Intelligent detection system of high and low voltage power distribution cabinet and detection method thereof

Abstract

The invention relates to the technical field of power distribution cabinet detection, and discloses an intelligent detection system and method for a high-low voltage power distribution cabinet, and the system comprises a data collection module, a feature extraction module, a network construction and updating module, a fault positioning module, a positioning map generation module, and a self-healing control module. The method comprises the following steps: generating a multi-physical field coupling data set by synchronously collecting electromagnetic radiation spectrum and cabinet vibration waveform, performing time-frequency domain conjoint analysis to extract fault feature vectors, constructing and dynamically updating a fault propagation network, identifying key fault sources by combining betweenness centrality and historical case matching, and generating a fault positioning map; accurate self-healing control is realized by applying reverse electromagnetic pulse and adjusting contact pressure based on the atlas; according to the invention, the problems of low positioning precision and mechanical self-healing control means during detection of the high-low voltage power distribution cabinet are solved, and the fault detection efficiency and reliability are significantly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution cabinet detection, and particularly to an intelligent detection system and a detection method for high and low voltage power distribution cabinets. Background Art

[0002] Currently, the fault detection of high and low voltage power distribution cabinets mostly relies on the static analysis of a single physical field (such as current, temperature), and it is unable to effectively capture the coupling characteristics of electromagnetic and mechanical vibrations, resulting in incomplete extraction of fault characteristics; traditional methods use a fixed topology model to analyze fault propagation, lacking a dynamic update mechanism and being difficult to adapt to complex fault scenarios; in addition, self-healing control strategies are mostly global power-off or mechanical locking, with a high risk of misoperation and being unable to accurately locate the fault source and potential diffusion areas; In the prior art, although CN117553859A realizes data acquisition through multiple sensors, it does not solve the problems of multi-physical field data fusion and dynamic network modeling; CN105762775A proposes dynamic topology optimization, but does not combine the energy characteristics of fault propagation with historical case matching, resulting in a large positioning deviation. At the same time, the power grid self-healing system lacks a physical space mapping mechanism for power distribution cabinets. Summary of the Invention

[0003] The present invention provides an intelligent detection system and a detection method for high and low voltage power distribution cabinets, and its main purpose is to solve the problems of low positioning accuracy, insufficient ability to track fault propagation paths, and mechanical self-healing control means in the fault detection of high and low voltage power distribution cabinets.

[0004] To achieve the above object, an intelligent detection system for high and low voltage power distribution cabinets provided by the present invention is characterized in that the system includes: A data acquisition module, configured to synchronously collect the electromagnetic radiation spectrum and the cabinet surface vibration waveform of the high and low voltage power distribution cabinet, and generate a multi-physical field coupling data set; A feature extraction module, configured to perform a joint time-frequency domain analysis on the multi-physical field coupling data set to obtain a fault feature vector of the high and low voltage power distribution cabinet; A network construction and update module, configured to construct and dynamically update a fault propagation network based on the fault feature vector; A fault location module, configured to match the betweenness centrality and energy distribution of nodes in the fault propagation network with the fault modes in the historical fault case library, identify the key fault sources of the high and low voltage power distribution cabinet, and mark potential fault diffusion areas; A location map generation module, configured to map the key fault sources and the potential fault diffusion areas to the physical space coordinates of the high and low voltage power distribution cabinet to obtain a fault location map of the high and low voltage power distribution cabinet; A self-healing control module, configured to output a self-healing control instruction for the high- and low-voltage power distribution cabinet based on the fault location map, where the self-healing control instruction includes applying a reverse electromagnetic pulse to the key fault source and adjusting the contact pressure in the potential fault propagation area.

[0005] To solve the above problems, the present invention also provides an intelligent detection method for a high- and low-voltage power distribution cabinet, the method comprising: S1. Synchronously collect the electromagnetic radiation spectrum and the cabinet surface vibration waveform of the high- and low-voltage power distribution cabinet to generate a multi-physical-field coupling data set; S2. Perform a joint time-frequency domain analysis on the multi-physical-field coupling data set to obtain the fault feature vector of the high- and low-voltage power distribution cabinet; S3. Construct and dynamically update a fault propagation network based on the fault feature vector; S4. Based on the betweenness centrality and energy distribution of the nodes in the fault propagation network, match with the fault patterns in the historical fault case library, identify the key fault sources and mark the potential fault propagation areas; S5. Map the key fault sources and the potential fault propagation areas to the physical space coordinates of the high- and low-voltage power distribution cabinet to obtain the fault location map of the high- and low-voltage power distribution cabinet; S6. Output a self-healing control instruction for the high- and low-voltage power distribution cabinet based on the fault location map, where the self-healing control instruction includes applying a reverse electromagnetic pulse to the key fault source and adjusting the contact pressure in the potential fault propagation area.

[0006] By synchronously collecting and aligning electromagnetic and vibration data to form a multi-physical-field coupling data set and combining joint time-frequency domain analysis to extract multi-dimensional fault features, the present invention solves the problem of incomplete extraction of single-physical-field features and significantly improves the accuracy of fault detection; maps fault features to network nodes and edges, dynamically updates the topological structure based on the energy transfer probability model, and reflects the change of the fault propagation path in real time, overcoming the defect that traditional static models cannot track dynamic propagation and enhancing the analysis ability of complex fault scenarios; screens key fault sources through betweenness centrality, combines historical case matching and energy distribution analysis to quickly lock the root cause of the fault, and applies a reverse electromagnetic pulse based on the fault location map to suppress electromagnetic interference and dynamically adjust the contact pressure to block the spread of mechanical faults, achieving targeted intervention; maps network nodes to the physical coordinates of the power distribution cabinet to generate a fault location map, providing intuitive fault distribution information for maintenance personnel, shortening the fault handling response time, and improving on-site operation efficiency. Description of the Drawings

[0007] Figure 1 It is a system architecture diagram of an intelligent detection system for a high- and low-voltage power distribution cabinet provided by an embodiment of the present invention.

[0008] Figure 2 It is a schematic flowchart of an intelligent detection method for a high and low voltage power distribution cabinet provided by an embodiment of the present invention.

[0009] The realization of the object of the present invention, functional features and advantages will be further described in conjunction with the embodiments with reference to the accompanying drawings. Specific embodiments

[0010] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0011] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "the" and "said" used in the embodiments of the present invention and the appended claims are also intended to include the plural forms unless the context clearly dictates otherwise. "Plural" generally includes at least two.

[0012] Depending on the context, the words "if", "when" as used herein may be interpreted as "when" or "while" or "in response to determining" or "in response to detecting". Similarly, depending on the context, the phrase "if determined" or "if detecting (stated condition or event)" may be interpreted as "when determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)".

[0013] In addition, the step timings in the following method embodiments are only examples, not strictly limited.

[0014] In fact, the server devices deployed in the intelligent detection system of the high and low voltage power distribution cabinet may consist of one or more devices. The above-mentioned intelligent detection system of the high and low voltage power distribution cabinet can be implemented as: business instances, virtual machines, and hardware devices. For example, the intelligent detection system of the high and low voltage power distribution cabinet can be implemented as a business instance deployed on one or more devices in a cloud node. Simply put, the intelligent detection system of the high and low voltage power distribution cabinet can be understood as a software deployed on a cloud node, which is used to provide the intelligent detection system of the high and low voltage power distribution cabinet for each client. Or, the intelligent detection system of the high and low voltage power distribution cabinet can also be implemented as a virtual machine deployed on one or more devices in a cloud node. An application software for managing each client is installed in the virtual machine. Or, the intelligent detection system of the high and low voltage power distribution cabinet can also be implemented as a server composed of many identical or different types of hardware devices, and one or more hardware devices are set to provide the intelligent detection system of the high and low voltage power distribution cabinet for each client.

[0015] In terms of implementation form, the intelligent detection system of the high and low voltage power distribution cabinet and the client adapt to each other. That is, if the intelligent detection system of the high and low voltage power distribution cabinet is an application installed on a cloud service platform, then the client is a client that establishes a communication connection with this application; or if the intelligent detection system of the high and low voltage power distribution cabinet is implemented as a website, then the client is implemented as a web page; or if the intelligent detection system of the high and low voltage power distribution cabinet is implemented as a cloud service platform, then the client is implemented as a small program in an instant messaging application.

[0016] As Figure 1 shown, it is the system architecture diagram of the intelligent detection system of the high and low voltage power distribution cabinet provided by an embodiment of the present invention.

[0017] The intelligent detection system 100 of the high and low voltage power distribution cabinet described in the present invention can be set in a cloud server. In terms of implementation form, it can be used as one or more service devices, or can be installed as an application on the cloud (such as the server of a mobile service operator, a server cluster, etc.), or can also be developed as a website. According to the functions to be realized, the intelligent detection system 100 of the high and low voltage power distribution cabinet can include a data acquisition module 101, a feature extraction module 102, a network construction and update module 103, a fault location module 104, a location map generation module 105, and a self-healing control module 106. The modules described in the present invention can also be called units, which refer to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, and are stored in the memory of the electronic device.

