A method and system for live-line work of a cable-type power distribution network

Abstract

The present application relates to the field of power system distribution network automation and operation and maintenance, and particularly relates to a cable type distribution network non-power-off operation method and system. The present application constructs a temporary bypass system key parameter model, isolates the operation area from the running power grid at the physical level, and ensures that the user continuously obtains power supply during the entire operation process. By introducing a comprehensive cost function including geometric length, specific attributes, risk and other multi-dimensions, the definition of path advantages and disadvantages can be more accurately matched to the actual application scenario. At the same time, a pre-screening strategy is added in the expansion process, and low-value nodes in the Open list are dynamically filtered according to the total cost, greatly reducing the number of nodes that need to be expanded. Experiments show that in a large-scale network scenario, this strategy can reduce the time complexity of the algorithm by 30%-50%, effectively alleviating the performance bottleneck of the traditional A algorithm under large data volume, and enabling the algorithm to output path planning results within a reasonable time.

Description

Technical Field

[0001] This invention relates to the field of power system distribution network automation and operation and maintenance technology, and in particular to a method and system for uninterrupted power operation in cable distribution networks. Background Technology

[0002] In existing urban power grids, cable-type distribution networks are widely used. When it is necessary to inspect, replace, or troubleshoot a section of line, ring main unit, or transformer, the traditional method is usually to directly disconnect the power supply, forcing downstream users to experience power outages. This not only affects power supply reliability indicators (such as RS-3), but may also cause huge economic losses to important users such as hospitals, data centers, and precision manufacturing industries.

[0003] Existing technologies suffer from low decision-making efficiency and poor implementation results when determining live-line working schemes. Furthermore, while existing technologies utilize path search algorithms to determine live-line routes, these algorithms often focus solely on geometric distance when applied to scenarios such as cable laying and logistics distribution in complex environments. For example, in cable laying, only the shortest straight-line distance between two points is considered, neglecting differences in material costs across different cable sections, impedance variations during power transmission, and the impact of high impedance due to differences in cable materials and layout. High impedance leads to additional power loss and voltage drop, as well as environmental risks. This single-factor approach fails to meet the comprehensive requirements of safety, economy, and reliability in practical engineering projects. Meanwhile, although the traditional A algorithm introduces heuristic functions to guide the search direction, in complex and large-scale path networks, if the heuristic function is designed in a single way, it is easy to have blind search. On the one hand, it may over-expand a large number of irrelevant nodes, causing the algorithm's time complexity to soar and wasting computing resources. On the other hand, because it does not fully combine the definition of path quality by factors such as impedance and risk in actual engineering, the so-called shortest path found is not truly optimal in actual engineering scenarios. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method and system for uninterrupted power supply operation in cable-type distribution networks, which solves the problems existing in the prior art.

[0005] This invention provides a method for uninterrupted power supply operation in cable-type distribution networks, the method comprising the following steps: S1: Collect multi-source heterogeneous data of cable-type distribution networks and perform data preprocessing operations on the multi-source heterogeneous data; S2: Establish the topology logic model of the cable-type distribution network; S3: Establish the operation influence domain and key parameter models of the temporary bypass system based on the aforementioned distribution network topology logic model; The key parameter model of the temporary bypass system includes the bypass path model, capacity requirement model, and equipment selection model. The bypass path model includes a subgraph extraction module, a path search module, and an optimal path output module. The subgraph extraction module extracts a path from the distribution network topology logic model using a graph neural network model, containing only nodes within the influence domain D and the determined power supply connection point. The determined load lead-out point The relevant key nodes are connected subgraphs G′; the path search module evaluates each edge in the connected subgraph G′ using an improved A algorithm to obtain a recommended temporary bypass cable laying path; the optimal path output module outputs the recommended temporary bypass cable laying path. The improved A algorithm is used to evaluate each edge in the connected subgraph G′, and the recommended temporary bypass cable laying path is as follows: Sa: Add risk factor parameters to each edge in the connected subgraph G′; Sb: Define and initialize the actual cumulative cost function, heuristic function value, and total cost function of the node; Sc: Initialize the list and set the node affiliation; Sd: In the main loop search phase, expand the node and update the cost, and select the optimal node and determine the termination; Se: In the path backtracking phase, construct the reverse path to obtain the recommended temporary bypass cable laying path; S4: Perform uninterrupted power supply work based on the calculation results of the key parameter model of the temporary bypass system.

