Fault analysis method for optical storage and distribution network based on permeability quantification

Abstract

The present application relates to the technical field of power distribution network fault analysis, and particularly relates to a photovoltaic storage power distribution network fault analysis method based on permeability quantification, comprising: S1, obtaining the operation state data of the photovoltaic storage power distribution network in real time, establishing a dynamic equivalent model of the photovoltaic storage system based on the operation state data, and calculating a comprehensive permeability factor reflecting the current photovoltaic storage access degree; S2, when detecting that a fault occurs in the power distribution network, controlling the distributed power source accessed to the power distribution network to inject a preset characteristic signal into the power grid, and obtaining the signal transmission characteristics of the characteristic signal in the propagation process through the signal acquisition devices distributed at each detection point of the power distribution network. The present application realizes real-time online perception and dynamic tracking of the photovoltaic storage access degree by constructing a multi-dimensional permeability quantification index system and fusing it into a comprehensive permeability factor.

Description

Technical Field

[0001] This invention relates to the field of distribution network fault analysis technology, and specifically to a photovoltaic-storage distribution network fault analysis method based on penetration rate quantification. Background Technology

[0002] With the accelerated construction of new power systems, the large-scale integration of photovoltaic (PV) power generation and energy storage systems into distribution networks has become an inevitable trend. Distribution networks are transforming from traditional unidirectional radial passive networks to multi-source, bidirectional interactive active networks. Inverter-type distributed power sources (such as PV inverters and energy storage converters) serve as the core interface for grid connection of PV-storage systems. Their control strategies and operating characteristics have profoundly changed the fault response mechanism of distribution networks. Due to the combined influence of current limiting of power electronic devices and grid connection control strategies, the fault current output by PV-storage power sources during grid faults exhibits "weak feeder" characteristics, characterized by limited amplitude, controlled phase, and a large number of non-power frequency components. This poses a severe challenge to traditional overcurrent protection and distance protection methods based on power frequency fault component amplitude discrimination.

[0003] Existing technologies for fault analysis methods in distribution networks with distributed power sources have the following shortcomings: First, regarding the impact of photovoltaic penetration rate, existing solutions mostly remain at the level of qualitative analysis or offline scenario division, lacking a real-time online quantitative sensing mechanism for the degree of photovoltaic and energy storage integration, and failing to introduce penetration rate as a dynamic variable into the entire fault analysis process. This results in protection settings being unable to adaptively adjust to real-time changes in penetration rate, making it difficult to match the dynamic evolution of system operation. Second, facing the dilemma of severely weakened fault characteristics under high penetration rates, existing solutions are mostly passively adaptive, attempting to extract limited fault information from weakened power frequency electrical quantities, but failing to fundamentally change the problem of missing fault characteristics, and lacking technical means to actively enhance fault characteristics by utilizing the controllability of photovoltaic and energy storage power sources themselves. Summary of the Invention

[0004] The purpose of this invention is to provide a fault analysis method for photovoltaic-storage-distribution networks based on penetration rate quantification, thereby solving the above-mentioned technical problems:

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] A fault analysis method for photovoltaic-storage-distribution networks based on penetration rate quantification includes:

[0007] S1. Real-time acquisition of the operation status data of the photovoltaic-storage distribution network, establishment of a dynamic equivalent model of the photovoltaic-storage system based on the operation status data, and calculation of the comprehensive penetration rate factor reflecting the current degree of photovoltaic-storage access.

[0008] S2. When a fault is detected in the distribution network, the distributed power source connected to the distribution network is controlled to inject a preset characteristic signal into the power grid, and the signal transmission characteristics of the characteristic signal during the propagation process are obtained by the signal acquisition device distributed at each detection point of the distribution network.

[0009] S3. Obtain the power frequency electrical quantity information of the distribution network, input the power frequency electrical quantity information, the signal transmission characteristics and the comprehensive penetration rate factor into the fault diagnosis model, obtain the fault judgment result, and determine the fault location based on the signal transmission characteristics of each detection point.

[0010] S4. Based on the magnitude of the comprehensive penetration rate factor, adaptively adjust the protection settings of the distribution network protection device.

[0011] As a further technical solution, the dynamic equivalent model is a dynamic Thevenin equivalent model, which updates the equivalent impedance and equivalent potential of the system in real time through online parameter identification.