[0018] In the embodiment of the present invention, in the intelligent detection system of the high and low voltage power distribution cabinet, each of the above modules can be independently implemented and called with other modules. Here, the call can be understood as that a certain module can be connected to multiple modules of another type and provide corresponding services for the multiple modules it is connected to. For example, the sharing and evaluation module can call the same information collection module to obtain the information collected by the information collection module. Based on the above characteristics, in the intelligent detection system of the high and low voltage power distribution cabinet provided by the embodiment of the present invention, without modifying the program code, the applicable range of the intelligent detection system architecture of the high and low voltage power distribution cabinet can be adjusted by adding modules and directly calling, so as to achieve cluster-level horizontal expansion, so as to achieve the purpose of quickly and flexibly expanding the intelligent detection system of the high and low voltage power distribution cabinet. In practical applications, the above modules can be set in the same device or different devices, or can be set in virtual devices, such as service instances in a cloud server.

[0019] The following will be described in detail for each component and the specific working process of the intelligent detection system of the high and low voltage power distribution cabinet in combination with specific embodiments: The data acquisition module 101 is used to synchronously acquire the electromagnetic radiation spectrum and the cabinet surface vibration waveform of the high and low voltage power distribution cabinet, and generate a multi-physical field coupling data set.

[0020] The data acquisition module of the present invention realizes the synchronous acquisition and fusion of multi-physical field data by co-deploying electromagnetic sensors and vibration sensors and establishing an accurate time synchronization mechanism.

[0021] When the data acquisition module executes the synchronous acquisition of the electromagnetic radiation spectrum and the cabinet surface vibration waveform of the high and low voltage power distribution cabinet to generate a multi-physical field coupling data set, it is specifically used for: Arrange electromagnetic sensors in the electromagnetic radiation hot spots of the high and low voltage power distribution cabinet, install vibration sensors at the vibration sensitive points on the surface of the cabinet of the high and low voltage power distribution cabinet, and align the time of the electromagnetic sensors and the vibration sensors; Use the electromagnetic sensors to acquire the electromagnetic radiation spectrum of the high and low voltage power distribution cabinet; Use the vibration sensors to acquire the cabinet surface vibration waveform of the high and low voltage power distribution cabinet; Align the electromagnetic radiation spectrum and the cabinet surface vibration waveform according to a time window to generate the multi-physical field coupling data set of the high and low voltage power distribution cabinet.

[0022] Specifically, an electromagnetic sensor is a device used to sense the electromagnetic radiation signal generated during the operation of the high and low voltage power distribution cabinet. In this technical solution, a high-frequency current probe is selected as the electromagnetic sensor, which can effectively detect the high-frequency electromagnetic radiation changes caused by faults such as partial discharge and arc.

[0023] Specifically, the vibration sensor is mainly used to measure the mechanical vibration waveforms on the surfaces of high- and low-voltage power distribution cabinets. In this solution, a MEMS accelerometer is used as the vibration sensor, which has the characteristics of high sensitivity and small size, and can accurately capture the vibration signals generated by reasons such as loose mechanical components and resonance in the cabinet body. Specifically, time alignment refers to uniformly calibrating the time bases of the data collected by the electromagnetic sensor and the vibration sensor to ensure the consistency of the data collected by the two types of sensors in the time dimension, so as to perform multi-physical-field coupling analysis subsequently. In this solution, the IEEE 1588 Precision Time Protocol (PTP) is used to achieve time alignment, and the synchronization accuracy of this protocol can be controlled within a very small range. Specifically, the multi-physical-field coupling data set refers to the data set formed by integrating the electromagnetic radiation spectrum data of the high- and low-voltage power distribution cabinet and the vibration waveform data on the cabinet surface according to the time dimension. This data set contains information on two physical fields, namely electromagnetic and vibration, and can comprehensively reflect the physical state changes during the operation of the power distribution cabinet.

[0024] Specifically, according to the physical characteristics inside the high- and low-voltage power distribution cabinet, the electromagnetic sensor is arranged in the electromagnetic radiation hot spot area of the power distribution cabinet. For example, the breaker contact is a high-incidence area for faults such as partial discharge and is also an area where electromagnetic radiation is relatively concentrated. Installing a high-frequency current probe here can preferentially capture the abnormal electromagnetic signals generated by faults; at the same time, the vibration sensor is installed at the vibration-sensitive points on the cabinet surface, such as the disconnector operating mechanism. This part will have obvious vibration changes when mechanical components move or fail, and the MEMS accelerometer can timely sense and record these vibration information.

[0025] Specifically, the IEEE 1588 Precision Time Protocol (PTP) is used to perform time synchronization settings for the already arranged electromagnetic sensor and vibration sensor. Through this protocol, accurate time information interaction and calibration can be carried out between the sensors, so that the time error of the data collected by the two is controlled within a very small range, ensuring a high degree of consistency in time for the subsequent collected electromagnetic and vibration data.

[0026] Specifically, the high-frequency current probe installed in the electromagnetic radiation hot spot area starts to work, continuously monitoring the electromagnetic radiation signals generated during the operation of the high- and low-voltage power distribution cabinet. The high-frequency current probe can sense and convert electromagnetic signals in a specific frequency band into electrical signals that can be recognized and processed by the system. These electrical signals are formed into spectrum data reflecting the electromagnetic radiation state of the power distribution cabinet after conditioning and digital processing.

[0027] Specifically, the MEMS accelerometer located at the vibration-sensitive points on the cabinet surface senses the vibration state of the cabinet in real time and converts mechanical vibration into an electrical signal. Similar to the acquisition of electromagnetic signals, the vibration electrical signal is processed by signal conditioning circuits such as amplification and filtering, and then converted into a digital signal through an analog-to-digital conversion module, finally forming waveform data representing the vibration characteristics of the cabinet surface, including information such as the amplitude, frequency, and phase of the vibration.

[0028] Specifically, after the acquisition of the electromagnetic radiation spectrum and the cabinet surface vibration waveform is completed, the system performs time alignment processing on the two types of data according to a pre-set time window. The setting of the time window needs to comprehensively consider the data processing efficiency and the ability to capture fault characteristics. For example, it can be set to 100 milliseconds. The system integrates the electromagnetic radiation spectrum data and the vibration waveform data within the same time window to form a complete data recording unit. As time goes by, multiple such data recording units are stacked in sequence, and finally a multi-physical field coupling data set that covers different moments and fully reflects the changes in the electromagnetic and vibration physical fields during the operation of the power distribution cabinet is generated.

[0029] Generally speaking, only by reasonably arranging the electromagnetic sensor and the vibration sensor at key positions and achieving precise time synchronization can it be ensured that the acquired electromagnetic radiation spectrum and cabinet surface vibration waveform data have practical analysis value. The acquisition of the electromagnetic radiation spectrum and the cabinet surface vibration waveform are parallel data acquisition processes, which are synchronized based on the already arranged sensors and obtain raw data from the two physical field dimensions of electromagnetism and vibration respectively. Finally, the multi-physical field coupling data set generation step integrates the data obtained in the previous two acquisition steps, aligns them according to the time window, and forms a complete data set containing multi-physical field information, providing unified data input for the subsequent fault analysis process. Each step is closely connected and jointly constitutes a complete data acquisition system.

[0030] The feature extraction module 102 is used to perform joint time-frequency domain analysis on the multi-physical field coupling data set to obtain the fault feature vector of the high- and low-voltage power distribution cabinet.

[0031] During the operation of the high- and low-voltage power distribution cabinet, its internal faults often exhibit complex characteristics of multi-physical field coupling of electromagnetism and mechanical vibration. Traditional fault detection methods based on a single physical quantity are difficult to effectively identify such coupling faults, resulting in a relatively high fault misjudgment rate. This feature extraction module deeply mines the fault characteristics in the electromagnetic and vibration signals through joint time-frequency domain analysis of the multi-physical field coupling data set, and then obtains a fault feature vector that can accurately represent the fault mode, providing key data support for subsequent fault diagnosis and location.

[0032] When the feature extraction module performs joint time-frequency domain analysis on the multi-physical field coupling data set to obtain the fault feature vector of the high- and low-voltage power distribution cabinets, it specifically is used for: Perform frequency domain analysis on the electromagnetic radiation spectrum in the multi-physical field coupling data set, and extract the abnormal frequency band features with sudden energy changes; Perform time domain phase analysis on the cabinet surface vibration waveform in the multi-physical field coupling data set, and extract the phase mutation features related to the time sequence; Fuse the abnormal frequency band features and the phase mutation features to obtain a fault feature vector characterizing the fault mode.

[0033] Specifically, frequency domain analysis is a signal processing technique that converts a time domain signal to the frequency domain for analysis. Through frequency domain analysis, the energy distribution of the signal at different frequency components can be revealed, and it is often used to analyze the frequency characteristics, harmonic components, etc. of the signal. In this solution, frequency domain analysis is mainly used to process the electromagnetic radiation spectrum to discover abnormal frequency components and changes in energy distribution therein. Specifically, time domain phase analysis is a method for analyzing the phase information in a time domain signal. Phase reflects the relative position relationship of the signal on the time axis. By analyzing the time domain phase, the instantaneous change characteristics of the signal in the time series can be captured. In this technology, time domain phase analysis is mainly applied to the cabinet surface vibration waveform to extract phase mutation information related to faults. Specifically, the abnormal frequency band features refer to the feature information of a specific frequency range where significant sudden changes in energy occur after performing frequency domain analysis on the electromagnetic radiation spectrum. These abnormal frequency bands are often closely related to electromagnetic faults inside the power distribution cabinet (such as partial discharge, arc, etc.), and are important bases for judging electromagnetic faults. Specifically, the phase mutation feature is a feature detected by performing time domain phase analysis on the cabinet surface vibration waveform, where the phase suddenly jumps in the time series. Phase mutation is usually caused by mechanical component faults (such as looseness, wear, resonance, etc.), and is one of the key features reflecting mechanical faults. Furthermore, it is the process of integrating the features extracted from different physical field signals. Through feature fusion, the fault information contained in the electromagnetic and vibration signals can be organically combined to form a more comprehensive and representative fault feature vector, so as to more accurately describe the fault mode. The fault feature vector is a multi-dimensional vector, composed of the abnormal frequency band features and phase mutation features after feature fusion. This vector comprehensively characterizes the possible fault modes of the high- and low-voltage power distribution cabinets, and is the core data basis for subsequent fault diagnosis, location, and network modeling.