[0006] Preferably, in Sb, let g(n) represent the actual cumulative cost function from the starting node star to node n. The initialization rule for the actual cumulative cost function of the node is as follows: In the initial stage of the improved A algorithm, g(n) of all other nodes in the connected subgraph G′ except the starting node start is initialized to infinity; the starting node start is the starting point of the path, and its actual cumulative cost to itself is naturally 0, that is, g(start) = 0. set up The optimized heuristic function value, wherein the heuristic function value The calculation formula is: ; In the formula, This represents the geometric distance from node n to the target node goal; This represents the average impedance of the edges in the entire path network of the connected subgraph G′. Let α, β, and γ be the average risk factor of the edges in the entire path network of the connected subgraph G′, where α, β, and γ are the weight coefficients of the three factors of distance, impedance, and risk, respectively, and satisfy α+β+γ=1, and v is the velocity constant. The heuristic function value The initialization rule is as follows: for each node n in the connected subgraph G′, its initial heuristic function value must be calculated according to the above formula. ; Let the total cost function f(n) be the total cost of node n, obtained by adding the actual cumulative cost parameter of node n and the heuristic function value. That is, the expression of the total cost function is: ; The initialization rule for the total cost function is: combining g(n) and The initialization result, the total cost function of the starting node start is: For the other nodes of the connected subgraph G′, the total cost function is: .

[0007] Preferably, in the Sc, the list initialization includes the initialization of an open list (Open) and a closed list (Closed). The open list (Open) is used to store nodes to be expanded in the improved A algorithm. During the initialization phase, the starting node (start) is added to the open list, at which point the open list contains only this one node. The closed list (Closed) is used to store nodes in the improved A algorithm that have already undergone expansion, meaning that all adjacent nodes of these nodes have been explored, and the node will not be expanded again. During the initialization phase, the closed list is empty. The node attribution setting is specifically as follows: after the initialization of the open list and the closed list is completed, the attribution of the starting node (start) needs to be confirmed again: that is, the starting node is only in the open list and not in the closed list.

[0008] Preferably, in Sd, the node with the smallest total cost function f(n) in the open list Open is selected as the current node current, and it is determined whether the current node current is the goal node: if yes, proceed to the next traversal operation; if no, move the current node current from the open list Open to the closed list Closed. Iterate through the set of all neighboring nodes of the current node (current), and calculate the new actual cumulative cost function for each neighboring node. The calculation formula is as follows: ; In the formula, The comprehensive cost function is calculated using the following formula: ; In the formula, e represents the edge from the current node to an adjacent node; in the formula, This represents the geometric distance from the current node to a certain adjacent node; This represents the average impedance of the edge formed by all adjacent nodes of the current node. Let α be the average risk factor of the edges formed by all adjacent nodes of the current node, where α, β, and γ are the weight coefficients of the three factors of distance, impedance, and risk, respectively. If the corresponding adjacent node is not in the open list, add it to the open list as a new node to be expanded, and then set... Calculate the heuristic function values ​​from the adjacent nodes to the target node as described above. Calculate the total cost function of the corresponding adjacent nodes; and record the parent node of the corresponding adjacent node as node current, providing parent-child relationship links for subsequent path backtracking; If the corresponding neighboring nodes are in the open list, and the new actual cumulative cost function of the corresponding neighboring nodes is better; update the relevant cost information and parent node pointers of the neighboring nodes, specifically: update This ensures that the latest minimum actual cost to the neighbor node is recorded; the total cost function is recalculated and updated synchronously; the parent node of the corresponding adjacent node is reset to the current node to ensure that the path backtracking can point to a better parent node link.

[0009] Preferably, after traversing all adjacent nodes of the current node, a pre-filtering operation is performed on the nodes in the open list Open. The specific rule of the pre-filtering operation is: set a preset threshold, filter and discard nodes in the open list Open whose total cost function value exceeds the preset threshold, and do not perform further expansion processing on them.