[0012] As a further technical solution, the comprehensive penetration rate factor is obtained by weighted fusion of multi-dimensional penetration rate indicators, which include capacity penetration rate, real-time output penetration rate, and fault contribution penetration rate. Among them, the capacity penetration rate is the ratio of the installed capacity of photovoltaic and energy storage to the maximum load of the distribution network, the real-time output penetration rate is the ratio of the current actual output of photovoltaic and energy storage to the total load of the system, and the fault contribution penetration rate is the ratio of the current injected into the fault point by the photovoltaic and energy storage system during a fault to the total fault current.

[0013] As a further technical solution, the distributed power source includes an energy storage converter and a photovoltaic inverter, and the characteristic signal includes a non-power frequency characteristic identification signal injected by the energy storage converter and a characteristic harmonic signal injected by the photovoltaic inverter.

[0014] As a further technical solution, the frequency of the non-power frequency feature identification signal is selected from the non-power frequency band, and different energy storage units inject feature identification signals of different frequencies or different codes; the amplitude of the feature signal does not exceed a preset proportion of the rated current.

[0015] As a further technical solution, the energy storage converter injects the non-power frequency characteristic identification signal within a first preset time delay after the fault occurs, and the photovoltaic inverter switches to the characteristic injection mode and injects the characteristic harmonic signal within a second preset time delay after the fault detection.

[0016] As a further technical solution, the signal transmission characteristics include amplitude attenuation information and phase shift information of the characteristic signal during propagation, and a signal propagation feature matrix is ​​constructed based on the amplitude attenuation information and phase shift information obtained at each detection point.

[0017] As a further technical solution, the fault diagnosis model is a graph neural network model, which uses the distribution network topology as the graph structure input to realize the propagation and aggregation of fault information between nodes.

[0018] As a further technical solution, the method of determining the fault location based on the signal transmission characteristics of each detection point specifically includes a three-level progressive positioning:

[0019] The first level determines the fault location based on the arrival time and energy direction of the characteristic signals at each detection point;

[0020] The second level involves identifying fault sections within a defined area by combining the power grid topology and load characteristics.

[0021] The third level involves using the propagation delay of characteristic signals to accurately measure distances along critical paths.

[0022] The beneficial effects of this invention are:

[0023] This invention constructs a multi-dimensional penetration rate quantification index system and integrates it into a comprehensive penetration rate factor, enabling real-time online sensing and dynamic tracking of the degree of photovoltaic-storage integration. Based on this, it leverages the rapid response advantage of energy storage converters and the flexible and controllable characteristics of photovoltaic inverters to proactively inject coded non-power frequency characteristic signals into the grid after a fault occurs, fundamentally solving the core problem of traditional protection failure caused by weakened fault characteristics in high penetration scenarios. At the same time, it deeply integrates power frequency electrical quantity information, characteristic signal transmission characteristics, and comprehensive penetration rate factor and inputs them into a graph neural network for collaborative discrimination, overcoming the shortcomings of traditional solutions such as single information utilization and insufficient fault location accuracy. Attached Figure Description

[0024] The invention will now be further described with reference to the accompanying drawings.