[0034] Specifically, the fast Fourier transform (FFT) is used to transform the electromagnetic radiation spectrum in the multi-physical field coupling dataset from the time domain to the frequency domain. FFT can efficiently decompose the time-domain signal into the superposition of different frequency components, thereby obtaining the amplitude and phase information of the electromagnetic radiation signal at each frequency and forming a complete frequency-domain spectrum diagram.

[0035] Furthermore, energy analysis is performed on the converted frequency-domain spectrum diagram to calculate the energy distribution in each frequency band. By setting a reasonable energy threshold, the frequency bands with significant energy mutations are screened out. For example, when the energy in a certain frequency band shows a significant increase or decrease compared to the energy in the normal operating state, this frequency band is marked as an abnormal frequency band. At the same time, key parameters of the abnormal frequency band, such as the center frequency, bandwidth, and energy mutation amplitude, are extracted as the characteristics of the abnormal frequency band. These characteristics can intuitively reflect the abnormal changes in the electromagnetic radiation signal and provide important clues for judging electromagnetic faults.

[0036] Specifically, the original vibration signal is transformed into an analytic signal through the Hilbert transform, and then the instantaneous phase information of the signal is obtained. The instantaneous phase can accurately describe the phase state of the vibration signal at each moment.

[0037] Specifically, the calculated instantaneous phase is analyzed to detect the phase mutation points. By setting an appropriate phase change threshold, when the change amount of the instantaneous phase exceeds this threshold, it is determined that a phase mutation has occurred. For example, when the derivative of the instantaneous phase exceeds a certain value (such as 0.5 rad / s), it is considered that a phase mutation occurs at this moment. Information such as the moment when the phase mutation occurs and the phase values before and after the mutation is recorded as the characteristics of the phase mutation. These characteristics can effectively reflect the fault state of mechanical components, such as abnormal changes in the vibration phase caused by component loosening.

[0038] Specifically, the characteristics of the abnormal frequency band and the characteristics of the phase mutation are combined. According to the pre-defined feature vector format, the parameters of the abnormal frequency band (such as the center frequency, energy mutation amplitude, etc.) and the relevant information of the phase mutation (such as the mutation moment, phase change amount, etc.) are arranged in sequence to form a multi-dimensional vector.

[0039] Furthermore, in order to eliminate the differences in dimension and numerical range between different characteristic parameters, the combined feature vector is normalized by Z-score. Through the normalization process, each element in the feature vector is within a reasonable numerical range.

[0040] Generally speaking, by performing targeted frequency-domain analysis on the electromagnetic radiation spectrum and time-domain phase analysis on the vibration waveform of the cabinet surface respectively, the fault characteristics contained in the two physical field signals can be deeply explored. Compared with the traditional single signal analysis method, the joint time-frequency domain analysis can capture the change characteristics of the signal more comprehensively, thus significantly improving the ability to identify fault characteristics. For example, for the fault of the coupling of partial discharge and mechanical resonance, the abnormal frequency band in the electromagnetic radiation can be found through frequency-domain analysis, and the phase mutation of the vibration waveform can be detected through time-domain phase analysis, effectively avoiding the omission of fault characteristics.

[0041] Generally speaking, the feature fusion technology organically integrates the fault characteristics from different physical fields, and the formed fault feature vector contains richer and more comprehensive fault information. This multi-dimensional fault feature description method can more accurately reflect the actual fault situation, effectively reduce the fault misjudgment rate, and improve the accuracy and reliability of fault diagnosis.

[0042] Generally speaking, the frequency-domain analysis of the electromagnetic radiation spectrum and the time-domain phase analysis of the vibration waveform of the cabinet surface are two parallel and independent feature extraction processes. They analyze and process the multi-physical field coupling data set from the two physical field dimensions of electromagnetism and vibration respectively, and extract the fault characteristics in their respective fields. These two steps are the core basic operations of the feature extraction module, providing the necessary original feature data for the subsequent feature fusion. The feature fusion step integrates the abnormal frequency band characteristics and phase mutation characteristics extracted in the above two steps, and generates the final fault feature vector through combination and normalization processing. As the output result of the entire feature extraction module, the fault feature vector not only synthesizes the fault characteristics of the electromagnetic and vibration signals, but also provides key input data for the subsequent network construction and update module, fault location module, etc. It is an important bridge connecting data acquisition and subsequent processing links such as fault diagnosis and location, and plays a key role in the entire intelligent detection system of high and low voltage switch cabinets.

[0043] The network construction and update module 103 is used to construct and dynamically update the fault propagation network based on the fault feature vector.

[0044] In the field of high and low voltage switch cabinet fault detection, the traditional technology uses a static topology model to analyze faults, which cannot dynamically reflect the energy transfer path during the fault propagation process, resulting in large positioning deviations and difficulty in adapting to complex fault scenarios. This network construction and update module constructs and dynamically updates the fault propagation network based on the fault feature vector, and realizes the accurate modeling and real-time tracking of the fault propagation process through innovative node mapping, edge generation and topology optimization mechanisms.

[0045] When the network construction and update module executes the construction and dynamic update of the fault propagation network based on the fault feature vector, it is specifically used for: Map the abnormal frequency band features in the fault feature vector to the nodes of the fault propagation network, where the weight of each node is determined by the electromagnetic radiation energy density corresponding to the abnormal frequency band feature; Generate directed connection edges between the nodes based on the time correlation of the phase mutation features in the fault feature vector, where the direction of the directed connection edge is determined by the timing relationship of the occurrence of the phase mutation features; Dynamically update the topological structure of the fault propagation network through an energy transfer probability model.

[0046] Specifically, the fault propagation network is a network model that represents the internal fault propagation path and state of high and low voltage switchgear in the form of a graph structure. The nodes in the network represent fault features, and the edges represent fault propagation relationships. The development process of the fault in the switchgear is reflected through the dynamic changes of the nodes and edges.

[0047] Specifically, in the fault propagation network, the nodes are mapped from the abnormal frequency band features in the fault feature vector, and each node corresponds to a specific electromagnetic radiation abnormal frequency band. The node weight is determined by the electromagnetic radiation energy density corresponding to this frequency band, which is used to quantify the importance of this frequency band in the fault features.

[0048] Specifically, the directed connection edge is a directed line segment connecting the nodes in the fault propagation network, and its direction is determined by the time sequence of the phase mutation features in the fault feature vector, representing the direction of fault propagation; the weight of the edge reflects the probability of the fault propagating through this path and is dynamically adjusted through the energy transfer probability model.

[0049] Specifically, the energy transfer probability model is a mathematical model constructed based on probability theory, which is used to describe the transfer probability of fault energy between the nodes of the fault propagation network. This model calculates and updates the energy transfer probability between the nodes by analyzing the energy relationship between the nodes and historical fault data, providing a basis for the dynamic update of the network topological structure.

[0050] Specifically, topological optimization refers to the process of adjusting and optimizing the node connection relationship, node attributes, etc. of the fault propagation network. By merging redundant nodes, adding key connection edges, etc., the network structure is simplified, the model calculation efficiency is improved, and at the same time the network can more accurately reflect the fault propagation law.

[0051] Furthermore, node mapping refers to using the abnormal frequency band features in the fault feature vector as the basis for constructing network nodes. For example, if the fault feature vector contains energy mutation information in the 30 - 50 kHz frequency band, a corresponding node is created in the fault propagation network. The weight of each node is determined according to the electromagnetic radiation energy density corresponding to this abnormal frequency band. The higher the energy density, the greater the node weight. For example, if the energy density of a certain frequency band is relatively high, the weight of its mapped node can be set to a relatively high value (such as 0.8), indicating that this frequency band occupies an important position in the fault features and may be a key manifestation of the fault.

[0052] Furthermore, the generation of directed connection edges refers to determining the connection relationship between nodes based on the time correlation of the phase mutation features in the fault feature vector. When it is detected that there is a phase mutation in the vibration signals associated with the abnormal frequency band features corresponding to two nodes, and there is a sequential time relationship in the phase mutation, a directed connection edge is generated from the node corresponding to the earlier phase mutation to the node corresponding to the later phase mutation. For example, if the vibration signal associated with node A undergoes a phase mutation at time t1, and the vibration signal associated with node B undergoes a phase mutation at time t2 (t2 > t1), a directed edge from node A to node B is generated to clarify the possible propagation path of the fault.

[0053] The network construction and update module performs dynamic update of the topological structure of the fault propagation network through the energy transfer probability model, specifically for: When it is detected that the difference in electromagnetic radiation energy density between nodes in the fault propagation network exceeds a preset difference threshold, an edge weight attenuation operation is triggered, and based on the energy transfer probability model, the edge weight corresponding to the low - probability propagation path in the directed connection edge is attenuated; When it is detected that the graph structure similarity between the newly added nodes in the fault propagation network and the historical topological structure is lower than a preset similarity threshold, a topological optimization instruction is triggered, and according to the optimization instruction, the topological structure is dynamically optimized, and the model parameters of the energy transfer probability model are updated; Synchronize the updated edge weights, the optimized topological structure, and the updated energy transfer probability model to the fault propagation network.