[0010] Preferably, the preset threshold is defined as the total cost function value exceeding 120% of the cost of the currently found optimal path, based on the cost of the optimal path already found.

[0011] Preferably, the capacity demand model is used to calculate the minimum current-carrying capacity that the recommended temporary bypass cable laying path must have; the inputs of the capacity demand model include the historical or predicted maximum load current of all load nodes within the operation influence domain D, the impedance parameters of each path in the distribution network topology logic model G, and the set safety margin coefficient.

[0012] Preferably, the expression for the capacity demand model is: ; In the formula, Minimum current-carrying capacity requirements for temporary bypass cable laying paths. The actual current value flowing through the temporary bypass cable laying path. The historical or predicted maximum load current for all load nodes within the operation's influence domain D.

[0013] Preferably, the equipment selection model matches and generates a complete list of temporary bypass system equipment and its connection topology that can be directly purchased and deployed from a standardized equipment library based on the calculation results of the bypass path model and the capacity requirement model; the input of the equipment selection model is the recommended temporary bypass cable laying path, the minimum current carrying capacity requirement of the temporary bypass cable laying path, the operating voltage level, and the standardized equipment database, which contains various types of bypass switches, cables, connectors, mobile transformers and their technical parameters; The equipment selection model includes a bypass cable selection submodule, a bypass switch car selection submodule, a connection and protection equipment selection submodule, and a topology model generation submodule.

[0014] According to another aspect of the present invention, a live-line working system for a cable-type distribution network is provided, the system employing the above-described live-line working method for a cable-type distribution network, the system comprising: The data acquisition and preprocessing module is used to acquire multi-source heterogeneous data from cable-type distribution networks and perform data preprocessing operations on the multi-source heterogeneous data. The distribution network topology logic model building module is used to build the distribution network topology logic model of cable-type distribution networks. The module for establishing the key parameter model of the operation impact domain and temporary bypass system is used to establish the key parameter model of the operation impact domain and temporary bypass system based on the distribution network topology logic model. The execution module is used to perform uninterrupted power supply operations based on the calculation results of the key parameter model of the temporary bypass system.

[0015] The embodiments of the present invention have the following technical effects: This embodiment constructs a key parameter model of the temporary bypass system, which physically isolates the work area from the operating power grid, ensuring that users receive a continuous power supply throughout the entire operation. This embodiment introduces graph theory, optimization algorithms, and other methods to establish an accurate mathematical model, making key aspects such as bypass path selection and capacity calculation more scientific and objective, reducing reliance on human experience, and improving decision-making efficiency and the optimality of the solution.

[0016] This embodiment introduces a comprehensive cost function that incorporates multiple dimensions, including geometric length, specific attributes, and risk, enabling a more precise alignment with the definition of path quality in real-world application scenarios. Different scenarios can be flexibly adapted to business needs simply by adjusting the weight coefficients. Simultaneously, a pre-screening strategy is incorporated during the expansion process, dynamically filtering low-value nodes from the Open list based on the total cost, significantly reducing the number of nodes requiring expansion. Experiments demonstrate that in large-scale network scenarios, this strategy can reduce the algorithm's time complexity by 30%-50%, effectively alleviating the performance bottleneck of the traditional A algorithm under large data volumes, allowing the algorithm to output path planning results within a reasonable timeframe.

[0017] By optimizing the heuristic function to incorporate average information from multiple dimensions such as distance, impedance, and risk, the heuristic guidance is no longer limited to geometric distance but better aligns with the comprehensive consideration of multiple factors required in practical engineering. During the search process, the improved A algorithm can select expansion nodes based on a more comprehensive estimate of the remaining path cost, significantly improving the accuracy of the search direction and reducing ineffective expansion of irrelevant nodes, thus finding the optimal path more efficiently across multiple dimensions. The actual cumulative cost function is initialized to reflect the multi-dimensional cumulative cost from the starting point to the current node, and the total cost function is defined as the sum of the actual cumulative cost and the heuristically estimated remaining cost, allowing the node cost parameters to comprehensively reflect the actual value of the path in terms of length, impedance, and risk. This multi-dimensional cost evaluation method allows the improved A algorithm to consider not only how far the path has been traveled but also the quality of the path taken, thereby selecting the optimal path that truly meets engineering requirements. Attached Figure Description

