Fusion type self-healing strategy switching method, system, equipment and medium

Abstract

The invention discloses a fusion type self-healing strategy switching method, system and device and a medium, and belongs to the technical field of power system operation control, and the method comprises the steps: obtaining power distribution network data, carrying out the preprocessing, obtaining a real-time operation data set, determining a node connection relation, constructing an edge weight matrix, and generating a topological graph; assigning a corresponding risk level to the topological graph and identifying a congested node set; searching the congestion node set according to the risk level and power flow constraint to obtain a candidate power supply path set; according to the candidate power supply path set, calculating path stability characteristics, performing topology disorder degree evaluation, and screening power supply paths; and based on the power supply path, generating and outputting a self-healing strategy switching instruction, writing the power supply path into the topological graph, and synchronizing the power supply path to the real-time operation data set. According to the invention, the problem of load balancing failure caused by insufficient power supply path generation logic and real-time network matching capability in the prior art is solved, and the secondary fault risk after strategy switching is reduced.

Description

Technical Field

[0001] This invention relates to the field of power system operation control technology, specifically to a fusion-type self-healing strategy switching method, system, device, and medium. Background Technology

[0002] In complex distribution networks with multiple sources including distributed power sources and energy storage systems, self-healing strategies need to dynamically switch power supply paths based on real-time network conditions to achieve rapid power restoration. Existing technologies typically generate fixed power supply paths based on preset topology connections and static load distribution. While such methods can meet basic requirements in scenarios dominated by a single power source or with stable operating conditions, their power supply path generation mechanisms are difficult to effectively adapt to real-time operating conditions in dynamic scenarios with multiple power sources coordinating power supply and frequent network topology adjustments.

[0003] In complex distribution networks with multiple power source points and dynamic coordination, existing self-healing strategies suffer from insufficient dynamic matching capabilities between the power supply path generation logic and real-time network status. This leads to the failure of global load balancing of the power supply recovery path after strategy switching, causing a chain reaction of line overload and limited power restoration range, thus hindering the improvement of power supply reliability. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by this invention is: how to provide a fusion-type self-healing strategy switching method, which constructs a topology graph through real-time running data, identifies congested nodes and classifies risk levels based on outlier analysis, performs a two-layer feasible flow search to generate a set of candidate power supply paths by combining risk constraints and power flow constraints, selects the optimal power supply path by integrating the dynamic stability characteristics of the path and the topology disorder evaluation, and synchronizes the strategy switching command to the real-time running dataset through a closed-loop feedback mechanism, thereby achieving efficient adaptation of power supply paths and real-time operating status in complex distribution networks, improving power supply reliability in multi-source collaborative scenarios, and reducing the risk of secondary faults after strategy switching.

[0006] To address the aforementioned technical problems, this invention provides the following technical solution: a fusion-based self-healing strategy switching method, comprising the following steps,

[0007] The system acquires distribution network data, preprocesses it to obtain a real-time operational dataset, establishes node connectivity relationships and constructs an edge weight matrix, generates a topology graph, assigns corresponding risk levels to the topology graph, and identifies congested node sets. Based on the risk levels and power flow constraints, it searches the congested node sets to obtain a set of candidate power supply paths. Based on the candidate power supply path set, it calculates path stability characteristics, performs topology disorder assessment, and filters power supply paths. Based on the power supply paths, it generates and outputs self-healing strategy switching instructions, writes the power supply paths into the topology graph, and synchronizes them to the real-time operational dataset.

[0008] As a preferred embodiment of the integrated self-healing strategy switching method described in this invention, the following steps are included: acquiring distribution network data, preprocessing it to obtain a real-time operating dataset, and constructing a topology graph based on the real-time operating dataset. This includes: generating a real-time operating dataset by preprocessing the acquired data; determining node connection relationships based on parameters in the real-time operating dataset; calculating a first transmission capacity weight; constructing an edge weight matrix using the first transmission capacity weight and the power output fluctuation rate; and then generating the topology graph by mapping the node connection relationships.