[0054] Specifically, edge weight attenuation refers to real - time monitoring of the difference in electromagnetic radiation energy density between nodes in the fault propagation network. When the energy density difference between nodes exceeds the preset difference threshold, it indicates that the fault propagation probability of some paths is relatively low, and an edge weight attenuation operation is triggered. Based on the energy transfer probability model, the edge weight corresponding to the low - probability propagation path in the directed connection edge is attenuated. For example, if the energy density difference between node X and node Y exceeds the threshold, the weight of the connection edge between them is reduced from the initial value of 0.6 to 0.2, weakening the role of this low - probability propagation path in the network and making the network more focused on high - probability fault propagation paths.

[0055] Specifically, the triggering condition for edge weight decay is as follows: ; Among them, is the judgment value of the triggering condition for edge weight decay, is the electromagnetic radiation energy density of node is the electromagnetic radiation energy density of node is the larger value of the energy densities of the two nodes, is the preset difference threshold (the value can be 0.3 - 0.5).

[0056] Specifically, the triggering condition for edge weight decay determines whether the connection edges between nodes in the fault propagation network need to have their weights decayed. When the difference in the electromagnetic radiation energy density between node and exceeds the preset threshold , decay is triggered; otherwise it is not triggered.

[0057] Furthermore, the value of the preset difference threshold is determined based on the statistical analysis of the energy density differences of 95% of the high-probability propagation paths in the historical fault case database, and is used to distinguish between "high-probability propagation paths" and "low-probability propagation paths"; the electromagnetic radiation energy density can be measured by electromagnetic sensors; when the energy density difference is large, decay is triggered.

[0058] Specifically, is used to normalize the difference to avoid judgment deviation caused by the absolute value difference of the energy density (for example, a small difference between high-energy density nodes may be more meaningful than a large difference between low-energy density nodes); the triggering condition judgment value (0 or 1) is used to control whether to perform the edge weight decay operation.

[0059] Furthermore, edge weight decay involves local updates of the fault propagation network, while topology optimization and parameter updates involve global updates of the fault propagation network.

[0060] Specifically, update result synchronization means synchronizing the edge weight values after edge weight decay, the network structure after topology optimization, and the updated energy transfer probability model parameters to the fault propagation network, so that the network can reflect the latest state of fault propagation in real time and provide an accurate analysis model for subsequent fault location.

[0061] Generally speaking, by converting the fault feature vectors into network nodes and edges and using the energy transfer probability model for dynamic updates, the dynamic modeling of the fault propagation process is realized, effectively solving the problem of difficult tracking of fault propagation paths and providing a reliable basis for quickly locating the fault source.

[0062] Generally speaking, the edge weight decay and topology optimization mechanisms endow the fault propagation network with adaptability. When there are changes in the energy distribution or new fault features in the network, through local updates and global optimization, the network can automatically adjust its structure and parameters, remove redundant information, strengthen key propagation paths, avoid redundant network structures, improve the computational efficiency of the model, ensure that the network always operates in an optimal state, and adapt to the analysis requirements under different fault scenarios.

[0063] Generally speaking, a network is constructed by combining the electromagnetic radiation energy density and the vibration phase mutation characteristics, making full use of the advantages of multi-physical field coupling data. The electromagnetic energy characteristics are reflected by the node weights, and the time correlation of the vibration phase is reflected by the directed edges, realizing the deep fusion of multi-physical field fault characteristics, effectively capturing the complex relationships of fault propagation, and significantly improving the analysis ability of multi-physical field coupling faults and reducing the risk of false fault judgment compared with single physical quantity analysis methods.

[0064] When the network construction and update module executes the dynamic optimization of the topology structure according to the optimization instruction and updates the model parameters of the energy transfer probability model, it is specifically used for: Adjust the node connection relationship of the fault propagation network according to the optimization instruction to generate an optimized topology structure, including: When the difference in electromagnetic radiation energy density between two nodes in the fault propagation network is less than the preset merging threshold and the connection edge directions are the same, they are merged into one node; Add key connection edges between unconnected nodes according to the high-probability propagation paths in the historical fault case library; Calculate the energy transfer probability between nodes in the optimized topology structure according to the Markov chain transition probability, and iteratively update the model parameters of the energy transfer probability model through the gradient descent algorithm.

[0065] Specifically, the Markov chain transition probability is based on the Markov chain theory and describes the probability of the node state transition in the fault propagation network from one moment to the next. Using this probability to construct an energy transfer probability model can effectively analyze the dynamic process of fault propagation and provide quantitative support for network update.

[0066] Furthermore, according to the Markov chain transition probability model based on fault energy transfer, calculate the energy transfer probability between nodes in the optimized topology structure, where the initial value of the transition probability is determined by the high-probability propagation paths in the historical fault case library.

[0067] Specifically, the calculation formula of the transition probability is as follows: ; Among them, is the probability of the fault propagating from node a to node b through the edge, is the weight of the directed connection edge from node a to node b, is the sum of the weights of all outgoing edges of node a (i.e., the edges pointing from a to other nodes).

[0068] Furthermore, the calculation formula of the transition probability calculates the energy transfer probability from node a to node b in the fault propagation network, which is used to describe the probability distribution of the fault energy transferring from the current node to the adjacent node, reflects the "intensity" of the fault propagation of the path from node a to node b, which is determined by the difference in the electromagnetic radiation energy density between nodes, is used for normalization to ensure that the sum of the transition probabilities is 1.

[0069] Specifically, the greater the difference in the electromagnetic radiation energy density, the lower the transition probability.

[0070] Furthermore, the calculation formula of the directed connection edge weight is as follows: ; where, is the weight of the directed connection edge from node a to node b, is the electromagnetic radiation energy density of node a, is the electromagnetic radiation energy density of node b, is the energy attenuation coefficient.

[0071] Furthermore, assume a preset merging threshold , when the difference in the electromagnetic radiation energy density between node a and node b in the fault propagation network is less than , and the directions of the connection edges are the same, then node a and node b are merged into one node.

[0072] Specifically, the energy attenuation coefficient is obtained through the fitting and optimization of the energy transfer paths in the historical fault case library to ensure that the model error is less than 5%.

[0073] Furthermore, the greater the energy density difference , the smaller the exponential function value, and the lower the edge weight . At the same time, due to the characteristics of the exponential function, is always in the interval (0,1], ensuring the effectiveness of the transition probability calculation.

[0074] Furthermore, when node a and node b meet the following two conditions, they are merged into one node: the difference in the electromagnetic radiation energy density (the preset merging threshold ); the directions of the connection edges are the same, ensuring the consistency of the merged node in the fault propagation direction.

[0075] Specifically, the gradient descent algorithm is a commonly used optimization algorithm that iteratively adjusts model parameters by calculating the gradient direction of the objective function to minimize the value of the objective function. In this solution, it is used to update the parameters of the energy transfer probability model to optimize the model's ability to describe fault propagation.

[0076] Furthermore, the loss function of the gradient descent algorithm is the mean squared error function MSE, and the calculation formula of the loss function is as follows: ; where, is the value of the loss function, is the number of samples, is the predicted energy transfer frequency between node and node , is the actual energy transfer frequency between node and node recorded in the historical fault case library.

[0077] Furthermore, the value of the loss function represents the overall measure of the model prediction error, and its value range is [0, +∞). The smaller the value, the higher the prediction accuracy; the number of samples is the number of node pairs with energy transfer records in the historical fault case library; the predicted energy transfer frequency is calculated through the energy transfer probability model; the actual energy transfer frequency reflects the actual propagation probability of faults between node and node in the historical data and is used to supervise model training.

[0078] For example: Extract the actual propagation frequencies of all node pairs from the historical case library, calculate the predicted frequencies through the current model parameters, calculate the MSE as the loss, and backpropagate the gradient to update parameters such as β, and iterate until the loss converges or the error < 5%.

[0079] Furthermore, the actual energy transfer frequency is obtained by dividing the number of fault propagations from node to node by the total number of fault propagations of the node.

[0080] Specifically, topology optimization refers to calculating the graph structure similarity between newly added nodes and the historical topology structure when new nodes are detected in the fault propagation network. If the similarity is lower than the preset similarity threshold, a topology optimization instruction is triggered. For example, the graph edit distance can be used to calculate the graph structure similarity, and the similarity threshold is set to 0.7. At this time, the network topology structure is adjusted according to the instruction: when the difference in electromagnetic radiation energy density between two nodes is less than the preset merging threshold and the connection edge directions are the same, these two nodes are merged into one node to reduce redundant nodes; at the same time, referring to the high-probability propagation paths recorded in the historical fault case library, key connection edges are added between nodes that were not originally connected but conform to the high-probability propagation logic to improve the network topology structure and make it more conform to the actual fault propagation law.

[0081] Specifically, parameter update refers to calculating the energy transfer probability between nodes in the optimized topology structure using the Markov chain transition probability and iteratively updating the parameters of the energy transfer probability model through the gradient descent algorithm. In each iteration, according to the difference between the calculated energy transfer probability and the actual fault data, the model parameters are adjusted to make the model more accurately describe the fault propagation process and improve the adaptability of the network to the dynamic changes of faults.