[0018] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is a flowchart of a method for uninterrupted power supply operation in a cable-type distribution network provided by an embodiment of the present invention; Figure 2This is a flowchart provided by an embodiment of the present invention, which uses an improved A algorithm to evaluate each edge in the connected subgraph to obtain a recommended temporary bypass cable laying path. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0021] Example 1, Figure 1 A flowchart of a live-line working method for cable-type distribution networks is shown, such as... Figure 1 As shown, a method for uninterrupted power supply operation in a cable-type distribution network includes the following steps: S1: Collect multi-source heterogeneous data of cable-type distribution networks and perform data preprocessing operations on the multi-source heterogeneous data; The multi-source heterogeneous data includes GIS geographic information system data, SCADA data, and materials management data.

[0022] The GIS data includes cable trench routes, ring main unit / transformer location coordinates, road red lines, building outlines, etc.; the SCADA data includes equipment ledgers, historical fault records, real-time / historical load data, etc.; the material management data includes the inventory status, technical parameters, and physical storage location of existing bypass operation equipment, etc.

[0023] The data preprocessing operations include outlier removal, unit unification, and coordinate calibration. Outlier removal involves eliminating obviously erroneous data from the multi-source heterogeneous data. Unit unification involves standardizing all electrical parameters to the International System of Units (SI), for example, standardizing cable cross-sectional area to mm². 2 The length is standardized to m; the coordinate calibration involves converting the GIS geographic information system data into UTM coordinates to ensure the geometric accuracy of distance calculation.

[0024] S2: Establish the topology logic model of the cable-type distribution network; The physical equipment of the cable-type distribution network is abstracted as nodes and edges of the distribution network topology logic model to construct a directed graph model.

[0025] The node set of the power distribution network topology logic model includes power supply nodes, facility nodes, and virtual nodes. In this embodiment, the power supply node is the 10kV outgoing line bay of the substation, the facility nodes include ring main units (RMU), prefabricated substations, and cable branch boxes, and the virtual nodes are virtual nodes set up at the incoming and outgoing ends of the ring main units respectively.

[0026] The edge set of the distribution network topology logic model includes physical edges and logical edges. The physical edges are the actual cable segments connecting two nodes. Each edge e needs to record its first and last nodes and its length L. e Resistance R e Reactance X e The logic edge identifies the switch status. For circuit breakers and load switches, a "closed / open" status variable is established. In the initial state, all switches of unmaintained lines are closed by default.

[0027] S3: Establish the operation influence domain and key parameter models of the temporary bypass system based on the aforementioned distribution network topology logic model; In this step, the operation influence domain D is the influence domain determined in the distribution network topology logic model G with the target operation point as the center. The influence domain includes all critical load nodes and lines that may be subject to voltage fluctuations, power flow changes, or power outage risks due to the operation; it is obtained through SCADA data and GIS geographic information system data in the multi-source heterogeneous data.

[0028] The key parameter model of the temporary bypass system includes the bypass path model, capacity requirement model, and equipment selection model.

[0029] The bypass path model is used to find an optimal physical path in the distribution network topology logic model for laying temporary bypass cables to achieve seamless load transfer. The input data of the bypass path model includes the distribution network topology logic model G, the operation influence domain D, and the determined power access point. and the determined load lead-out point The determined power access point is a reliable power source upstream of the work site, and the determined load output point is a busbar or main line downstream of the work site.

[0030] The bypass path model includes a subgraph extraction module, a path search module, and an optimal path output module. The subgraph extraction module extracts a path from the distribution network topology logic model using a graph neural network model, containing only nodes within the influence domain D and the determined power supply connection point. The determined load lead-out point The relevant key nodes are connected subgraphs G′, which can effectively reduce the search space and improve computational efficiency. The connected subgraph G′ has a clear start node and a goal node. The path calculation module evaluates each edge in the connected subgraph G′ using an improved A algorithm to obtain a recommended temporary bypass cable laying path. The optimal path output module is used to output the recommended temporary bypass cable laying path.