[0025] Figure 1 The flowchart is for the photovoltaic-storage distribution network fault analysis method provided by the present invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] Please see Figure 1As shown, the fault analysis method for photovoltaic-storage distribution networks based on penetration rate quantification includes: S1, acquiring real-time operating status data of the photovoltaic-storage distribution network, establishing a dynamic equivalent model of the photovoltaic-storage system based on the operating status data, and calculating a comprehensive penetration rate factor reflecting the current degree of photovoltaic-storage access; S2, when a fault is detected in the distribution network, controlling the distributed power sources connected to the distribution network to inject preset characteristic signals into the grid, and acquiring the signal transmission characteristics of the characteristic signals during propagation through signal acquisition devices distributed at each detection point in the distribution network; S3, acquiring the power frequency electrical quantity information of the distribution network, inputting the power frequency electrical quantity information, signal transmission characteristics, and comprehensive penetration rate factor into the fault diagnosis model to obtain the fault discrimination result, and determining the fault location based on the signal transmission characteristics of each detection point; S4, adaptively adjusting the protection settings of the distribution network protection device according to the magnitude of the comprehensive penetration rate factor. Through the above technical solution, this embodiment provides a fault analysis method for photovoltaic-storage distribution networks based on penetration rate quantification. Its core lies in first establishing a dynamic equivalent model of the photovoltaic-storage system using real-time acquired operational status data and calculating a comprehensive penetration rate factor that integrates multi-dimensional penetration rate indicators, thereby achieving precise quantitative perception of the photovoltaic-storage integration level. When a fault is detected, the distributed power source is actively controlled to inject a preset characteristic signal into the grid and its signal transmission characteristics at each detection point are collected, thus overcoming the defect of weakened fault characteristics under high penetration rates. Then, the power frequency electrical quantity information, signal transmission characteristics, and comprehensive penetration rate factor are jointly input into the fault diagnosis model for collaborative discrimination to obtain the fault discrimination result, and the fault location is determined based on the signal transmission characteristics. Finally, the protection settings are adaptively adjusted according to the magnitude of the comprehensive penetration rate factor, thus forming a complete fault analysis technical solution from real-time penetration rate perception and proactive fault characteristic enhancement to multi-source information intelligent decision-making and dynamic adaptation of protection settings.

[0028] The dynamic equivalent model is a dynamic Thevenin equivalent model, which updates the system's equivalent impedance and equivalent potential in real time through online parameter identification. This embodiment employs a dynamic Thevenin equivalent model and updates the system's equivalent impedance and equivalent potential in real time through online parameter identification technology. The Thevenin equivalent model can represent a complex photovoltaic-storage system as a simplified form of a voltage source series impedance, facilitating rapid calculation of the system's fault current contribution characteristics under different operating conditions. Online parameter identification overcomes the limitation of traditional offline modeling in tracking the system's time-varying characteristics, continuously updating equivalent parameters with changes in illumination, temperature, and photovoltaic-storage output, ensuring that the established dynamic model always remains consistent with the actual system state. This constraint provides reliable basic parameter support for the accurate calculation of the comprehensive permeability factor, enabling subsequent fault characteristic analysis and protection setting adjustments to be based on a real-time and accurate system model.

[0029] The comprehensive penetration rate factor is obtained by weighted fusion of multiple penetration rate indicators, including capacity penetration rate, real-time output penetration rate, and fault contribution penetration rate. Specifically, capacity penetration rate is the ratio of installed photovoltaic (PV) and energy storage (ESS) capacity to the maximum load of the distribution network; real-time output penetration rate is the ratio of the current actual output of PV and ESS to the total system load; and fault contribution penetration rate is the ratio of the current injected into the fault point by the PV and ESS system during a fault to the total fault current. Through the above technical solution, this embodiment elaborates on the comprehensive penetration rate factor, constructing it as a comprehensive quantitative parameter formed by weighted fusion of three dimensions: capacity penetration rate, real-time output penetration rate, and fault contribution penetration rate. Capacity penetration rate is defined as the ratio of installed PV and ESS capacity to the maximum load of the distribution network, used to characterize the static access scale of the PV and ESS system; real-time output penetration rate is defined as the ratio of the current actual output of PV and ESS to the total system load, used to reflect the dynamic fluctuation characteristics of PV and ESS output; and fault contribution penetration rate is defined as the ratio of the current injected into the fault point by the PV and ESS system during a fault to the total fault current, used to measure the actual impact of the PV and ESS system on the fault current during the fault transient process. The penetration rate of the photovoltaic-storage system is characterized by three indicators from the perspectives of static capacity, dynamic operation, and fault response. These indicators are then weighted and fused to form a single comprehensive indicator. This approach avoids the problem of incomplete representation by a single indicator and facilitates direct use by subsequent fault discrimination models and protection setting adjustment modules. This achieves a scientific transformation of penetration rate from a qualitative description to a quantitative parameter.

