A method and system for power distribution network dispatching

Abstract

This invention discloses a distribution network scheduling method and system, applied in the field of power system technology. The method includes: in response to a fault occurrence signal of a first target distribution network, determining each fault node; determining a topology reconfiguration trigger signal for the first target distribution network based on the number of fault nodes and the time persistence characteristics of the neighborhood connectivity deviation index of all fault nodes; in response to the topology reconfiguration trigger signal, dynamically reconfiguring the first target distribution network using a determined target fault control signal to obtain a second target distribution network; acquiring real-time power data of all second nodes in the second target distribution network, constructing a matching power migration feature vector, determining the target compression parameters of the second target distribution network, processing the real-time power data, and executing a matching scheduling strategy. This invention provides a distribution network scheduling method and system that improves the efficiency of distribution network scheduling when a fault occurs.

Description

Technical Field

[0001] This invention relates to the field of power system technology, and in particular to a distribution network dispatching method and system. Background Technology

[0002] Distribution network dispatching refers to the technical and management process of ensuring the safe, reliable, and economical distribution of electricity to end users by real-time monitoring, analysis, and control of the distribution network's operating status. With the continuous expansion of the distribution network scale and the increasing complexity of the operating environment, efficient dispatching of the distribution network can ensure the stable operation of the system in dynamic changes, improve the power supply capacity and quality of the distribution network, and ensure the reliability and security of power supply.

[0003] However, existing distribution network dispatching methods rely on a few fixed dispatching strategies to handle faults when they occur in the distribution network. These strategies are insufficient to cope with new types of faults, resulting in low efficiency in distribution network dispatching.

[0004] Therefore, improving the efficiency of distribution network dispatch when faults occur has become a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] This invention provides a distribution network dispatching method and system to solve the technical problem that existing distribution network dispatching methods use a few fixed dispatching strategies to handle faults when they occur in the distribution network, which are difficult to cope with new faults, resulting in low efficiency of distribution network dispatching.

[0006] To address the aforementioned technical problems, embodiments of the present invention provide a power distribution network dispatching method.

[0007] In response to a fault occurrence signal in the first target distribution network, a neighborhood connection deviation index for each first node is calculated based on the differences in power fluctuation characteristics between adjacent first nodes within the first target distribution network. The neighborhood connection deviation index reflects the fault degree of each first node. All the neighborhood connection deviation indices are compared and analyzed to determine each fault node.

[0008] Based on the number of faulty nodes and the time persistence characteristics of the neighborhood connection deviation index of all faulty nodes, the topology reconfiguration trigger signal of the first target distribution network is determined.

[0009] In response to the topology reconfiguration trigger signal, a target fault control signal is determined based on the correlation characteristics of historical power data of all first nodes of the first target distribution network and the neighborhood connection deviation index of the fault node; the first target distribution network is dynamically reconfigured using the target fault control signal to obtain a second target distribution network;

[0010] Real-time power data of all second nodes of the second target distribution network is acquired. Based on the spatial and temporal distribution characteristics of the real-time power data, matching power migration feature vectors are constructed. Based on the comparison and analysis results of the power migration feature vectors, the target compression parameters of the second target distribution network are determined.

[0011] The real-time power data is processed using the target compression parameters, and a scheduling strategy matching the processing result is executed.

[0012] As one preferred embodiment, the calculation of the neighborhood connectivity deviation index for each first node based on the differences in power fluctuation characteristics between adjacent first nodes within the first target distribution network includes:

[0013] The fluctuation statistical feature values ​​of all the historical power data of each first node in the first target distribution network are vectorized to obtain the fluctuation feature vector of each first node; wherein, the fluctuation statistical feature values ​​include at least the mean, standard deviation and range;

[0014] Based on the current topology of the first target distribution network, the neighboring nodes of each first node are obtained, and the average dynamic time warping distance between the first fluctuation feature vector of each first node and the second fluctuation feature vector of all the corresponding neighboring nodes is used as the neighborhood connection deviation index of each first node.

[0015] As one preferred embodiment, the topology reconstruction trigger signal includes a local topology reconstruction trigger signal and a global topology reconstruction trigger signal;

[0016] The step of determining the topology reconfiguration trigger signal of the first target distribution network based on the number of faulty nodes and the time persistence characteristics of the neighborhood connection deviation index of all faulty nodes includes:

[0017] The normalized value of the average of the neighborhood connection deviation indices of all the fault nodes in the first target distribution network is used as the node connection imbalance index.