[0009] As a preferred embodiment of the integrated self-healing strategy switching method described in this invention, the method of assigning corresponding risk levels to the topology graph and identifying congested node sets includes: performing outlier analysis using the edge weight matrix and the node connection relationship to identify candidate outlier node sets; and dividing the risk levels and generating congested node sets based on the second transmission capacity weight of the candidate outlier node sets and the output volatility.

[0010] As a preferred embodiment of the integrated self-healing strategy switching method described in this invention, the method involves: searching the congested node set according to the risk level and power flow constraints to obtain a candidate power supply path set, including: filtering a node subset based on the risk level and the second transmission capacity weight; and performing a two-layer feasible flow search on the node subset under the power flow constraints to obtain the candidate power supply path set. The beneficial effect of this preferred technical solution is that by filtering a high-priority node subset based on risk level and transmission capacity weight, and performing a two-layer feasible flow search under power flow constraints, it effectively generates candidate power supply paths that balance line thermal stability and voltage stability, ensuring the adaptability of the power supply path to the real-time network status.

[0011] As a preferred embodiment of the fusion-based self-healing strategy switching method of the present invention, the following steps are included: calculating path stability characteristics and performing topological disorder evaluation based on the candidate power supply path set, including: calculating path stability characteristics based on the node sequence in the power supply path set; and calculating topological disorder using the node connection relationship and the edge weight matrix.

[0012] As a preferred embodiment of the integrated self-healing strategy switching method described in this invention, the power supply path selection includes: selecting a subset of candidate paths that are greater than a preset stability characteristic threshold and less than a preset topological disorder threshold based on the path stability characteristics and the topological disorder degree; and performing multi-level sorting of the candidate path subset according to the path stability characteristics in descending order as the primary sorting rule and the topological disorder degree in ascending order as the secondary sorting rule to generate power supply paths. The beneficial effect of this preferred technical solution is that through selection and multi-level sorting, it can ensure path adaptation and structural rationality, avoid temporary instability and structural fragility caused by path switching, and guarantee continuous power supply.

[0013] As a preferred embodiment of the integrated self-healing strategy switching method described in this invention, the method includes: generating and outputting a self-healing strategy switching instruction based on the power supply path, writing the power supply path into the topology graph, and synchronizing it to the real-time running dataset. This includes: writing the node sequence and sorting result of the power supply path into the strategy execution queue to trigger the strategy switching instruction; writing the node sequence of the power supply path into the edge weight matrix to update the first transmission capacity weight, the risk level, and the topology graph; and synchronizing the updated topology graph to the real-time running dataset, covering the node connection relationships in the real-time running dataset. The beneficial effect of this preferred technical solution is that, through a feedback mechanism, the optimal path is written back to the topology graph and updated, forming a closed-loop decision-making link, ensuring the synchronization of the power supply path with its evolution.

[0014] This invention provides a fusion-type self-healing strategy switching system.

[0015] To address the aforementioned technical problems, the present invention further provides the following technical solution: a fusion-based self-healing strategy switching system, comprising: a data acquisition and preprocessing module, which acquires distribution network data, performs preprocessing to obtain a real-time operating dataset, and constructs a topology graph based on the real-time operating dataset; a congestion node set construction module, which assigns corresponding risk levels to the topology graph and identifies congestion node sets; a candidate power supply path construction module, which searches the congestion node set according to the risk level and power flow constraints to obtain a candidate power supply path set; a power supply path filtering module, which calculates path stability characteristics based on the candidate power supply path set, performs topology disorder evaluation, and filters power supply paths; and a strategy switching and data feedback module, which generates and outputs self-healing strategy switching instructions based on the power supply paths, writes the power supply paths into the topology graph, and synchronizes them to the real-time operating dataset.

[0016] The present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the aforementioned integrated self-healing strategy switching method.

[0017] The present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the steps of the aforementioned integrated self-healing strategy switching method.