[0082] Generally speaking, through node mapping and directed connection edge generation, the fault feature vector is transformed into a network structure. Based on the constructed network, the network is adjusted through edge weight attenuation, topology optimization, and parameter update. Edge weight attenuation optimizes the local low-probability propagation paths in the network, while topology optimization and parameter update improve the network structure and model parameters from a global perspective. The two cooperate with each other. The local update quickly responds to the energy distribution changes, and the global update adapts to the major adjustments of the network structure, jointly ensuring the accuracy and efficiency of the network. Finally, the optimized network state is fed back to the entire system to provide the latest and accurate network model for the fault location module, forming a tightly collaborative closed loop among modules to achieve efficient detection and analysis of high- and low-voltage switchgear faults.

[0083] The fault location module 104 is used to match the betweenness centrality and energy distribution of nodes in the fault propagation network with the fault modes in the historical fault case library to identify the key fault sources of the high- and low-voltage switchgear and mark the potential fault diffusion areas.

[0084] During the operation and maintenance of high- and low-voltage switchgear, traditional fault location technologies have large positioning deviations due to their inability to dynamically track fault propagation paths, making it difficult to quickly and accurately lock the fault sources and potential diffusion areas. This fault location module realizes the accurate identification of key fault sources and the effective marking of potential fault diffusion areas by integrating the topological characteristics and physical features of the fault propagation network and combining with the historical fault case library, fundamentally solving the positioning problem in the background technology.

[0085] When the fault location module performs matching with the fault patterns in the historical fault case library based on the betweenness centrality and energy distribution of the nodes in the fault propagation network to identify the key fault sources of the high- and low-voltage switchgears and mark the potential fault diffusion areas, it is specifically used for: Using the Brandes algorithm to calculate the betweenness centrality of each node in the fault propagation network, including: Traversing all nodes in the fault propagation network as source nodes, calculating the shortest paths from the source nodes to other nodes based on breadth-first search, and counting the number of occurrences of each node in all the shortest paths and normalizing them to betweenness values; Extracting the energy distribution of the electromagnetic radiation energy density from the node weights of the fault propagation network; Constructing a multi-dimensional feature vector of the betweenness centrality and the energy distribution, and calculating the matching degree between the multi-dimensional feature vector and the fault patterns in the historical fault case library. The historical fault case library stores the mapping relationship between the fault patterns and the node attributes through an incidence matrix; According to the sequence list of the betweenness centrality and the matching degree, screening the nodes with betweenness centrality higher than the preset betweenness centrality threshold and the highest matching degree score as the key fault sources; Traversing the adjacent nodes of the key fault sources, and identifying the adjacent nodes with a high-probability propagation path to the key fault sources according to the energy transfer probability model, and marking them as potential fault diffusion areas.

[0086] Specifically, betweenness centrality is an index used to measure the importance of nodes in a network. In a fault propagation network, the higher the betweenness centrality of a certain node, the more the number of shortest paths passing through this node, which means that this node plays a stronger hub role in the fault propagation process and is more likely to be a key fault source or a key node in the fault propagation.

[0087] Specifically, the Brandes algorithm is an efficient algorithm for calculating the betweenness centrality of network nodes, and its time complexity is O(nm) (n is the number of nodes, m is the number of edges). This algorithm traverses all nodes in the network as source nodes, calculates the shortest paths from the source nodes to other nodes, and counts the number of occurrences of each node in these shortest paths, so as to obtain the betweenness centrality of the nodes.

[0088] Specifically, energy distribution refers to the distribution of the electromagnetic radiation energy density corresponding to each node in the fault propagation network. The node weights are determined by the electromagnetic radiation energy density, and the energy distribution reflects the energy aggregation degree of the fault in different regions, which is an important basis for judging the severity and propagation trend of the fault.

[0089] Specifically, the multi-dimensional feature vector is a vector composed of features in multiple dimensions such as the betweenness centrality and energy distribution of nodes in the fault propagation network. This vector comprehensively describes the topological attributes and physical characteristics of nodes and is used for matching analysis with fault patterns in the historical fault case database.

[0090] Specifically, the cosine similarity algorithm is an algorithm used to calculate the similarity between two vectors. By calculating the cosine value of the angle between two vectors, it measures the degree of similarity in their directions. In this solution, it is used to calculate the matching degree between the current multi-dimensional feature vector and the fault pattern vector in the historical fault case database. The closer the cosine value is to 1, the higher the matching degree.

[0091] Specifically, the historical fault case database is a database that stores a large amount of historical fault information of high and low voltage distribution cabinets. In the form of an association matrix, it records the mapping relationships between different fault patterns and node attributes (such as betweenness centrality, energy distribution, fault feature vectors, etc.), providing a reference basis for current fault location.

[0092] Specifically, the key fault source refers to the core location or component that causes faults in high and low voltage distribution cabinets. Through the analysis of the fault propagation network, combined with betweenness centrality and the matching degree with historical fault patterns, the node most likely to be the origin of the fault is selected as the key fault source.

[0093] Specifically, the potential fault spread area is the area where faults may further spread and have an impact, identified based on the connection relationships and energy transfer probabilities between the key fault source and its adjacent nodes in the fault propagation network. Marking the potential fault spread area helps maintenance personnel take measures in advance to prevent the expansion of faults.

[0094] Furthermore, the Brandes algorithm is used to process the fault propagation network. Calculating the betweenness centrality of nodes includes the following steps: First, all nodes in the network are traversed in turn as source nodes. For each source node, the shortest paths from this source node to all other nodes in the network are calculated based on the breadth-first search algorithm. During the calculation process, the nodes passed by each shortest path are recorded; after the traversal is completed, the number of occurrences of each node in all shortest paths is counted. To facilitate the comparison of the importance of nodes in different-scale networks, the number of occurrences is normalized to obtain the betweenness value of each node. For example, if node A appears 50 times in all shortest path calculations, after normalization, its betweenness value may be 0.7 (the specific value is determined according to the network scale and calculation method), indicating that node A has a high betweenness centrality in the fault propagation network and may play a key role in the fault propagation process.

[0095] Furthermore, the electromagnetic radiation energy density information corresponding to each node is directly extracted from the node weights of the fault propagation network to form the energy distribution of the entire network. Since the node weights have been determined by the electromagnetic radiation energy density corresponding to the abnormal frequency band characteristics during the network construction process, the energy-related data of each node can be directly obtained. By analyzing these data, the distribution of fault energy at different positions inside the power distribution cabinet can be understood. For example, if the energy density of certain nodes is relatively high, it indicates that relatively serious faults or fault development trends may exist in these areas.

[0096] Specifically, the calculated node betweenness centrality and the extracted energy distribution information are combined to construct a multi-dimensional feature vector. For example, for a certain node, its multi-dimensional feature vector may be [betweenness centrality = 0.85, energy density = 0.7], and this vector comprehensively describes the attributes of the node in terms of network topology and physical characteristics.

[0097] Furthermore, using the cosine similarity algorithm, the constructed multi-dimensional feature vector is calculated for its matching degree with the fault mode vectors stored in the historical fault case library. The historical fault case library stores the node attribute feature vectors corresponding to different fault modes through an association matrix. For example, the arc fault mode may correspond to a feature vector of [betweenness centrality = 0.82, energy density = 0.68]. Calculate the cosine similarity between the multi-dimensional feature vector of the current node and each historical fault mode vector to obtain the matching degree score of the node with different fault modes. For example, if the calculated cosine similarity is 0.98, it means that the node has a high matching degree with the corresponding historical fault mode, and the fault situation represented by this node is relatively similar to a certain fault mode in the historical case.

[0098] Specifically, according to the calculated betweenness centrality sequence list and the matching degree score, a preset betweenness centrality threshold (such as 0.8) is set. Screen out the nodes with betweenness centrality higher than this threshold and the highest matching degree score, and determine them as the key fault sources. For example, the betweenness centrality of node B is 0.9, and its matching degree score with the historical fault mode is 95%, which is higher than other nodes and meets the screening conditions. Then, it can be determined that node B is the key fault source of the current high and low voltage power distribution cabinet, that is, the fault is most likely to originate from the position or component represented by this node.

[0099] Specifically, after determining the key fault source, traverse its adjacent nodes. According to the energy transfer probability model, analyze the energy transfer probability between the key fault source and each adjacent node. Identify the adjacent nodes with a high-probability propagation path to the key fault source, and label these nodes and their regions as potential fault diffusion regions. For example, if the energy transfer probability between node C and the key fault source node B is high (such as exceeding 0.7), then label node C and its surrounding relevant regions as potential fault diffusion regions, prompting the operation and maintenance personnel that this region may be affected by the fault and requires key attention and protective measures.

[0100] Specifically, by calculating the node betweenness centrality, nodes that play a key role in fault propagation can be screened from the perspective of the network topology; combined with the matching of energy distribution and the historical fault case library, the relevance between nodes and fault modes is further verified from the physical characteristics and historical experience levels. This multi-dimensional analysis method, compared with the traditional single-location method based on physical quantities or simple topology, significantly improves the fault identification accuracy to 98%, effectively solves the problem of large fault location deviation in the background technology, and can quickly and accurately lock the root cause of the fault.

[0101] Specifically, the potential fault diffusion region is marked based on the energy transfer probability model, fully considering the propagation law and possibility of faults inside the power distribution cabinet. By identifying high-probability affected regions in advance, the operation and maintenance personnel can take targeted preventive measures, such as strengthening monitoring, arranging maintenance in advance, etc., to avoid the further expansion of faults, reduce the power outage range and economic losses, and improve the reliability and stability of the power system operation.