[0018] Record the duration for which the node connection imbalance index exceeds a first preset threshold;

[0019] If the number of faulty nodes is greater than a second preset threshold and the duration is greater than a third preset threshold, a global topology reconfiguration trigger signal matching the first target distribution network is generated; otherwise, a local topology reconfiguration trigger signal matching each faulty node is generated.

[0020] As one preferred embodiment, the target fault control signal includes a fault processing signal and a reconfiguration signal;

[0021] The determination of the target fault control signal based on the correlation characteristics of historical power data of all first nodes in the first target distribution network and the fault node includes:

[0022] The fault handling signal for each fault node is determined based on the neighborhood connection deviation index of each fault node in the first target distribution network.

[0023] The Gran causality test is used to perform causal analysis on all the remaining historical power data of the first node, and a correlation feature matrix is ​​constructed based on the results of the causal analysis.

[0024] A mixed-integer second-order cone programming model is used to generate candidate fault control signals based on the current topology of the first target distribution network and the correlation feature matrix;

[0025] Each candidate fault control signal is simulated, and the target fault control signal is determined based on the simulation results.

[0026] As one preferred embodiment, the step of constructing a matching power migration feature vector based on the spatial and temporal distribution characteristics of the real-time power data includes:

[0027] A sliding window is used to perform time-series analysis on the real-time power data of each second node to determine the power time-series distribution feature vector of each second node; all the power time-series distribution feature vectors of the second target distribution network are processed to obtain the power time-series migration feature vector of the second target distribution network;

[0028] Spatial difference analysis is performed on the real-time power data of all second nodes of the second target distribution network at the same time to obtain the power spatial migration feature vector of the second target distribution network; wherein, the spatial difference analysis reflects the spatial difference characteristics of the real-time power data of all second nodes;

[0029] The power time-series migration feature vector and the power spatial migration feature vector are concatenated using a first vector concatenation process to obtain the power migration feature vector of the second target distribution network.

[0030] As one preferred embodiment, the step of performing time-series analysis on the real-time power data of each second node using a sliding window to determine the power time-series distribution feature vector of each second node includes:

[0031] A first sliding window is used to perform volatility analysis on each of the real-time power data points to obtain the instantaneous volatility feature vector of each second node; a second sliding window is used to perform trend analysis on each of the real-time power data points to obtain the trend change feature vector of each second node; wherein, the length of the second sliding window is greater than the length of the first sliding window;

[0032] The instantaneous fluctuation feature vector and the corresponding trend change feature vector of each second node are concatenated using a second vector concatenation process to obtain the power time series distribution feature vector of each second node.

[0033] As one preferred embodiment, determining the target compression parameters of the second target distribution network based on the comparison and analysis results of the power migration feature vector includes:

[0034] The power migration feature vector and the power migration pattern database are compared and analyzed, and the target compression parameters of the second target distribution network are determined based on the results of the comparison and analysis.

[0035] As one preferred embodiment, the power migration mode database includes at least normal migration mode, fault migration mode, and segmented migration mode.

[0036] As one preferred embodiment, the method further includes:

[0037] The bandwidth of the second target distribution network is adjusted based on the predictive analysis results of the power migration feature vector using a resource reservation mechanism.

[0038] Another embodiment of the present invention provides a power distribution network dispatching system, comprising:

[0039] The fault identification module is used to respond to the fault occurrence signal of the first target distribution network, calculate the neighborhood connection deviation index of each first node based on the difference in power fluctuation characteristics between adjacent first nodes in the first target distribution network, wherein the neighborhood connection deviation index reflects the fault degree of each first node; compare and analyze all the neighborhood connection deviation indices to determine each fault node;

[0040] The reconfiguration analysis module is used to determine the topology reconfiguration trigger signal of the first target distribution network based on the number of faulty nodes and the time persistence characteristics of the neighborhood connection deviation index of all faulty nodes.

[0041] A topology reconfiguration module is used to respond to the topology reconfiguration trigger signal, determine a target fault control signal based on the correlation characteristics of historical power data of all first nodes of the first target distribution network and the neighborhood connection deviation index of the fault node; and dynamically reconfigure the first target distribution network with the target fault control signal to obtain a second target distribution network.