[0018] The beneficial effects of this invention are as follows: By constructing a coupling mechanism between the topology graph and the real-time running dataset, the problem of path generation lag caused by the reliance on static topology in traditional methods is solved. Based on node connectivity and the number of edge matrices, combined with outlier analysis, high-risk congested node sets are accurately identified. Under risk level constraints, a candidate path set that balances line thermal stability and voltage stability is generated through a two-layer flow search, achieving deep adaptation between power supply paths and real-time network conditions. By using multi-dimensional evaluation of dynamic stability characteristics and topology disorder to select the optimal path, the risks of transient instability and structural fragility caused by path switching are effectively avoided, significantly improving power supply continuity. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 The above is a flowchart of a fusion-type self-healing strategy switching method provided in one embodiment of the present invention. Detailed Implementation

[0021] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0022] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides a fusion-based self-healing strategy switching method, including:

[0023] S100: Acquire distribution network data, preprocess it to obtain a real-time running dataset, establish node connection relationships and construct edge weight matrix to generate a topology graph.

[0024] S200: Assign corresponding risk levels to the topology graph and identify the set of congested nodes.

[0025] S300: Search the set of congested nodes according to the risk level and power flow constraints to obtain a set of candidate power supply paths.

[0026] S400: Based on the set of candidate power supply paths, calculate the path stability characteristics, perform topology disorder evaluation, and select power supply paths.

[0027] S500: Based on the power supply path, it generates and outputs self-healing strategy switching instructions, writes the power supply path into the topology graph, and synchronizes it to the real-time running dataset.

[0028] It should be noted that existing technologies typically generate fixed power supply paths based on preset topology connections and static load distribution. Such methods can meet basic requirements in scenarios dominated by a single power source or with stable operating conditions. However, in dynamic scenarios with multiple power sources coordinating power supply and frequent adjustments to the network topology, their power supply path generation mechanism is difficult to effectively adapt to real-time operating conditions. Furthermore, in complex distribution networks with multiple power sources dynamically coordinating, existing self-healing strategies suffer from insufficient dynamic matching capabilities between the power supply path generation logic and the real-time network status. This leads to the failure of global load balancing of the power supply recovery path after strategy switching, causing a chain reaction of line overload and limited power restoration range, thus hindering the goal of improving power supply reliability.

[0029] Therefore, to address the issues of power supply path generation mechanism and power supply reliability mentioned above, the steps S100-S500 are as follows: First, a topology graph is constructed using real-time operational data, and congested nodes are identified and risk levels are classified based on outlier analysis. Then, a two-layer feasible flow search is performed to generate a set of candidate power supply paths by combining risk constraints and power flow constraints. Next, the optimal power supply path is selected by fusing path dynamic stability characteristics and topology disorder evaluation. Finally, the strategy switching command is synchronized to the real-time operational dataset through a closed-loop feedback mechanism. This achieves efficient adaptation of power supply paths and real-time operational status in complex distribution networks, improves power supply reliability in multi-source collaborative scenarios, and reduces the risk of secondary faults after strategy switching.

[0030] Example 2, refer to Figure 1 This is the second embodiment of the present invention, which provides a fusion-type self-healing strategy switching method.

[0031] In this embodiment of the invention, step S100 involves acquiring distribution network data, preprocessing it to obtain a real-time running dataset, establishing node connection relationships and constructing an edge weight matrix to generate a topology graph, including steps A1 and A2:

[0032] A1: After preprocessing the collected data, a real-time running dataset is generated. The node connection relationship is determined based on the parameters in the real-time running dataset, and the first transmission capacity weight is calculated.

[0033] A2: Construct an edge weight matrix using the first transmission capacity weight and the output fluctuation rate, and then generate a topology graph by mapping the node connection relationship.

[0034] It should be noted that in step A1, the data collected specifically includes voltage amplitude, current phase, output fluctuation rate, load power measurement value, and power supply output impedance parameters; based on the parameters in the real-time running dataset, namely voltage amplitude, current phase, and power supply output impedance parameters, the node connection relationship is established.

[0035] Specifically, in step A1, the first transmission capacity weight is calculated based on the node connection relationship and the load power measurement value.

[0036] The first transmission capacity weight is the real-time transmission capacity weight of each connection edge in the edge matrix.