[0030] Distributed power sources include energy storage converters and photovoltaic inverters. The characteristic signals include non-power frequency characteristic identification signals injected by the energy storage converters and characteristic harmonic signals injected by the photovoltaic inverters. Through the above technical solution, this embodiment defines the specific type of distributed power source and the types of characteristic signals injected in step S2. Distributed power sources include two types of devices: energy storage converters and photovoltaic inverters. Correspondingly, the injected characteristic signals include non-power frequency characteristic identification signals emitted by the energy storage converters and characteristic harmonic signals emitted by the photovoltaic inverters. This distinction is made because energy storage converters have the advantages of direct energy source and fast response speed, making them suitable for injecting non-power frequency characteristic signals in the very short time after a fault occurs, acting as an "active probe." Photovoltaic inverters are more widely distributed and numerous in the distribution network, and their injected characteristic harmonic signals can provide richer spatial distribution information. The two types of signals are designed differently in terms of frequency or coding, which not only makes the signal source identifiable, but also enhances the redundancy and coverage of fault information through multi-point heterogeneous source injection. This enables subsequent fault identification and location to obtain more sufficient and multi-dimensional signal evidence, significantly improving the reliability and accuracy of fault analysis in complex scenarios.

[0031] The frequency of the non-power frequency characteristic identification signal is selected from the non-power frequency band, and different energy storage units inject characteristic identification signals of different frequencies or different codes; the amplitude of the characteristic signal does not exceed a preset proportion of the rated current. Through the above technical solution, this embodiment provides specific limitations on the frequency selection principle, amplitude constraint conditions, and signal differentiation design of different energy storage units for the non-power frequency characteristic identification signal; specifically, the frequency of the non-power frequency characteristic identification signal is selected from the non-power frequency band, the purpose of which is to avoid aliasing with the dominant 50Hz power frequency component in the distribution network, and also to avoid common integer harmonic frequencies such as 100Hz and 150Hz, thereby ensuring the identifiability and anti-interference capability of the characteristic signal in the spectrum; different energy storage units inject characteristic identification signals of different frequencies or different codes, so that each detection point can reverse the signal source based on the received signal characteristics, providing signal source identification information for subsequent fault location; in addition, the amplitude of the characteristic signal is limited to a preset proportion of the rated current, ensuring that the actively injected signal will not have an adverse impact on the normal operation of the distribution network and power quality, reflecting the coordination and unity between actively enhancing fault characteristics and ensuring the normal operation of the system.

[0032] Within a first preset time delay after a fault occurs, the energy storage converter injects a non-power frequency characteristic identification signal, while the photovoltaic inverter switches to the characteristic injection mode and injects a characteristic harmonic signal within a second preset time delay after fault detection. Through the above technical solution, this embodiment provides a timing coordination mechanism for the energy storage converter and photovoltaic inverter to inject characteristic signals after a fault occurs. Specifically, the energy storage converter completes the injection of the non-power frequency characteristic identification signal within the first preset time delay after a fault occurs, while the photovoltaic inverter completes the control mode switching and injects the characteristic harmonic signal within the second preset time delay after fault detection. The second preset time delay is set to be greater than or equal to the first preset time delay. Specifically, since the energy storage converter directly outputs energy using the energy storage medium, its control system does not need to undergo complex calculations such as maximum power point tracking, and its response speed is significantly better than that of the photovoltaic inverter. The purpose of setting it as the priority injection node is to take full advantage of the fast response of the energy storage system and inject the first set of characteristic signals into the grid in the shortest time after the fault occurs, thus gaining a valuable time window for subsequent signal detection. The photovoltaic inverter responds slightly slower, and the signal it injects is a characteristic harmonic, which lags behind the energy storage signal in time but is connected to it. The two types of signals form an orderly injection sequence on the time axis, which not only avoids the spectrum congestion and mutual interference caused by the simultaneous injection of different signal sources, but also ensures that each detection point can clearly distinguish the characteristic signals from different sources, thereby providing additional discriminative information in the time dimension for fault analysis.