[0042] The compression analysis module is used to acquire real-time power data of all second nodes of the second target distribution network, construct matching power migration feature vectors based on the spatial and temporal distribution characteristics of the real-time power data, and determine the target compression parameters of the second target distribution network based on the comparison and analysis results of the power migration feature vectors.

[0043] The power dispatching module is used to process the real-time power data with the target compression parameters and execute a dispatching strategy that matches the processing result.

[0044] Compared with the prior art, the beneficial effects of the embodiments of the present invention are at least one of the following:

[0045] In response to a fault occurrence signal in the first target distribution network, a neighborhood connection deviation index is calculated for each first node based on the differences in power fluctuation characteristics between adjacent first nodes within the first target distribution network. This neighborhood connection deviation index reflects the fault severity of each first node. Comparative analysis of all neighborhood connection deviation indices is performed to identify each fault node. By utilizing the differences in power fluctuation characteristics between adjacent nodes to identify fault nodes, rather than detecting faults using absolute thresholds, the system can adapt to dynamic changes in grid load and improve the reliability of fault node identification. Based on the number of fault nodes and the temporal persistence characteristics of the neighborhood connection deviation indices of all fault nodes, a topology reconfiguration trigger signal for the first target distribution network is determined. Combining the duration of the fault node and the intensity of the deviation index, transient interference and permanent faults are distinguished. The reconfiguration process is initiated only when both the fault scale and duration meet the requirements, avoiding false triggering of reconfiguration, reducing unnecessary reconfiguration operations, and preventing system oscillations caused by frequent reconfiguration. In response to the topology reconfiguration trigger signal, a target fault control signal is determined based on the correlation characteristics of historical power data of all first nodes in the first target distribution network and the neighborhood connection deviation index of the fault node. The target fault control signal is used to control the first target distribution network. Dynamic reconfiguration yields a second target distribution network, breaking the limitations of fixed topology. Real-time reconfiguration optimizes power flow distribution, improving distribution network scheduling efficiency. By eliminating faulty nodes and reconfiguring only the local network topology associated with normal nodes, fault propagation is avoided, reconfiguration time is shortened, and distribution network scheduling efficiency is improved. Considering that non-faulty nodes consume significant redundant resources, and that the second target distribution network is obtained through network reconfiguration, each second node contains a large amount of information and long data fields. Directly transmitting all data from the distribution network would lead to communication congestion. To improve subsequent scheduling efficiency, real-time power data from all second nodes of the second target distribution network is acquired. Based on the spatial and temporal distribution characteristics of the real-time power data, matching power migration feature vectors are constructed. Based on the comparative analysis of the power migration feature vectors, target compression parameters for the second target distribution network are determined. These compression parameters adaptively match real-time requirements, reduce redundant data transmission, alleviate communication bandwidth pressure, and improve scheduling real-time performance. Real-time power data is processed using the target compression parameters, and a scheduling strategy matching the processing results is executed. Data in non-faulty areas is compressed, and concentrated bandwidth is used to transmit fault handling instructions, further improving distribution network scheduling efficiency. Attached Figure Description

[0046] Figure 1 This is a flowchart of a power distribution network dispatching method in one embodiment of the present invention;

[0047] Figure 2 This is a structural block diagram of a power distribution network dispatching system according to one embodiment of the present invention;

[0048] Figure label:

[0049] 11. Fault Identification Module; 12. Reconfiguration Analysis Module; 13. Topology Reconfiguration Module; 14. Compression Analysis Module; 15. Power Dispatch Module. Detailed Implementation

[0050] 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. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0051] In the description of this application, the terms "first," "second," "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first," "second," "third," etc., may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0052] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections or electrical connections; they can refer to direct connections or indirect connections through an intermediate medium; and they can refer to the internal communication between two components. The terms "vertical," "horizontal," "left," "right," "upper," "lower," and similar expressions used herein are for illustrative purposes only and do not indicate or imply that the device or component referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as limiting the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0053] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is for the purpose of describing specific embodiments only and is not intended to limit the invention. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0054] One embodiment of the present invention provides a distribution network dispatching method; for details, please refer to [link to relevant documentation]. Figures 1-2 , Figure 1 The flowchart shown is a distribution network dispatching method according to one embodiment of the present invention. Figure 2 The diagram shown is a structural block diagram of a power distribution network dispatching system according to one embodiment of the present invention.