[0037] It should be noted that the specific method for determining node connection relationships is to compare the absolute value of the voltage phase difference between two nodes with a preset phase difference threshold to determine whether a physical connection exists. If the absolute value of the voltage phase difference is less than or equal to the preset phase difference threshold and the current direction is consistent with the power transmission direction, it is determined to be a direct connection. The power supply output impedance parameter is used to verify the validity of the connection. If the equivalent impedance value between two nodes is less than the theoretical lower limit of the distribution network line impedance, it is determined to be an invalid connection and excluded. The preset phase difference threshold is determined by the statistical distribution of phase differences of normally connected nodes in historical operating data.

[0038] For example, a connection is considered unconnected when the absolute value of the voltage phase difference exceeds 5 degrees. The results of the electrical connection determination are stored in the form of an adjacency matrix. The row and column indices of the matrix correspond to node identifiers, and the matrix element values ​​are Boolean types, where 1 indicates a valid connection and 0 indicates no connection or an invalid connection.

[0039] It should be noted that in step A2, the node connection relationship determination provides a topological skeleton for the construction of the topology graph. The real-time transmission capacity weight and output fluctuation rate are fused to form an edge weight matrix, which is finally mapped to a weighted directed graph structure to generate the topology graph.

[0040] In an alternative implementation, node connection relationships can also be realized through distribution automation topology identification technology. In this case, determining the connection based on voltage phase difference and impedance can be replaced by using switch status signals and power flow direction, and using the switch position and power flow direction collected by the distribution terminal, combined with the adjacency matrix to determine the node on / off status, thereby reducing phase difference misjudgment caused by harmonics.

[0041] In another alternative implementation, the node connection relationship can also be realized through a graph neural network. In this network, based on parameter rule mapping, a distributed sensor network and topology inference algorithm can be used to infer the connection relationship in reverse using real-time data such as the communication signal strength and phase difference between nodes, thereby reducing the dependence on preset physical parameters.

[0042] In this embodiment of the invention, step S200, which assigns a corresponding risk level to the topology graph and identifies the congested node set, specifically includes steps B1 and B2:

[0043] B1: Utilize the edge weight matrix and node connectivity to perform outlier analysis and identify candidate outlier node sets.

[0044] B2: Based on the second transmission capacity weight and output volatility of the candidate outlier node set, risk levels are divided and a congestion node set is generated.

[0045] Specifically, in step B1, the local network density and betweenness centrality of each node are calculated using the edge weight matrix and node connection relationship. Based on the outlier score of the local network density and betweenness centrality, the set of candidate outlier nodes that exceeds the preset outlier threshold is identified.

[0046] It should be noted that the local network density is calculated by counting the actual number of connecting edges in the subgraph formed by the target node and its neighboring nodes. The actual number of connecting edges is obtained by summing the Boolean values ​​of the corresponding rows and columns in the adjacency matrix. The theoretical maximum possible number of connecting edges is calculated based on the number of neighboring nodes, which is the number of neighboring nodes multiplied by the number of neighboring nodes minus one and then divided by two. The local network density is the ratio of the actual number of connecting edges to the theoretical maximum possible number of connecting edges, and the result is rounded to two decimal places, with a value range of [0.00, 1.00].

[0047] It should be noted that the node betweenness centrality is calculated by taking the target node as the intermediate node and counting the proportion of all shortest paths passing through that node to the total number of global shortest paths. The search range of the shortest path is limited to the local network within three hops. The path weight is calculated based on the real-time transmission capacity weight in the edge weight matrix. The higher the real-time transmission capacity weight value, the lower the path priority.

[0048] It should be noted that the calculation logic of the outlier score is to perform a weighted sum of the local network density per unit value and the node betweenness centrality per unit value for each node. The weight ratio is dynamically adjusted according to the contribution of the two types of indicators to network congestion events in historical operating data.

[0049] Specifically, in step B2, the risk level is determined by calculating the weighted sum of the current value of the real-time transmission capacity weight of the candidate outlier node and the normalized value of the output volatility.

[0050] In one alternative implementation, congestion node identification can also be achieved through voltage sensitivity analysis. By calculating the sensitivity of node voltage to power changes, nodes at risk of voltage exceeding limits can be identified in advance. This avoids the lag caused by relying solely on power flow calculations, based on historical power exceeding limits indicators.