[0031] When traditional path search algorithms are applied to scenarios such as cable laying and logistics distribution in complex environments, traditional shortest path algorithms mostly focus on the single factor of geometric distance to evaluate the quality of a path. For example, in cable laying, only the shortest straight-line distance between two points is considered, while ignoring the differences in material costs when laying cables in different sections, the impedance effects during power transmission, and the fact that different sections have varying impedances due to differences in cable materials and layouts. High impedance can lead to additional power loss and voltage drop, as well as environmental risks in the section. This single-factor approach cannot meet the comprehensive requirements of safety, economy, and reliability in actual engineering projects. Meanwhile, although the traditional A algorithm introduces heuristic functions to guide the search direction, in complex and large-scale path networks, if the heuristic function design is simple, it is easy to have blind search. On the one hand, it may over-expand a large number of irrelevant nodes, causing the algorithm's time complexity to soar and wasting computational resources. On the other hand, because it does not fully combine the definition of path quality based on factors such as impedance and risk in actual engineering, the so-called shortest path found is not truly optimal in actual engineering scenarios. Based on this, this embodiment provides an improved A algorithm to evaluate each edge in the connected subgraph G′ to obtain the recommended temporary bypass cable laying path.

[0032] Specifically, the improved A algorithm is used to evaluate each edge in the connected subgraph G′ to obtain the recommended temporary bypass cable laying path as follows: Sa: Add a risk factor parameter to each edge in the connected subgraph G′; In this step, to reduce the complexity of the power distribution network topology logic model, a risk factor parameter is added to each edge in the connected subgraph G′. The risk of the edge is a dimensionless coefficient used to quantify various risks brought about by the environmental attributes of the edge, such as construction risk, operation and maintenance risk, and natural disaster risk. Its value is assigned according to the specific environmental characteristics of the edge, and the value range can usually be set to [0,1]. The assignment process of the risk factor of the edge is as follows: if the edge is located above a busy road with frequent vehicle traffic and there are construction and operation and maintenance risks such as collisions and scratches, the traffic environment risk factor of such edge is set to a higher value, which is 0.7; if the edge is located below a green belt, next to a sidewalk, or in an area with low traffic flow, the traffic environment risk factor of the edge is lower, which is 0.1, otherwise it is 0.2. In this embodiment, the busy road is an urban main road, highway, etc.

[0033] Overhead cables or lines are exposed to the external environment and are susceptible to lightning strikes, wind disasters, and human-caused damage. Therefore, the risk factor for laying them at the edges is relatively high, at 0.6. Underground direct-buried sections are protected by soil and other media and are relatively safe, so the risk factor for laying them at the edges is relatively low, at 0.4.

[0034] If there are flammable and explosive sites, hospitals, schools, government buildings, or other important public facilities or densely populated areas such as commercial areas or large residential areas near the edge, and a malfunction could cause significant casualties or property damage, the risk factor for the surrounding facilities and population density of the edge should be increased accordingly and set to 0.7. In other cases, the risk factor is 0.3.

[0035] For each edge, considering the traffic environment risk factors, the edge's construction method risk factors, and the surrounding facilities and population density risk factors, the Analytic Hierarchy Process (AHP) is used to construct a risk assessment model, yielding the final edge risk factor Risk(e). This systematic definition and data collection process ensures that the risk factor scientifically and reasonably reflects the actual risk level of the edge, providing an accurate basis for avoiding high-risk road sections during path searching.

[0036] Sb: Defines and initializes the actual cumulative cost function, heuristic function value, and total cost function of the node; In this step, let g(n) represent the actual cumulative cost function from the starting node star to node n. The initialization rule for the actual cumulative cost function of a node is as follows: In the initial stage of the improved A algorithm, g(n) of all nodes in the graph except the starting node start is initialized to infinity. This is because before the search begins, the algorithm has not yet explored any paths to these nodes, so it is assumed by default that the actual cost to reach these nodes is unknown and extremely high. As the starting node start is the starting point of the path, its actual cumulative cost to itself is naturally 0, i.e., g(start) = 0. This initialization method clarifies the starting state of the search and provides a benchmark for subsequently expanding outward from the starting node and gradually updating the g(n) values ​​of other nodes.