[0055] A flowchart of a power distribution network dispatching method in one embodiment of the present invention includes the following steps S1 to S5, as detailed below:

[0056] Step S1: In response to the fault occurrence signal of the first target distribution network, calculate the neighborhood connection deviation index of each first node based on the difference in power fluctuation characteristics between adjacent first nodes in the first target distribution network. The neighborhood connection deviation index reflects the fault degree of each first node. Compare and analyze all neighborhood connection deviation indices to determine each fault node.

[0057] Step S2: Determine the topology reconfiguration trigger signal for the first target distribution network based on the number of faulty nodes and the time persistence characteristics of the neighborhood connection deviation index of all faulty nodes.

[0058] Step S3: In response to the topology reconfiguration trigger signal, determine the target fault control signal based on the correlation characteristics of the historical power data of all first nodes of the first target distribution network and the neighborhood connection deviation index of the fault node; use the target fault control signal to dynamically reconfigure the first target distribution network to obtain the second target distribution network.

[0059] Step S4: Obtain real-time power data of all second nodes of the second target distribution network, construct matching power migration feature vectors based on the spatial and temporal distribution characteristics of the real-time power data, and determine the target compression parameters of the second target distribution network based on the comparison and analysis results of the power migration feature vectors.

[0060] Step S5: Process the real-time power data with the target compression parameters and execute a scheduling strategy that matches the processing result.

[0061] This embodiment provides a distribution network scheduling method that quantifies the degree of fault by calculating the differences in power fluctuations between adjacent nodes. Based on the number of faulty nodes and the time persistence characteristics of the neighborhood connection deviation index of all faulty nodes, a multi-condition triggering mechanism is adopted to avoid equipment wear caused by frequent reconfiguration. Considering that the data of non-faulty nodes occupies a large amount of redundant resources, and that the second target distribution network is obtained by network reconfiguration, each second node contains a lot of information and the data fields are long. Directly transmitting all the data of the distribution network would lead to communication congestion. In order to improve the subsequent scheduling efficiency, the target compression parameters of the second target distribution network are determined based on the power migration feature vector. Different compression parameters are used for nodes of different importance to improve the efficiency of subsequent distribution network scheduling.

[0062] It should be noted that compression parameters include at least the compression ratio and the compression method, and the compression method includes lossless compression and lossy compression.

[0063] In one embodiment, step S1, which calculates the neighborhood connectivity deviation index of each first node based on the differences in power fluctuation characteristics between adjacent first nodes within the first target distribution network, includes:

[0064] The fluctuation statistical feature values ​​of all historical power data of each first node in the first target distribution network are vectorized to obtain the fluctuation feature vector of each first node; wherein the fluctuation statistical feature values ​​include at least the mean, standard deviation and range.

[0065] Based on the current topology of the first target distribution network, the neighboring nodes of each first node are obtained, and the average dynamic time warping distance between the first fluctuation feature vector of each first node and the second fluctuation feature vector of all corresponding neighboring nodes is used as the neighborhood connection deviation index of each first node.

[0066] It should be noted that in a normally operating power distribution network, adjacent nodes are directly connected by lines. Parameters such as voltage, current, and power are affected by the same power source and load, and their fluctuation trends are consistent. Adjacent nodes usually have similar power behaviors. When the fluctuation characteristic vector between a certain node and its adjacent nodes differs more, it indicates that the node is more likely to have a fault.

[0067] This embodiment provides a power distribution network scheduling method that quantifies the fault degree of each first node by calculating the dynamic time warping distance between its second fluctuation feature vector and the second fluctuation feature vectors of all neighboring nodes. The higher the neighborhood connection deviation index, the greater the difference in power behavior between the node and its neighboring nodes, and the higher the fault probability.

[0068] In one embodiment, the topology reconstruction trigger signal in step S2 includes a local topology reconstruction trigger signal and a global topology reconstruction trigger signal.

[0069] In one embodiment, step S2 determines the topology reconfiguration trigger signal of the first target distribution network based on the number of faulty nodes and the time persistence characteristics of the neighborhood connection deviation index of all faulty nodes, including:

[0070] The normalized value of the average neighborhood connection deviation index of all fault nodes in the first target distribution network is used as the node connection imbalance index; the duration of the node connection imbalance index being greater than the first preset threshold is recorded; if the number of fault nodes is greater than the second preset threshold and the duration is greater than the third preset threshold, a global topology reconfiguration trigger signal matching the first target distribution network is generated; otherwise, a local topology reconfiguration trigger signal matching each fault node is generated.