[0051] In another alternative implementation, congestion node identification can also be achieved by introducing a graph convolutional network. The topology graph is used as input, and the power coupling relationship between nodes is learned through the graph convolutional network to directly output the congestion probability, thus ensuring that the self-healing strategy of the distribution network meets the requirements of timeliness and accuracy.

[0052] In this embodiment of the invention, step S300 involves searching the congested node set according to risk level and power flow constraints to obtain a candidate power supply path set, specifically including steps C1 and C2:

[0053] C1: Select a subset of nodes based on risk level and first transmission capacity weight.

[0054] C2: Under power flow constraints, perform a two-layer feasible flow search on a subset of nodes to obtain a set of power supply paths.

[0055] Specifically, step C2 includes steps C21-C23:

[0056] C21: Perform a first-level feasible flow search on a subset of high-priority nodes under power flow constraints to generate an initial power supply path set that satisfies the line thermal stability limit.

[0057] C22: Based on the initial power supply path set, and combining the output fluctuation rate and the node betweenness centrality, perform a second-level feasible flow search to generate an optimized power supply path set.

[0058] C23: Merge the power supply path set with the optimized power supply path set, remove duplicate paths and sort them to generate a candidate power supply path set.

[0059] Specifically, in step C1, only nodes with a risk level of high or medium risk and a first transmission capacity weight exceeding 0.7 are retained. The threshold of 0.7 is determined by analyzing the correlation between the first transmission capacity weight and line overload events in historical operational data.

[0060] Specifically, in step C21, the power flow constraints of the first layer search include line thermal stability limit and node power balance constraints, with the thermal stability limit determined based on the material parameters of the line conductor.

[0061] It should be noted that the specific parameters of the material are the conductor cross-sectional area and conductor properties.

[0062] For example, the copper core cross-sectional area is 50mm².2 The circuit has a thermal stability limiting current of 500 amperes and an aluminum core cross-sectional area of ​​70 mm². 2 The circuit has a maximum current limit of 600 amperes.

[0063] Specifically, in step C22, the voltage stability limit constraint of the second-level search requires that the voltage amplitude deviation of the end node of the path does not exceed ±5% of the rated value.

[0064] It should be noted that in step C23, the rule for determining duplicate paths is that if the node sequences of two paths are completely identical or are in reverse order, they are considered duplicate paths. The criterion for determining inversion is that the node sequences are traversed in reverse order and the first transmission capacity weights in the edge weight matrix are identical.

[0065] In one possible implementation, the first-level power flow constraints can also be addressed by using a constraint optimization algorithm to directly solve a planning problem containing thermal stability constraints and voltage constraints, thereby finding an initial path that satisfies the physical limits and handling the power flow reversal problem caused by a high proportion of distributed power sources.

[0066] In another possible implementation, the removal and sorting of duplicate paths can also be achieved through a graph embedding algorithm, which maps the paths to a low-dimensional vector space, identifies duplicate paths through vector similarity, and then sorts them by constructing a scoring function, which can support the updating of changing sorting rules.

[0067] In this embodiment of the invention, step S400 involves calculating path stability characteristics based on the candidate power supply path set, evaluating topological disorder, and screening power supply paths. Specifically, this includes the following steps D1-D4:

[0068] D1: Calculate the path stability characteristics based on the node sequence in the power supply path set.

[0069] D2: Calculate the topological disorder degree using the node connectivity and edge weight matrix.

[0070] D3: Based on path stability characteristics and topological disorder, select a subset of candidate paths that are greater than the preset stability characteristic threshold and less than the preset topological disorder threshold.

[0071] D4: The candidate path subset is sorted in descending order based on path stability characteristics as the primary sorting rule, and topological disorder as ascending order as the secondary sorting rule to generate power supply paths.

[0072] It should be noted that in step D1, the path stability feature is calculated by calculating the path Lyapunov stability margin. The path Lyapunov stability margin is defined as the weighted sum of the reciprocal of the recovery time and the reciprocal of the oscillation decay time. The weight ratio is dynamically adjusted according to the contribution of voltage and frequency to system stability. The dynamic adjustment rule is to analyze the correlation strength between voltage stability and frequency stability and system collapse events in historical operating data.

