and inspection results back to the power grid robot to perform optimized control of the power grid operation and inspection robot.

9. The optimization control method for power grid operation and inspection according to claim 8, characterized in that: Step S52 includes the following steps: Step S521: performing error tree splicing on the error classification data set to obtain an operation and inspection error tree; Step S522: performing category proportion statistics on the error classification data set to generate error category proportion data; Step S523: quantifying the error contribution of the operation and inspection error tree based on the error category proportion data to obtain error contribution degree data; Step S524: perform error source backtracking on the error classification data set according to the error contribution data to generate operation and inspection error source data.

10. An optimization control system for power grid operation and inspection, characterized in that: The optimization control system for power grid operation inspection is used to execute the optimization control method for power grid operation inspection according to claim 1, and the optimization control system for power grid operation inspection comprises: The current fluctuation identification module is used to obtain the current data of each section of the transmission line; extract the current fluctuation characteristics of the current data of each section of the transmission line to obtain the current fluctuation characteristic data of the power grid; perform noise oscillation analysis on the current data of each section of the transmission line to obtain transient fluctuation noise data; perform fluctuation pattern recognition on the current fluctuation characteristic data of the power grid and the transient fluctuation noise data to generate the current fluctuation pattern; The operation and inspection analysis module is used to obtain the historical operation and inspection database; based on the historical operation and inspection database, the current fluctuation pattern is analyzed by current abnormality feature fusion to generate the robot operation and inspection results; The electromagnetic field simulation module is used to map the electromagnetic spectrum of the current data of each section of the transmission line to obtain the electromagnetic spectrum data of the power grid; construct the electromagnetic coupling field of the electromagnetic spectrum data of the power grid to generate an electromagnetic field simulation environment; The deviation calculation module is used to simulate the electric potential field of the power grid in the electromagnetic field simulation environment and generate a simulated electric potential field; based on the robot operation and inspection results, the simulated electric potential field is analyzed for measured deviations to obtain measured deviation data; The robot optimization module is used to trace the error source of the measured deviation data and generate operation and inspection error source data; based on the operation and inspection error source data, the robot operation and inspection results are optimized to perform optimized control of the power grid operation and inspection robot.

Images