[0071] This embodiment provides a distribution network scheduling method that distinguishes between global topology reconfiguration trigger signals and local topology reconfiguration trigger signals based on the number of faulty nodes and the time persistence characteristics of the neighborhood connection deviation index of all faulty nodes. The global topology reconfiguration trigger signal is triggered only when the risk of fault propagation is high, thereby avoiding unnecessary network-wide reconfiguration and improving the scheduling efficiency of the subsequent distribution network.

[0072] In another embodiment, step S2 determines the topology reconfiguration trigger signal of the first target distribution network based on the number of faulty nodes and the time persistence characteristics of the neighborhood connection deviation index of all faulty nodes, including:

[0073] Record the duration for which the node connection imbalance index of each fault node is greater than the first preset threshold; if the number of fault nodes is greater than the second preset threshold and the duration of all fault nodes is greater than the third preset threshold, then generate a global topology reconfiguration trigger signal that matches the first target distribution network; otherwise, generate a local topology reconfiguration trigger signal that matches each fault node.

[0074] In one embodiment, the global topology reconfiguration trigger signal is designed to activate the backup power supply and divide the islands, while the local topology reconfiguration trigger signal is designed to adjust only the switching state.

[0075] In one embodiment, the target fault control signal in step S3 includes a fault processing signal and a reconstruction signal;

[0076] Based on the correlation characteristics of historical power data of all first nodes in the first target distribution network and the fault nodes, the target fault control signal is determined, including:

[0077] The fault handling signal for each fault node is determined based on the neighborhood connection deviation index of each fault node in the first target distribution network; the Gran causality test is used to perform causal analysis on the historical power data of all remaining first nodes, and the correlation feature matrix is ​​constructed based on the results of the causal analysis.

[0078] A mixed-integer second-order cone programming model is used to generate candidate fault control signals based on the current topology and correlation feature matrix of the first target distribution network. Simulation is performed on each candidate fault control signal, and the target fault control signal is determined based on the simulation results.

[0079] This embodiment provides a distribution network scheduling method that generates targeted fault handling signals for each faulty node based on a neighborhood connectivity deviation index. For example, nodes with a high neighborhood connectivity deviation index may require priority isolation or load transfer to prevent fault propagation. By performing causal analysis on the historical power data of the remaining nodes, the causal relationships between nodes are identified. Candidate fault control signals are generated based on the current distribution network topology. Each candidate control signal is simulated to evaluate its effectiveness under different fault scenarios. For example, the system recovery process after a node failure is simulated, and key indicators such as load recovery rate and voltage stability are recorded. Based on the simulation results, the control signal with the best overall performance is selected as the target fault control signal. This improves the accuracy of distribution network fault control, avoids adverse effects on distribution network scheduling, and thus improves the efficiency of subsequent distribution network scheduling.

[0080] In one embodiment, step S4 involves constructing a matching power migration feature vector based on the spatial and temporal distribution characteristics of real-time power data, including:

[0081] Step S401: Perform time-series analysis on the real-time power data of each second node using a sliding window to determine the power time-series distribution feature vector of each second node; process all power time-series distribution feature vectors of the second target distribution network to obtain the power time-series migration feature vector of the second target distribution network;

[0082] Step S402: Perform spatial difference analysis on the real-time power data of all second nodes of the second target distribution network at the same time to obtain the power spatial migration feature vector of the second target distribution network; wherein, the spatial difference analysis reflects the spatial difference characteristics of the real-time power data of all second nodes;

[0083] Step S403: Perform a first vector concatenation process on the power time-series migration feature vector and the power spatial migration feature vector to obtain the power migration feature vector of the second target distribution network.

[0084] This embodiment provides a power distribution network scheduling method that segments the real-time power data of each second node using a sliding window technique, extracts statistical features within each window, constructs a power time-series distribution feature vector for each node, captures its dynamic changes in the time dimension, and performs a horizontal comparison of the real-time power data of all second nodes at the same time to calculate the differences between nodes, such as Euclidean distance, correlation coefficient, and cluster analysis, to capture their dynamic changes in the spatial dimension. By integrating information from both time and spatial dimensions, a unified feature representation is constructed, breaking through the limitations of traditional single-dimensional analysis. This provides a basis for determining subsequent target compression parameters, adaptively matches real-time requirements, reduces redundant data transmission, alleviates communication bandwidth pressure, and improves the real-time performance of subsequent scheduling.