[0073] For example, the path Lyapunov stability margin obtains the recovery process data of the voltage amplitude and frequency of the path end node under transient disturbance through time-domain simulation. The recovery process data includes the time it takes for the voltage amplitude to recover from the transient drop value to the rated value and the oscillation decay time for the frequency to return from the offset value to the reference value.

[0074] It should be noted that in step D1, calculating the topological disorder degree is equivalent to calculating the information entropy increase rate. The information entropy increase rate is the weighted sum of the node degree entropy value and the weight entropy value. The weight ratio is set based on the contribution of the topological structure and operating parameters to system instability events in historical data. The information entropy increase rate is calculated by extracting the real-time transmission capacity weight values ​​of the node degree distribution and the edge weight matrix on the path, statistically analyzing their probability distributions, and then calculating the information entropy of the node degree distribution and the information entropy of the edge weight distribution.

[0075] In one alternative implementation, topological disorder can also be evaluated using information entropy theory. Information entropy quantifies the uncertainty of node connectivity and edge weight matrices; a lower entropy value indicates a more ordered topological structure. Calculated based on probability distribution, it reflects the degree of disorder among nodes and edges in the path, ensuring the capture of implicit topological disorder characteristics.

[0076] In another alternative implementation, the multi-level ranking technique can also use fuzzy hierarchical analysis to construct a fuzzy judgment matrix, transform the path stability characteristics and topological disorder into fuzzy weights, and then calculate the comprehensive score for ranking by weighting, thereby improving the robustness and objectivity of path ranking.

[0077] In this embodiment of the invention, step S500 involves generating and outputting a self-healing strategy switching instruction based on the power supply path, writing the power supply path into the topology graph, and synchronizing it to the real-time running dataset. Specifically, this includes the following steps E1-E5:

[0078] E1: Generate a self-healing strategy switching instruction based on the node sequence and sorting results of the optimal power supply path.

[0079] E2: Write the node sequence of the power supply path into the edge weight matrix of the topology graph, and update the first transmission capacity weight and node risk level label of the associated edge.

[0080] E3: Synchronize the updated topology graph to the real-time running dataset, overwriting the topological connections and edge weights in the original dataset.

[0081] E4: Triggers the update process from steps S200 to S300, re-topologically mapping based on the synchronized real-time running dataset and performing outlier analysis.

[0082] E5: Verify the consistency between the topology diagram and the real-time running dataset before and after the policy switch. If the verification fails, roll back to the previous version and trigger an alarm.

[0083] For example, in step E1, the self-healing strategy switching instruction includes a path identifier and a version number. The path identifier is generated by concatenating the identifiers of the start node and the end node of the path according to a preset format. For example, when the start node identifier is N001 and the end node is N005, the path identifier is N001-N005. The version number is generated by combining the strategy generation timestamp and the event type code.

[0084] It should be noted that in step E2, the node sequence is written to the real-time running dataset. By locking the write permission of the real-time running dataset, the adjacency matrix of the topology graph is overwritten item by item according to the row and column indices to cover the topological connection relationships in the original dataset, and the edge weight matrix is ​​overwritten according to the same index to cover the edge weight values ​​in the original dataset.

[0085] Example 3, referring to Figure 1 This is the third embodiment of the present invention, which provides a fusion-type self-healing strategy switching system, including: a data acquisition and preprocessing module, which acquires distribution network data, preprocesses it to obtain a real-time operating dataset, and constructs a topology graph based on the real-time operating dataset; a congested node set construction module, which assigns corresponding risk levels to the topology graph and identifies congested node sets; a candidate power supply path construction module, which searches the congested node set according to the risk level and power flow constraints to obtain a candidate power supply path set; a power supply path screening module, which calculates path stability characteristics based on the candidate power supply path set, performs topology disorder evaluation, and screens power supply paths; and a strategy switching and data feedback module, which generates and outputs self-healing strategy switching instructions based on the power supply paths, writes the power supply paths into the topology graph, and synchronizes them to the real-time operating dataset.