[0102] Specifically, the introduction of the historical fault case library provides a rich reference basis for fault location. By matching with historical fault modes, the experience of previous fault handling can be used to quickly judge the type and possible causes of the current fault, avoid misjudgment or missed judgment caused by lack of experience in new fault situations, and enhance the reliability and stability of fault location, especially suitable for complex and changeable power distribution cabinet fault scenarios.

[0103] Generally speaking, calculating the betweenness centrality of nodes and extracting the energy distribution are the basic steps of fault location. They analyze the fault propagation network from two dimensions of network topology and physical characteristics respectively, providing the original data for constructing the multi-dimensional feature vector later. The construction of the multi-dimensional feature vector and the matching degree calculation link integrate the features of the above two dimensions, establish a connection with the historical fault case library, and quantify the matching degree between the current fault features and the historical patterns through the cosine similarity algorithm. The key fault source identification step, based on the betweenness centrality and matching degree calculated previously, screens out the most likely fault sources by setting thresholds, which is the core link of fault location. Finally, the potential fault diffusion area annotation starts from the key fault source, analyzes the possibility of fault propagation using the energy transfer probability model, and determines the areas that may be affected.

[0104] The positioning map generation module 105 is used to map the key fault source and the potential fault diffusion area to the physical space coordinates of the high and low voltage distribution cabinets to obtain the fault positioning map of the high and low voltage distribution cabinets.

[0105] In the fault detection and handling process of the high and low voltage distribution cabinets, although the key fault source and the potential fault diffusion area have been identified through the fault location module, the abstract network node information is difficult to be directly applied to on-site operation and maintenance. The traditional method lacks an intuitive means of displaying the fault space distribution, resulting in difficulty for the operation and maintenance personnel to quickly master the fault location and affecting the fault handling efficiency. This positioning map generation module generates a visual fault positioning map by mapping the fault information to the physical space coordinates of the distribution cabinets, providing intuitive and accurate fault location information for the operation and maintenance personnel, and effectively solving the pain points in the background technology.

[0106] When the positioning map generation module executes mapping the key fault source and the potential fault diffusion area to the physical space coordinates of the high and low voltage distribution cabinets to obtain the fault positioning map of the high and low voltage distribution cabinets, it specifically is used for: Extracting the node identifier of the key fault source and the node set of the potential fault diffusion area from the fault propagation network; Invoking the pre-defined sensor layout coordinate mapping table of the high and low voltage distribution cabinets, and mapping the node identifier and the node set to the physical space coordinates of the high and low voltage distribution cabinets based on the sensor layout coordinate mapping table; Generating the fault positioning map of the high and low voltage distribution cabinets based on the physical space coordinates, and the fault positioning map marks the spatial positions of the key fault source and the potential fault diffusion area.

[0107] Specifically, the node identifier is the unique identification symbol for each node in the fault propagation network, used to distinguish different fault feature nodes. During the network construction process, each node mapped by the abnormal frequency band feature will be assigned a specific identifier, such as "node1", "node2", etc., which facilitates accurate identification and operation in subsequent processing. Specifically, the node set refers to a set composed of multiple nodes, which is used in this solution to represent all the nodes involved in the potential fault diffusion area. These nodes are identified as nodes with a high-probability propagation path to the key fault source through the energy transfer probability model, and they jointly constitute the area that may be affected by the fault. Specifically, the sensor layout coordinate mapping table is a pre-defined and stored table that records the correspondence between the installation positions of each sensor in the high- and low-voltage switchgear and the physical space coordinates (including X, Y, and Z-axis coordinates), and at the same time establishes the mapping association between the sensor detection data and the nodes of the fault propagation network. This mapping table is the key data basis for realizing the conversion from network nodes to physical space coordinates. Specifically, the physical space coordinates are used to describe the specific position information of each position inside the high- and low-voltage switchgear in three-dimensional space. The spatial position of each point is accurately determined by the X, Y, and Z-axis coordinate values, such as "(x = 1.2m, y = 0.8m, z = 0m)", so that the fault position can be clearly located in the actual cabinet.

[0108] Specifically, the fault location atlas is a visual chart that marks the information of the key fault source and the potential fault diffusion area in an intuitive way on the physical space layout of the high- and low-voltage switchgear. The key fault source and the potential diffusion area are distinguished by different visual elements such as colors and shapes, providing clear and easy-to-understand fault position distribution information for the operation and maintenance personnel. Furthermore, extracting node information is to accurately extract the node identifier corresponding to the key fault source and the node set of the potential fault diffusion area from the fault propagation network output by the fault location module. For example, in a certain fault analysis, the fault location module determines that the node "node5" is the key fault source, and at the same time identifies that nodes such as "node3", "node6", and "node8" constitute the potential fault diffusion area. Then, "node5" is extracted as the node identifier of the key fault source, and "{node3, node6, node8}" is extracted as the node set of the potential fault diffusion area. This step is the premise for subsequent mapping and atlas generation, ensuring the acquisition of accurate fault-related node information.

[0109] Furthermore, calling the mapping table means that the system calls the pre-stored sensor layout coordinate mapping table, which is configured during the installation and commissioning stage of the switchgear and records the association relationship between the detection area corresponding to each sensor and the network node, as well as the physical installation coordinates of the sensor. Furthermore, the coordinate transformation is based on the sensor layout coordinate mapping table, and the identified key fault source nodes and the set of nodes in the potential fault propagation area are mapped one by one to the physical space coordinates of the high- and low-voltage switchgear. In specific implementation, by querying the corresponding relationship between the node identifier and the sensor in the mapping table, the installation coordinates of the sensor are obtained, which are the physical space coordinates of the fault location represented by the node in the cabinet. For example, if the mapping table shows that the node "node5" corresponds to the sensor installed at the coordinate "(x = 1.5m, y = 1.0m, z = 0.5m)", then this coordinate is determined as the physical space coordinate of the key fault source "node5"; for the set of nodes in the potential fault propagation area, the coordinates corresponding to each node are also obtained through the mapping table.

[0110] Furthermore, the data processing is to organize and format the physical space coordinate data of the identified key fault source and the potential fault propagation area for the visualization engine to process. Furthermore, the visualization presentation refers to using professional visualization technologies such as the D3.js engine, ECharts, etc. to generate a fault location map of the high- and low-voltage switchgear based on the processed data. During the map generation process, different visualization elements are used to label the key fault source and the potential fault propagation area: usually, the key fault source is marked with eye-catching red and the specific coordinate information is labeled; the potential fault propagation area is marked with a yellow or orange range to clearly show the area that the fault may affect. At the same time, auxiliary information such as the outline of the cabinet and the sensor positions can also be added to the map, enabling the operation and maintenance personnel to more intuitively understand the relationship between the fault and the cabinet structure. For example, in the generated map, the key fault source is marked with a red dot at the coordinate "(x = 1.5m, y = 1.0m, z = 0.5m)", and the potential fault propagation area is covered with a yellow shadow over the relevant coordinate area, visually presenting the fault distribution.

[0111] Furthermore, transforming the abstract fault propagation network node information into an intuitive physical space position map avoids the difficulty for operation and maintenance personnel to understand complex network data. Through the visual fault location map, the operation and maintenance personnel can quickly locate the specific positions of the key fault source and the potential fault propagation area in the switchgear, without the need for complex data analysis and spatial imagination, significantly improving the efficiency of obtaining fault information.

[0112] Furthermore, the coordinate mapping step relies on a pre-defined sensor layout coordinate mapping table to transform node information into physical space coordinates, which is a key link connecting abstract network data and actual physical positions, providing an accurate position data basis for map generation.

[0113] The self-healing control module 106 is configured to output a self-healing control instruction for the high- and low-voltage power distribution cabinet based on the fault location map. The self-healing control instruction includes applying a reverse electromagnetic pulse to the key fault source and adjusting the contact pressure in the potential fault propagation area.

[0114] During the operation of the high- and low-voltage power distribution cabinet, traditional self-healing control strategies such as global power-off and mechanical locking, which are coarse-grained intervention methods, can inhibit the development of faults to a certain extent, but will inevitably cause misoperations in normal areas, resulting in unnecessary power outage losses and seriously affecting the reliability and stability of power supply. Based on the fault location map, this self-healing control module uses a precise control strategy of applying a reverse electromagnetic pulse to the key fault source and adjusting the contact pressure in the potential fault propagation area to achieve efficient suppression and isolation of faults, effectively solving the problem of insufficient accuracy of self-healing control in the background technology.

[0115] The execution of the self-healing control instruction in the self-healing control module includes: According to the coordinates of the key fault source in the fault location map, applying a reverse electromagnetic pulse in the corresponding area of the high- and low-voltage power distribution cabinet, the propagation direction of which is opposite to that of the electromagnetic wave with the abnormal frequency band characteristics; According to the coordinates of the potential fault propagation area in the fault location map, adjusting the contact pressure of the contacts in the potential fault propagation area to a target pressure value, and the target pressure value is dynamically calculated according to the propagation risk level of the energy transfer probability model.

[0116] Specifically, the reverse electromagnetic pulse refers to an electromagnetic pulse signal with the same frequency as the abnormal electromagnetic wave generated by the key fault source of the high- and low-voltage power distribution cabinet but with the opposite propagation direction. By emitting a reverse electromagnetic pulse, the abnormal electromagnetic radiation generated by the fault source is canceled out using the principle of electromagnetic interference, thereby inhibiting the further development of the fault.

[0117] Specifically, the parameters of the reverse electromagnetic pulse are as follows: The frequency range of the reverse electromagnetic pulse is the center frequency of the detected abnormal frequency band ±10%, and the pulse intensity is 120% of the electromagnetic radiation energy of the fault source.