[0085] In one embodiment, step S401 employs a sliding window to perform time-series analysis on the real-time power data of each second node, determining the power time-series distribution feature vector of each second node, including:

[0086] A first sliding window is used to perform volatility analysis on each real-time power data point to obtain the instantaneous volatility feature vector of each second node; a second sliding window is used to perform trend analysis on each real-time power data point to obtain the trend change feature vector of each second node; wherein, the length of the second sliding window is greater than the length of the first sliding window;

[0087] The instantaneous fluctuation feature vector and the corresponding trend change feature vector of each second node are concatenated to obtain the power time series distribution feature vector of each second node.

[0088] This embodiment provides a power distribution network dispatching method that uses a first sliding window to perform volatility analysis and quantify short-term time-series volatility characteristics, and a second sliding window to perform trend analysis and quantify long-term time-series volatility characteristics. By combining instantaneous volatility and trend change characteristics, the reliability of determining the power time-series distribution feature vector is improved.

[0089] In one embodiment, step S4, based on the comparison and analysis results of the power migration feature vectors, determines the target compression parameters of the second target distribution network, including:

[0090] The power migration feature vector and the power migration pattern database are compared and analyzed, and the target compression parameters of the second target distribution network are determined based on the results of the comparison and analysis.

[0091] It should be noted that the power migration pattern database contains compression parameters corresponding to different combinations of power temporal migration feature vectors and power spatial migration feature vectors.

[0092] In one embodiment, the power migration mode database includes at least normal migration mode, fault migration mode, and segmented migration mode.

[0093] In one embodiment, a resource reservation mechanism is used to adjust the bandwidth of the second target distribution network based on the predictive analysis results of the power migration feature vector.

[0094] The distribution network dispatching method provided in this embodiment of the invention has the following advantages compared to the prior art:

[0095] In response to a fault occurrence signal in the first target distribution network, a neighborhood connection deviation index is calculated for each first node based on the differences in power fluctuation characteristics between adjacent first nodes within the first target distribution network. This neighborhood connection deviation index reflects the fault severity of each first node. Comparative analysis of all neighborhood connection deviation indices is performed to identify each fault node. By utilizing the differences in power fluctuation characteristics between adjacent nodes to identify fault nodes, rather than detecting faults using absolute thresholds, the system can adapt to dynamic changes in grid load and improve the reliability of fault node identification. Based on the number of fault nodes and the temporal persistence characteristics of the neighborhood connection deviation indices of all fault nodes, a topology reconfiguration trigger signal for the first target distribution network is determined. Combining the duration of the fault node and the intensity of the deviation index, transient interference and permanent faults are distinguished. The reconfiguration process is initiated only when both the fault scale and duration meet the requirements, avoiding false triggering of reconfiguration, reducing unnecessary reconfiguration operations, and preventing system oscillations caused by frequent reconfiguration. In response to the topology reconfiguration trigger signal, a target fault control signal is determined based on the correlation characteristics of historical power data of all first nodes in the first target distribution network and the neighborhood connection deviation index of the fault node. The target fault control signal is used to control the first target distribution network. Dynamic reconfiguration yields a second target distribution network, breaking the limitations of fixed topology. Real-time reconfiguration optimizes power flow distribution, improving distribution network scheduling efficiency. By eliminating faulty nodes and reconfiguring only the local network topology associated with normal nodes, fault propagation is avoided, reconfiguration time is shortened, and distribution network scheduling efficiency is improved. Considering that non-faulty nodes consume significant redundant resources, and that the second target distribution network is obtained through network reconfiguration, each second node contains a large amount of information and long data fields. Directly transmitting all data from the distribution network would lead to communication congestion. To improve subsequent scheduling efficiency, real-time power data from all second nodes of the second target distribution network is acquired. Based on the spatial and temporal distribution characteristics of the real-time power data, matching power migration feature vectors are constructed. Based on the comparative analysis of the power migration feature vectors, target compression parameters for the second target distribution network are determined. These compression parameters adaptively match real-time requirements, reduce redundant data transmission, alleviate communication bandwidth pressure, and improve scheduling real-time performance. Real-time power data is processed using the target compression parameters, and a scheduling strategy matching the processing results is executed. Data in non-faulty areas is compressed, and concentrated bandwidth is used to transmit fault handling instructions, further improving distribution network scheduling efficiency.