[0033] Signal transmission characteristics include amplitude attenuation information and phase shift information of the characteristic signal during propagation. A signal propagation characteristic matrix is ​​constructed based on the amplitude attenuation information and phase shift information obtained at each detection point. Through the above technical solution, this embodiment provides the specific connotation of the characteristic signal transmission characteristics and its subsequent processing method. Specifically, the signal transmission characteristics include the amplitude attenuation information and phase shift information of the characteristic signal during propagation, and a signal propagation characteristic matrix is ​​constructed based on these two types of information obtained from each detection point. Specifically, when the characteristic signal propagates along the line, its amplitude will attenuate exponentially with the increase of propagation distance. At the same time, when passing through the fault point, a significant phase change will occur due to the introduction of fault impedance. These two aspects describe the impact of the fault on the propagation of the characteristic signal from the two physical dimensions of "energy loss" and "phase change". After the signal acquisition device of each detection point extracts these two types of information synchronously, it organizes them into a matrix form according to spatial distribution. The rows of the matrix correspond to the positions of different detection points, and the columns of the matrix correspond to the amplitude attenuation, phase shift, and the relative relationship between each detection point. The signal propagation characteristic matrix constructed in this way completely preserves the spatial position information of the fault point and its disturbance characteristics to signal propagation. It provides a structured and high-information-density data foundation for the input of the subsequent graph neural network model, enabling the fault discrimination model to make a comprehensive judgment based on the signal propagation distribution of the entire network rather than the information of a single detection point, which significantly improves the globality and accuracy of the judgment.

[0034] The fault diagnosis model is a graph neural network model, which uses the distribution network topology as the graph structure input to realize the propagation and aggregation of fault information between nodes. The fault location is determined based on the signal transmission characteristics of each detection point, specifically including a three-level progressive localization: Level 1: Determine the fault area based on the arrival time sequence and energy direction of the characteristic signals from each detection point; Level 2: Within the determined area, combine the power grid topology and load characteristics to determine the fault section; Level 3: For critical lines, accurately measure the distance using the propagation delay of the characteristic signals. Through the above technical solutions, this embodiment provides a fault discrimination method based on a graph neural network model and a fault location method based on a three-level progressive architecture. Specifically, in terms of fault discrimination, the graph neural network model uses the topology of the distribution network as the graph structure input—mapping devices such as buses, lines, and switches as nodes in the graph, and mapping electrical connections as edges in the graph. Through multi-layer graph convolution operations, each node can receive information from its neighboring nodes and aggregate multiple topological and electrical features within its neighborhood layer by layer, thereby achieving deep fusion and collaborative discrimination of fault information across the entire network, overcoming the limitation of traditional point-by-point independent judgment that cannot perceive global topological relationships. In terms of fault location, a regional-level location method is adopted. The system employs a three-tiered, progressive architecture consisting of segment-level positioning and precision-level positioning. Segment-level positioning initially delineates the approximate area of ​​the fault by comparing the timing and energy direction of characteristic signals received at each detection point. Segment-level positioning further determines the specific segment to which the fault belongs based on the signal amplitude attenuation gradient detected at each sectional switch within that area. Precision-level positioning utilizes the propagation delay generated by the reflection of characteristic signals at the fault point for high-precision ranging on critical lines. The three levels converge gradually and progress step by step, improving the positioning accuracy from the traditional line level to the segment level and even the tower level. The synergistic effect of the two levels enables fault identification and positioning to form a complete technical chain from global perception to regional locking and then to precise positioning.

[0035] It should also be noted that step S3 adaptively adjusts the protection settings based on the magnitude of the comprehensive penetration rate factor. Specifically, this includes: when the comprehensive penetration rate factor is lower than the first preset threshold, it indicates that the degree of photovoltaic and energy storage access is low and the impact on fault characteristics is limited. Therefore, the traditional protection settings are kept unchanged to avoid unnecessary setting disturbances. When the comprehensive penetration rate factor is between the first and second preset thresholds, it indicates that the photovoltaic and energy storage access has had a significant impact on fault characteristics. At this time, the setting correction algorithm is activated to make compensatory corrections to the traditional protection settings, so that the protection settings can be appropriately adjusted according to changes in penetration rate. When the comprehensive penetration rate factor is higher than the second preset threshold, it indicates that the system is in a high penetration rate condition and the traditional power frequency electrical quantity protection criteria are no longer reliable. At this time, the system switches to an active protection criterion based on characteristic signals to ensure the sensitivity and reliability of the protection. In particular, whether it is the setting correction under medium permeability conditions or the active protection criterion under high permeability conditions, the determination of the protection setting value adopts a multi-objective optimization algorithm. The algorithm aims to optimize the three core performance indicators of speed, sensitivity and selectivity simultaneously. This enables online optimization and updating of the protection setting value under dynamically changing permeability scenarios, ensuring that the protection device can maintain the best or second-best operating performance in the entire permeability range.