[0037] set up The optimized heuristic function value is used to estimate the remaining path cost from node n to the target node goal. Unlike the heuristic function in the traditional A algorithm, which is based solely on geometric distance, this improved algorithm... By incorporating average information from multiple dimensions such as edge length, impedance, and risk, the heuristic guidance becomes more aligned with the practical engineering needs for comprehensive consideration of multiple factors, thereby more accurately guiding the search towards the optimal path.

[0038] Wherein, the heuristic function value The calculation formula is: ; In the formula, This represents the geometric distance from node n to the target node goal; This represents the average impedance of the edges in the entire path network of the connected subgraph G′. Let α, β, and γ be the average risk factor of the edges in the entire path network of the connected subgraph G′, where α, β, and γ are the weight coefficients of the three factors of distance, impedance, and risk, respectively, and satisfy α+β+γ=1. v is the velocity constant used to consider the time cost of the path.

[0039] The heuristic function value The initialization rule is as follows: for each node n in the connected subgraph G′, its initial heuristic function value must be calculated according to the above formula. This step ensures that each node has a remaining path cost estimate based on multi-factor average information before the search begins, providing crucial heuristic guidance for the subsequent calculation of the total cost f(n) and the prioritization of node expansion.

[0040] Let the total cost function f(n) be the total cost of node n, obtained by adding the actual cumulative cost parameter of node n and the heuristic function value. That is, the expression of the total cost function is: ; The total cost function integrates the actual cost from the starting node to the current node and the remaining estimated cost from the current node to the target node. It is the core indicator for improving the A algorithm to determine the priority of node expansion. The smaller the total cost, the more likely the node is to be on the optimal path to the target node, and therefore it will be selected for expansion first.

[0041] The initialization rule for the total cost function is: combining g(n) and The initialization result, the total cost function of the starting node start is: For the other nodes of the connected subgraph G′, the total cost function is: This initialization result clarifies that in the initial stage of the search, only the total cost of the starting node is finite (equal to its heuristic estimate), while the total cost of other nodes is infinite. Therefore, the algorithm will prioritize the starting node for expansion, which is consistent with the logic of the search starting from the starting point.

[0042] Sc: List initialization and node ownership settings; In this step, the list initialization includes the initialization of the Open list and the Closed list. The Open list stores nodes to be expanded in the improved A algorithm, i.e., nodes that have been discovered but have not yet completed the expansion of all their adjacent nodes. During the initialization phase, the starting node is added to the Open list, at which point the Open list contains only this one node, and all other nodes are not in the Open list. The data structure of the Open list can be a priority queue to efficiently acquire the node with the smallest current total cost function f(n) each time it is selected for expansion. The Closed list stores nodes that have completed the expansion operation in the improved A algorithm, i.e., all adjacent nodes of these nodes have been explored, and the node will not be expanded again. During the initialization phase, the Closed list is empty because no node has completed the expansion operation. The Closed list can use conventional data structures such as arrays or linked lists to store node identifiers for quick determination of whether a node has been processed.

[0043] The initialization of the open and closed lists is a preparatory step in the improved A algorithm's search process. The open list initially contains the starting node, ensuring the search begins from the starting point; the closed list is initially empty, guaranteeing that all nodes have the potential to be explored in their initial state. Through the combined use of these two lists, the improved A algorithm can manage the expansion state of nodes in an orderly manner, avoiding the waste of computational resources and confusion in search logic caused by repeatedly expanding nodes. This forms a crucial data structure foundation for achieving efficient path search.

[0044] The node attribution setting specifically involves: after initializing the open and closed lists, the attribution of the starting node (`start`) needs to be confirmed again. This means the starting node must be in the open list and not in the closed list. This step is a compliance check of the initialization results, ensuring the accuracy of node state management during the subsequent execution of the improved A algorithm. For example, if the starting node is incorrectly added to the closed list, the algorithm will mistakenly believe that the node has completed expansion and will not explore its adjacent nodes, causing the entire search process to fail. By confirming node attribution, the rigor of the state after initialization of the improved A algorithm is ensured, providing a prerequisite for the smooth execution of the subsequent main loop search phase.