[0096] Another embodiment of the present invention provides a power distribution network dispatching system, comprising:

[0097] The fault identification module 11 is used to respond to the fault occurrence signal of the first target distribution network, calculate the neighborhood connection deviation index of each first node based on the difference in power fluctuation characteristics between adjacent first nodes in the first target distribution network, wherein the neighborhood connection deviation index reflects the fault degree of each first node; compare and analyze all neighborhood connection deviation indices to determine each fault node;

[0098] The reconfiguration analysis module 12 is used to determine the topology reconfiguration trigger signal of the first target distribution network based on the number of faulty nodes and the time persistence characteristics of the neighborhood connection deviation index of all faulty nodes.

[0099] Topology reconfiguration module 13 is used to respond to topology reconfiguration trigger signal, determine target fault control signal based on the correlation characteristics of historical power data of all first nodes of the first target distribution network and the neighborhood connection deviation index of the fault node; and dynamically reconfigure the first target distribution network with the target fault control signal to obtain the second target distribution network.

[0100] Compression analysis module 14 is used to acquire real-time power data of all second nodes of the second target distribution network, construct matching power migration feature vectors based on the spatial and temporal distribution characteristics of the real-time power data, and determine the target compression parameters of the second target distribution network based on the comparison and analysis results of the power migration feature vectors.

[0101] The power dispatching module 15 is used to process real-time power data with target compression parameters and execute dispatching strategies that match the processing results.

[0102] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A distribution network dispatching method, characterized in that, The method includes: In response to a fault occurrence signal in the first target distribution network, a neighborhood connection deviation index for each first node is calculated based on the differences in power fluctuation characteristics between adjacent first nodes within the first target distribution network. The neighborhood connection deviation index reflects the fault degree of each first node. All the neighborhood connection deviation indices are compared and analyzed to determine each fault node. Based on the number of faulty nodes and the time persistence characteristics of the neighborhood connection deviation index of all faulty nodes, the topology reconfiguration trigger signal of the first target distribution network is determined. In response to the topology reconfiguration trigger signal, a target fault control signal is determined based on the correlation characteristics of historical power data of all first nodes of the first target distribution network and the neighborhood connection deviation index of the fault node; the first target distribution network is dynamically reconfigured using the target fault control signal to obtain a second target distribution network; Real-time power data of all second nodes of the second target distribution network is acquired. Based on the spatial and temporal distribution characteristics of the real-time power data, matching power migration feature vectors are constructed. Based on the comparison and analysis results of the power migration feature vectors, the target compression parameters of the second target distribution network are determined. The real-time power data is processed using the target compression parameters, and a scheduling strategy matching the processing result is executed.

2. The distribution network dispatching method according to claim 1, characterized in that, The neighborhood connectivity deviation index of each first node is calculated based on the differences in power fluctuation characteristics between adjacent first nodes within the first target distribution network, including: The fluctuation statistical feature values ​​of all the historical power data of each first node in the first target distribution network are vectorized to obtain the fluctuation feature vector of each first node; wherein, the fluctuation statistical feature values ​​include at least the mean, standard deviation and range; Based on the current topology of the first target distribution network, the neighboring nodes of each first node are obtained, and the average dynamic time warping distance between the first fluctuation feature vector of each first node and the second fluctuation feature vector of all the corresponding neighboring nodes is used as the neighborhood connection deviation index of each first node.

3. The distribution network dispatching method according to claim 1, characterized in that, The topology reconstruction trigger signal includes a local topology reconstruction trigger signal and a global topology reconstruction trigger signal; The step of determining the topology reconfiguration trigger signal of the first target distribution network based on the number of faulty nodes and the time persistence characteristics of the neighborhood connection deviation index of all faulty nodes includes: The normalized value of the average of the neighborhood connection deviation indices of all the fault nodes in the first target distribution network is used as the node connection imbalance index. Record the duration for which the node connection imbalance index exceeds a first preset threshold; If the number of faulty nodes is greater than a second preset threshold and the duration is greater than a third preset threshold, a global topology reconfiguration trigger signal matching the first target distribution network is generated; otherwise, a local topology reconfiguration trigger signal matching each faulty node is generated.