[0036] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A fault analysis method for photovoltaic-storage distribution networks based on penetration rate quantification, characterized in that, include: S1. Real-time acquisition of the operation status data of the photovoltaic-storage distribution network, establishment of a dynamic equivalent model of the photovoltaic-storage system based on the operation status data, and calculation of the comprehensive penetration rate factor reflecting the current degree of photovoltaic-storage access. S2. When a fault is detected in the distribution network, the distributed power source connected to the distribution network is controlled to inject a preset characteristic signal into the power grid, and the signal transmission characteristics of the characteristic signal during the propagation process are obtained by the signal acquisition device distributed at each detection point of the distribution network. S3. Obtain the power frequency electrical quantity information of the distribution network, input the power frequency electrical quantity information, the signal transmission characteristics and the comprehensive penetration rate factor into the fault diagnosis model, obtain the fault judgment result, and determine the fault location based on the signal transmission characteristics of each detection point. S4. Based on the magnitude of the comprehensive penetration rate factor, adaptively adjust the protection settings of the distribution network protection device.

2. The method for fault analysis of photovoltaic-storage-distribution networks based on penetration rate quantification according to claim 1, characterized in that, The dynamic equivalent model is a dynamic Thevenin equivalent model, which updates the equivalent impedance and equivalent potential of the system in real time through online parameter identification.

3. The photovoltaic-storage-distribution network fault analysis method based on penetration rate quantification according to claim 2, characterized in that, The comprehensive penetration rate factor is obtained by weighted fusion of multi-dimensional penetration rate indicators, which include capacity penetration rate, real-time output penetration rate, and fault contribution penetration rate. Among them, the capacity penetration rate is the ratio of the installed capacity of photovoltaic and energy storage to the maximum load of the distribution network, the real-time output penetration rate is the ratio of the current actual output of photovoltaic and energy storage to the total load of the system, and the fault contribution penetration rate is the ratio of the current injected into the fault point by the photovoltaic and energy storage system during a fault to the total fault current.

4. The method for fault analysis of photovoltaic-storage-distribution networks based on penetration rate quantification according to claim 3, characterized in that, The distributed power source includes an energy storage converter and a photovoltaic inverter, and the characteristic signal includes a non-power frequency characteristic identification signal injected by the energy storage converter and a characteristic harmonic signal injected by the photovoltaic inverter.

5. The photovoltaic-storage-distribution network fault analysis method based on penetration rate quantification according to claim 4, characterized in that, The frequency of the non-power frequency feature identification signal is selected from the non-power frequency band, and different energy storage units inject feature identification signals of different frequencies or different codes; the amplitude of the feature signal does not exceed a preset proportion of the rated current.

6. The method for fault analysis of photovoltaic-storage-distribution networks based on penetration quantification according to claim 5, characterized in that, The energy storage converter injects the non-power frequency characteristic identification signal within a first preset time delay after the fault occurs, and the photovoltaic inverter switches to the characteristic injection mode and injects the characteristic harmonic signal within a second preset time delay after the fault is detected.

7. The photovoltaic-storage-distribution network fault analysis method based on penetration rate quantification according to claim 6, characterized in that, The signal transmission characteristics include amplitude attenuation information and phase shift information of the characteristic signal during propagation. A signal propagation feature matrix is ​​constructed based on the amplitude attenuation information and phase shift information obtained at each detection point.

8. The photovoltaic-storage-distribution network fault analysis method based on penetration rate quantification according to claim 7, characterized in that, The fault diagnosis model is a graph neural network model, which uses the distribution network topology as the graph structure input to realize the propagation and aggregation of fault information between nodes.

9. The method for fault analysis of photovoltaic-storage distribution networks based on penetration rate quantification according to claim 8, characterized in that, The method of determining the fault location based on the signal transmission characteristics of each detection point specifically includes a three-level progressive location method: The first level determines the fault location based on the arrival time and energy direction of the characteristic signals at each detection point; The second level involves identifying fault sections within a defined area by combining the power grid topology and load characteristics. The third level involves using the propagation delay of characteristic signals to accurately measure distances along critical paths.

Images