[0086] Example 4, the fourth embodiment of the present invention, differs from the previous three embodiments in that: if the function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0087] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0088] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0089] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0090] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for switching integrated self-healing strategies, characterized in that: include, The system acquires power distribution network data, performs preprocessing to obtain a real-time operational dataset, establishes node connection relationships and constructs an edge weight matrix, and generates a topology graph. Assign corresponding risk levels to the topology graph and identify the set of congested nodes; Based on the risk level and power flow constraints, a search is performed on the set of congested nodes to obtain a set of candidate power supply paths; Based on the candidate power supply path set, the path stability characteristics are calculated, and the topology disorder is evaluated to screen the power supply paths. Based on the power supply path, a self-healing strategy switching instruction is generated and output, the power supply path is written into the topology graph, and synchronized to the real-time running dataset.

2. The fusion-based self-healing strategy switching method as described in claim 1, characterized in that, Acquire distribution network data, preprocess it to obtain a real-time operational dataset, and construct a topology map based on the real-time operational dataset, including: After collecting and preprocessing the data, a real-time running dataset is generated. The node connection relationship is determined based on the parameters in the real-time running dataset, and the first transmission capacity weight is calculated. Using the first transmission capacity weight and output fluctuation rate, an edge weight matrix is ​​constructed, and then the topology graph is generated by mapping the node connection relationship.

3. The fusion-based self-healing strategy switching method as described in claim 2, characterized in that, Assigning corresponding risk levels to the topology graph and identifying the set of congested nodes includes: Using the edge weight matrix and the node connection relationship, outlier analysis is performed to identify candidate outlier node sets; Based on the second transmission capacity weight of the candidate outlier node set and the output volatility, risk levels are classified and a congestion node set is generated.

4. The fusion-based self-healing strategy switching method as described in claim 3, characterized in that, Based on the risk level and power flow constraints, a search is performed on the congested node set to obtain a set of candidate power supply paths, including: Based on the risk level and the first transmission capacity weight, a subset of nodes is selected; Under the power flow constraint, a two-layer feasible flow search is performed on the node subset to obtain a set of candidate power supply paths.

5. The fusion-based self-healing strategy switching method as described in claim 4, characterized in that, Based on the candidate power supply path set, path stability characteristics are calculated, and topological disorder is evaluated, including: Calculate the path stability characteristics based on the node sequence in the power supply path set; The topological disorder degree is calculated using the node connection relationships and the edge weight matrix.

6. The fusion-based self-healing strategy switching method as described in claim 5, characterized in that, Filter power supply paths, including: Based on the path stability characteristics and the topological disorder, a subset of candidate paths is selected that are greater than a preset stability characteristic threshold and less than a preset topological disorder threshold. The candidate path subset is sorted in multiple levels according to the path stability characteristics in descending order as the primary sorting rule and the topological disorder degree in ascending order as the secondary sorting rule to generate power supply paths.

7. The fusion-based self-healing strategy switching method as described in claim 6, characterized in that, Based on the power supply path, a self-healing strategy switching instruction is generated and output, and the power supply path is written into the topology graph and synchronized to the real-time running dataset, including: Write the node sequence and sorting result of the power supply path into the policy execution queue to trigger the policy switching instruction; Write the node sequence of the power supply path into the edge weight matrix, and update the first transmission capacity weight, the risk level, and the topology graph; The updated topology graph is synchronized to the real-time running dataset, overriding the node connection relationships in the real-time running dataset.

8. A fusion-based self-healing strategy switching system, employing a fusion-based self-healing strategy switching method as described in any one of claims 1 to 7, characterized in that, include: The data acquisition and preprocessing module acquires power distribution network data, performs preprocessing to obtain a real-time operating dataset, and constructs a topology map based on the real-time operating dataset. The congestion node set construction module assigns corresponding risk levels to the topology graph and identifies congestion node sets; The candidate power supply path construction module searches the set of congested nodes according to the risk level and power flow constraints to obtain a set of candidate power supply paths. The power supply path screening module calculates path stability characteristics and performs topological disorder evaluation based on the candidate power supply path set to screen power supply paths. The strategy switching and data feedback module generates and outputs a self-healing strategy switching instruction based on the power supply path, writes the power supply path into the topology map, and synchronizes it to the real-time running dataset.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the fusion self-healing strategy switching method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the integrated self-healing strategy switching method according to any one of claims 1 to 7.

Images