[0118] Furthermore, the algorithm for generating the parameters of the reverse electromagnetic pulse is as follows: ; where is the frequency of the reverse electromagnetic pulse, is the frequency of the fault source, is the phase of the reverse electromagnetic pulse, is the phase of the fault source, is the phase difference; Specifically, the phase difference The core parameters for achieving destructive interference ensure that the amplitudes cancel each other out when two electromagnetic waves are superimposed in space; the frequency of the reverse electromagnetic pulse is exactly the same as the frequency of the fault source to ensure that the reverse pulse is of the same frequency as the abnormal electromagnetic wave and meets the interference conditions; the phase of the reverse electromagnetic pulse (unit: rad) is obtained by adding a (180° phase difference) on the basis of the fault source phase and taking the modulus 2 to make the reverse pulse have the opposite phase to the abnormal electromagnetic wave.

[0119] Furthermore, only electromagnetic waves with the same frequency can produce a stable interference phenomenon, which is a necessary prerequisite for interference; the phase difference makes the electric field strength vector directions of the reverse pulse and the abnormal electromagnetic wave opposite, and destructive interference occurs after superposition, thereby suppressing the electromagnetic radiation of the fault source.

[0120] Specifically, parameters of a reverse electromagnetic pulse with the same frequency as the fault source but opposite in phase are generated, and the abnormal electromagnetic radiation generated by the fault source is cancelled out through the principle of electromagnetic interference to suppress the development of the fault.

[0121] Specifically, let the electromagnetic wave of the fault source be , and the reverse pulse be . After superposition, the total field strength is completely cancelled out (ideally). The formula achieves inversion through the phase difference , which conforms to Maxwell's equations and the principle of superposition of electromagnetic waves.

[0122] Specifically, the contact pressure of the electrical contacts inside the high and low voltage switchgear refers to the contact pressure between the electrical contacts. This pressure directly affects the electrical conductivity and mechanical stability of the contacts. When the contact pressure is insufficient, faults such as an increase in contact resistance, heating, and even arcing are likely to occur; reasonably adjusting the contact pressure can effectively improve the electrical connection performance and reduce the probability of faults. Specifically, the target pressure value is calculated based on the energy transfer probability model, which is the numerical value of the contact pressure that the contacts in the potential fault diffusion area should be adjusted to. This value is dynamically determined according to the risk level of fault propagation, aiming to suppress the propagation of faults in the diffusion area by optimizing the contact pressure. Furthermore, the formula for adjusting the contact pressure is as follows: ; where, is the target pressure value, is the basic pressure, is the risk level coefficient, is the propagation risk level (1 - 3).

[0123] Specifically, the formula for adjusting the contact pressure dynamically calculates the target contact pressure , improve the electrical connection performance by adjusting the contact pressure and suppress the spread of faults.

[0124] Further, the base pressure is set according to the minimum contact pressure requirement of the contacts in "GB / T 14048.1-2020 Low-voltage switchgear and controlgear" to ensure the basic electrical conductivity of the contacts; the risk level coefficient can be determined through experiments and is used to adjust the increment range of the pressure with the risk level. For example, in a high-risk scenario, the contact resistance needs to be reduced to below 5 mΩ, corresponding to .

[0125] Specifically, the propagation risk level is evaluated by the energy transfer probability model and takes values of 1 (low risk), 2 (medium risk), and 3 (high risk). The larger the value, the higher the probability of fault spread, which means the greater the pressure increment of the contacts.

[0126] Specifically, the energy transfer probability model calculates the fault propagation risk levels of different regions by analyzing the energy relationships between nodes in the fault propagation network and historical fault data, providing a quantitative basis for the self-healing control strategy.

[0127] Further, from the fault location map output by the fault location map generation module, extract the three-dimensional physical space coordinates (X, Y, and Z-axis coordinates) of the key fault source. For example, if the coordinates of the key fault source are (X = 1.2 m, Y = 0.8 m, Z = 0.5 m), this coordinate information will be used as the spatial positioning basis for subsequent electromagnetic pulse application. Further, device deployment and parameter configuration refer to deploying a phased array antenna device (such as a phased array antenna with a 4×4 array) near the area corresponding to the coordinates of the key fault source inside the high- and low-voltage switchboards. The phased array antenna can achieve the directional emission of electromagnetic wave beams by controlling the phases of each antenna element. According to the abnormal frequency band characteristics obtained by the fault feature extraction module, determine the electromagnetic wave frequency to be canceled (such as detecting an abnormal frequency band of 58 kHz), and set the emission frequency of the phased array antenna to this frequency; at the same time, adjust the antenna phase parameters so that the electromagnetic pulse it emits is in the opposite direction to the propagation direction of the electromagnetic wave generated by the fault source (with a phase difference of 180°). Further, pulse emission and interference cancellation refer to starting the phased array antenna to emit a reverse electromagnetic pulse towards the key fault source area. Using the principle of electromagnetic interference, when the reverse electromagnetic pulse meets the abnormal electromagnetic wave generated by the fault source, the two are superimposed and canceled, reducing the electromagnetic radiation intensity in the fault area. For example, in a certain partial discharge fault, after applying the reverse electromagnetic pulse, it is monitored that the electromagnetic interference intensity in this area has decreased by 15 dB, effectively suppressing the development of the partial discharge fault.

[0128] Furthermore, region location and data acquisition refer to obtaining the coordinate range of the potential fault diffusion region and the included node information from the fault location atlas. Meanwhile, the energy transfer probability model is retrieved to calculate the fault propagation risk level of this region. For example, if the potential fault diffusion region contains 3 nodes, after evaluation by the energy transfer probability model, its propagation risk level is "high". Furthermore, target pressure value calculation refers to dynamically calculating the target pressure value according to the preset pressure adjustment rule and the propagation risk level of the energy transfer probability model. The specific implementation method can adopt a pre-established pressure-risk level mapping table or be calculated through an algorithm. For example, when the risk level is "high", the target pressure value is determined to be 50N according to the mapping table; if the risk level is "low", the target pressure value is set to 30N. Furthermore, pressure adjustment execution means that within the potential fault diffusion region, a servo motor drive device (such as Panasonic MINASA6 series servo motors) is installed for each contact. The servo motor drives the contact adjustment mechanism to act through high-precision position control and torque control. The system converts the calculated target pressure value into a control signal for the servo motor to precisely adjust the contact pressure of the contact to the target value. For example, when the target pressure value is 45N, after receiving the control signal, the servo motor precisely adjusts the contact pressure to 45N through a transmission device such as a lead screw nut mechanism or a lever mechanism, enhancing the electrical connection performance of the contact, blocking the mechanical vibration propagation path, and preventing the further spread of the fault.

[0129] Generally speaking, through the accurate position information provided by the fault location atlas, the reverse electromagnetic pulse can directly act on the key fault source, and use the principle of electromagnetic interference to specifically eliminate abnormal electromagnetic radiation; the contact pressure adjustment is dynamically adjusted according to the actual risk level of the potential fault diffusion region and precisely acts on the contact parts that may be affected. Compared with traditional extensive intervention means such as global power-off, this solution realizes the precise location and precise control of faults, avoids misoperation of equipment in normal areas, and significantly improves the accuracy of self-healing control.

[0130] Generally speaking, the application of the reverse electromagnetic pulse can quickly reduce the electromagnetic interference intensity of the fault source and inhibit the development of electromagnetic faults such as partial discharge; the adjustment of the contact pressure can effectively improve the electrical connection performance, reduce the contact resistance and mechanical vibration, and prevent the further spread of the fault through the contact connection part. The combined action of the two can effectively suppress the fault within a short time, reduce the impact of the fault on the overall operation of the distribution cabinet, and improve the reliability and stability of the power system.

[0131] Generally speaking, both the steps of reverse electromagnetic pulse application and contact pressure adjustment are based on the fault location map to obtain the position information of the key fault sources and potential fault propagation regions, ensuring that the control strategy can accurately act on the fault-related regions. During the specific implementation process, the reverse electromagnetic pulse application mainly suppresses the root causes of electromagnetic faults, while the contact pressure adjustment focuses on blocking the fault propagation path from the aspects of mechanical connection and electrical performance. The two cooperate with each other and work together to form a complete self-healing control strategy. At the same time, the energy transfer probability model provides data support for the risk level of fault propagation during the contact pressure adjustment process, providing a basis for the calculation of the target pressure value.

[0132] Referring to Figure 2 As shown, it is a schematic flowchart of the intelligent detection method for high- and low-voltage switchgear provided by an embodiment of the present invention. In this embodiment, the intelligent detection method for high- and low-voltage switchgear includes: S1. Synchronously collect the electromagnetic radiation spectrum and the cabinet surface vibration waveform of the high- and low-voltage switchgear to generate a multi-physical field coupling data set; S2. Perform a joint time-frequency domain analysis on the multi-physical field coupling data set to obtain the fault feature vector of the high- and low-voltage switchgear; S3. Construct and dynamically update a fault propagation network based on the fault feature vector; S4. Based on the betweenness centrality and energy distribution of the nodes in the fault propagation network, match with the fault modes in the historical fault case library to identify the key fault sources and mark the potential fault propagation regions; S5. Map the key fault sources and the potential fault propagation regions to the physical space coordinates of the high- and low-voltage switchgear to obtain the fault location map of the high- and low-voltage switchgear; S6. Output the self-healing control instruction of the high- and low-voltage switchgear based on the fault location map, where the self-healing control instruction includes applying a reverse electromagnetic pulse to the key fault source and adjusting the contact pressure of the potential fault propagation region.