4. The distribution network dispatching method according to claim 1, characterized in that, The target fault control signal includes a fault handling signal and a reconfiguration signal; The determination of the target fault control signal based on the correlation characteristics of historical power data of all first nodes in the first target distribution network and the fault node includes: The fault handling signal for each fault node is determined based on the neighborhood connection deviation index of each fault node in the first target distribution network. The Gran causality test is used to perform causal analysis on all the remaining historical power data of the first node, and a correlation feature matrix is ​​constructed based on the results of the causal analysis. A mixed-integer second-order cone programming model is used to generate candidate fault control signals based on the current topology of the first target distribution network and the correlation feature matrix; Each candidate fault control signal is simulated, and the target fault control signal is determined based on the simulation results.

5. The distribution network dispatching method according to claim 1, characterized in that, The construction of a matching power migration feature vector based on the spatial and temporal distribution characteristics of the real-time power data includes: A sliding window is used to perform time-series analysis on the real-time power data of each second node to determine the power time-series distribution feature vector of each second node; all the power time-series distribution feature vectors of the second target distribution network are processed to obtain the power time-series migration feature vector of the second target distribution network; Spatial difference analysis is performed on the real-time power data of all second nodes of the second target distribution network at the same time to obtain the power spatial migration feature vector of the second target distribution network; wherein, the spatial difference analysis reflects the spatial difference characteristics of the real-time power data of all second nodes; The power time-series migration feature vector and the power spatial migration feature vector are concatenated using a first vector concatenation process to obtain the power migration feature vector of the second target distribution network.

6. A distribution network dispatching method according to claim 5, characterized in that, The step of performing time-series analysis on the real-time power data of each second node using a sliding window to determine the power time-series distribution feature vector of each second node includes: A first sliding window is used to perform volatility analysis on each of the real-time power data points to obtain the instantaneous volatility feature vector of each second node; a second sliding window is used to perform trend analysis on each of the real-time power data points to obtain the trend change feature vector of each second node; wherein, the length of the second sliding window is greater than the length of the first sliding window; The instantaneous fluctuation feature vector and the corresponding trend change feature vector of each second node are concatenated using a second vector concatenation process to obtain the power time series distribution feature vector of each second node.

7. The distribution network dispatching method according to claim 1, characterized in that, The step of determining the target compression parameters of the second target distribution network based on the comparison and analysis results of the power migration feature vector includes: The power migration feature vector and the power migration pattern database are compared and analyzed, and the target compression parameters of the second target distribution network are determined based on the results of the comparison and analysis.

8. A distribution network dispatching method according to claim 7, characterized in that, The power migration mode database includes at least normal migration mode, fault migration mode, and segmented migration mode.

9. A distribution network dispatching method according to claim 1, characterized in that, The method further includes: The bandwidth of the second target distribution network is adjusted based on the predictive analysis results of the power migration feature vector using a resource reservation mechanism.

10. A power distribution network dispatching system, characterized in that, The system includes: The fault identification module is used to respond to the fault occurrence signal of the first target distribution network, calculate the neighborhood connection deviation index of each first node based on the difference in power fluctuation characteristics between adjacent first nodes in the first target distribution network, wherein the neighborhood connection deviation index reflects the fault degree of each first node; compare and analyze all the neighborhood connection deviation indices to determine each fault node; The reconfiguration analysis module is used to determine the topology reconfiguration trigger signal of the first target distribution network based on the number of faulty nodes and the time persistence characteristics of the neighborhood connection deviation index of all faulty nodes. A topology reconfiguration module is used to respond to the topology reconfiguration trigger signal, determine a target fault control signal based on the correlation characteristics of historical power data of all first nodes of the first target distribution network and the neighborhood connection deviation index of the fault node; and dynamically reconfigure the first target distribution network with the target fault control signal to obtain a second target distribution network. The compression analysis module is used to acquire real-time power data of all second nodes of the second target distribution network, construct matching power migration feature vectors based on the spatial and temporal distribution characteristics of the real-time power data, and determine the target compression parameters of the second target distribution network based on the comparison and analysis results of the power migration feature vectors. The power dispatching module is used to process the real-time power data with the target compression parameters and execute a dispatching strategy that matches the processing result.

Images