[0133] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention.

[0134] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is a theory, method, technology and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0135] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An intelligent detection system for high and low voltage switchgear, characterized in that, The system includes: A data acquisition module, configured to synchronously acquire the electromagnetic radiation spectrum and the cabinet surface vibration waveform of the high- and low-voltage power distribution cabinets, and generate a multi-physical-field coupling data set; A feature extraction module, configured to perform joint time-frequency domain analysis on the multi-physical-field coupling data set to obtain the fault feature vector of the high- and low-voltage power distribution cabinets; A network construction and update module, configured to construct and dynamically update a fault propagation network based on the fault feature vector; A fault location module, configured to match the betweenness centrality and energy distribution of the nodes in the fault propagation network with the fault patterns in the historical fault case base, identify the key fault sources of the high- and low-voltage power distribution cabinets, and mark the potential fault diffusion regions; A location map generation module, configured to map the key fault sources and the potential fault diffusion regions to the physical space coordinates of the high- and low-voltage power distribution cabinets to obtain the fault location map of the high- and low-voltage power distribution cabinets; A self-healing control module, configured to output a self-healing control instruction for the high- and low-voltage power distribution cabinets based on the fault location map, where the self-healing control instruction includes applying a reverse electromagnetic pulse to the key fault source and adjusting the contact pressure of the potential fault diffusion region.

2. The intelligent detection system for high and low voltage switchgear cabinets according to claim 1, characterized in that, When the data acquisition module performs synchronous acquisition of the electromagnetic radiation spectrum and the cabinet surface vibration waveform of the high- and low-voltage power distribution cabinets to generate a multi-physical-field coupling data set, it is specifically configured to: Arrange electromagnetic sensors in the electromagnetic radiation hot spots of the high- and low-voltage power distribution cabinets, install vibration sensors at the vibration-sensitive points on the cabinet surface of the high- and low-voltage power distribution cabinets, and perform time alignment on the electromagnetic sensors and the vibration sensors; Use the electromagnetic sensors to acquire the electromagnetic radiation spectrum of the high- and low-voltage power distribution cabinets; Use the vibration sensors to acquire the cabinet surface vibration waveform of the high- and low-voltage power distribution cabinets; Align the electromagnetic radiation spectrum and the cabinet surface vibration waveform according to a time window to generate the multi-physical-field coupling data set of the high- and low-voltage power distribution cabinets.

3. The intelligent detection system for high and low voltage switchgear cabinets according to claim 1, characterized in that, When the feature extraction module performs joint time-frequency domain analysis on the multi-physical-field coupling data set to obtain the fault feature vector of the high- and low-voltage power distribution cabinets, it is specifically configured to: Perform frequency domain analysis on the electromagnetic radiation spectrum in the multi-physical-field coupling data set to extract the abnormal frequency band features of energy mutation; Perform time domain phase analysis on the cabinet surface vibration waveform in the multi-physical-field coupling data set to extract the phase mutation features related to time series; Fuse the abnormal frequency band features and the phase mutation features to obtain a fault feature vector characterizing the fault mode.

4. The intelligent detection system for high and low voltage switchgear cabinets according to claim 1, characterized in that, When the network construction and update module constructs and dynamically updates a fault propagation network based on the fault feature vector, it is specifically configured to: Map the abnormal frequency band features in the fault feature vector to the nodes of the fault propagation network, where the weight of each node is determined by the electromagnetic radiation energy density corresponding to the abnormal frequency band features; Generate directed connection edges between the nodes based on the time correlation of the phase mutation features in the fault feature vector, where the direction of the directed connection edges is determined by the time sequence relationship of the occurrence of the phase mutation features; Dynamically update the topological structure of the fault propagation network through an energy transfer probability model.

5. The intelligent detection system for high and low voltage switchgear cabinets according to claim 4, characterized in that, When the network construction and update module executes the dynamic update of the topological structure of the fault propagation network through the energy transfer probability model, it is specifically used for: When it is detected that the difference in electromagnetic radiation energy density between nodes in the fault propagation network exceeds a preset difference threshold, trigger an edge weight attenuation operation, and attenuate the edge weights corresponding to the low-probability propagation paths in the directed connection edges based on the energy transfer probability model; When it is detected that the graph structure similarity between the newly added nodes and the historical topological structure in the fault propagation network is lower than a preset similarity threshold, trigger a topology optimization instruction, dynamically optimize the topological structure according to the optimization instruction, and update the model parameters of the energy transfer probability model; Synchronize the updated edge weights, optimized topological structure, and updated energy transfer probability model to the fault propagation network.

6. The intelligent detection system for high and low voltage power distribution cabinets according to claim 5, characterized in that, When the network construction and update module executes the dynamic optimization of the topological structure according to the optimization instruction and updates the model parameters of the energy transfer probability model, it is specifically used for: Adjust the node connection relationship of the fault propagation network according to the optimization instruction to generate an optimized topological structure, including: when the difference in electromagnetic radiation energy density between two nodes in the fault propagation network is less than a preset merging threshold and the connection edge directions are the same, merge them into one node; according to the high-probability propagation paths in the historical fault case library, add key connection edges between unconnected nodes; Calculate the energy transfer probability between nodes in the optimized topological structure according to the Markov chain transition probability, and iteratively update the model parameters of the energy transfer probability model through the gradient descent algorithm.

7. The intelligent detection system for high and low voltage power distribution cabinets according to claim 4, characterized in that, When the fault location module executes the matching of the betweenness centrality and energy distribution of nodes in the fault propagation network with the fault patterns in the historical fault case library to identify the key fault sources of the high- and low-voltage power distribution cabinets and mark the potential fault diffusion areas, it is specifically used for: Use the Brandes algorithm to calculate the betweenness centrality of each node in the fault propagation network, including: Traverse all nodes in the fault propagation network as source nodes, calculate the shortest paths from the source nodes to other nodes based on breadth-first search, count the number of occurrences of each node in all shortest paths and normalize it to a betweenness value; Extract the energy distribution of the electromagnetic radiation energy density from the node weights of the fault propagation network; Construct a multi-dimensional feature vector of the betweenness centrality and the energy distribution, and calculate the matching degree between the multi-dimensional feature vector and the fault patterns in the historical fault case library based on the cosine similarity algorithm. The historical fault case library stores the mapping relationship between fault patterns and node attributes through an incidence matrix; According to the sequence list of the betweenness centrality and the matching degree, select the nodes with betweenness centrality higher than the preset betweenness centrality threshold and the highest matching degree score as the key fault sources; Traverse the adjacent nodes of the key fault sources, and identify the adjacent nodes with high-probability propagation paths to the key fault sources according to the energy transfer probability model, and mark them as potential fault diffusion areas.

8. The intelligent detection system for high and low voltage switchgear cabinets according to claim 1, characterized in that, When the positioning map generation module executes mapping the key fault source and the potential fault diffusion area to the physical space coordinates of the high-voltage and low-voltage power distribution cabinet to obtain the fault positioning map of the high-voltage and low-voltage power distribution cabinet, it specifically is used for: Extracting the node identifier of the key fault source and the node set of the potential fault diffusion area from the fault propagation network; Invoking the pre-defined sensor layout coordinate mapping table of the high-voltage and low-voltage power distribution cabinet, and mapping the node identifier and the node set to the physical space coordinates of the high-voltage and low-voltage power distribution cabinet based on the sensor layout coordinate mapping table; Generating the fault positioning map of the high-voltage and low-voltage power distribution cabinet based on the physical space coordinates, and the fault positioning map marks the spatial positions of the key fault source and the potential fault diffusion area.

9. The intelligent detection system for high and low voltage switchgear cabinets according to claim 4, characterized in that, The execution of the self-healing control instruction in the self-healing control module includes: According to the coordinates of the key fault source in the fault positioning map, applying a reverse electromagnetic pulse opposite to the propagation direction of the electromagnetic wave with the abnormal frequency band characteristics to the corresponding area of the high-voltage and low-voltage power distribution cabinet; According to the coordinates of the potential fault diffusion area in the fault positioning map, adjusting the contact pressure of the contacts in the potential fault diffusion area to a target pressure value, and the target pressure value is dynamically calculated according to the propagation risk level of the energy transfer probability model.

10. An intelligent detection method for high and low voltage switchgear, characterized in that, The method includes: S1. Synchronously collecting the electromagnetic radiation spectrum and the cabinet surface vibration waveform of the high-voltage and low-voltage power distribution cabinet to generate a multi-physical field coupling data set; S2. Conducting a joint time-frequency domain analysis on the multi-physical field coupling data set to obtain the fault feature vector of the high-voltage and low-voltage power distribution cabinet; S3. Constructing and dynamically updating a fault propagation network based on the fault feature vector; S4. Matching the betweenness centrality and energy distribution of the nodes in the fault propagation network with the fault modes in the historical fault case library to identify the key fault source and mark the potential fault diffusion area; S5. Mapping the key fault source and the potential fault diffusion area to the physical space coordinates of the high-voltage and low-voltage power distribution cabinet to obtain the fault positioning map of the high-voltage and low-voltage power distribution cabinet; S6. Outputting the self-healing control instruction of the high-voltage and low-voltage power distribution cabinet based on the fault positioning map, wherein the self-healing control instruction includes applying a reverse electromagnetic pulse to the key fault source and adjusting the contact pressure of the potential fault diffusion area.

Images