[0045] By optimizing the heuristic function to incorporate average information from multiple dimensions such as distance, impedance, and risk, the heuristic guidance is no longer limited to geometric distance but better aligns with the comprehensive consideration of multiple factors required in practical engineering. During the search process, the improved A algorithm can select expansion nodes based on a more comprehensive estimate of the remaining path cost, significantly improving the accuracy of the search direction and reducing ineffective expansion of irrelevant nodes, thus finding the optimal path more efficiently across multiple dimensions. The actual cumulative cost function is initialized to reflect the multi-dimensional cumulative cost from the starting point to the current node, and the total cost function is defined as the sum of the actual cumulative cost and the heuristically estimated remaining cost, allowing the node cost parameters to comprehensively reflect the actual value of the path in terms of length, impedance, and risk. This multi-dimensional cost evaluation method allows the improved A algorithm to consider not only how far the path has been traveled but also the quality of the path taken, thereby selecting the optimal path that truly meets engineering requirements.

[0046] Sd: During the main loop search phase, node expansion and cost updates are performed, and optimal node selection and termination judgment are made. In the open list Open, select the node with the smallest total cost function f(n) as the current node current. Determine whether the current node current is the goal node: if yes, proceed to the next traversal operation; if no, move the current node current from the open list Open to the closed list Closed. Iterate through the set of all neighboring nodes of the current node (current), and calculate the new actual cumulative cost function for each neighboring node. The calculation formula is as follows: ; In the formula, The comprehensive cost function is calculated using the following formula: ; In the formula, e represents the edge from the current node to an adjacent node; in the formula, This represents the geometric distance from the current node to a certain adjacent node; This represents the average impedance of the edge formed by all adjacent nodes of the current node. α is the average risk factor of the edges formed by all adjacent nodes of the current node, where α, β, and γ are the weight coefficients of the three factors of distance, impedance, and risk, respectively.

[0047] If the corresponding adjacent node is not in the open list, add it to the open list as a new node to be expanded, and then set... That is, record the current known minimum actual cumulative cost from the starting node to the corresponding neighbor node mentioned above; calculate the heuristic function value from the corresponding neighbor node to the target node. Calculate the total cost function of the corresponding adjacent nodes; and record the parent node of the corresponding adjacent node as node current, providing parent-child relationship links for subsequent path backtracking; If the corresponding neighbor node is in the open list and its new actual cumulative cost function is better, it means the path from the current node to the corresponding neighbor node is cheaper than the previously recorded path. Therefore, the relevant cost information and parent node pointer of the neighbor node need to be updated. Specifically, update... This ensures that the latest minimum actual cost to the neighbor node is recorded; the total cost function is recalculated and updated synchronously; the parent node of the corresponding adjacent node is reset to the current node to ensure that the path backtracking can point to a better parent node link.

[0048] Furthermore, after traversing all adjacent nodes of the current node, a pre-filtering operation is performed on the nodes in the open list Open. The specific rules of the pre-filtering operation are as follows: a preset threshold is set, and nodes in the open list Open that satisfy the condition that the total cost function value exceeds the preset threshold are filtered out and discarded, and no further expansion processing is performed on them.

[0049] In this embodiment, the preset threshold is defined as the total cost function value exceeding 120% of the cost of the currently found optimal path, based on the cost of the optimal path already found.

[0050] The core function of this pre-screening operation is to significantly improve the overall efficiency of the path search algorithm and alleviate the problem of excessive time complexity caused by too many nodes by actively eliminating nodes that are unlikely to generate the optimal path in a large-scale network environment.

[0051] Se: In the path backtracking phase, a reverse path is constructed to obtain the recommended temporary bypass cable laying path; Starting from the target node (goal), and based on the parent node information recorded by each node, trace backward to the starting node (start), and arrange the nodes in order to obtain the recommended temporary bypass cable laying path.

[0052] This embodiment introduces a comprehensive cost function that incorporates multiple dimensions, including geometric length, specific attributes, and risk, enabling a more precise alignment with the definition of path quality in real-world application scenarios. Different scenarios can be flexibly adapted to business needs simply by adjusting the weight coefficients. Simultaneously, a pre-screening strategy is incorporated during the expansion process, dynamically filtering low-value nodes from the Open list based on the total cost, significantly reducing the number of nodes requiring expansion. Experiments demonstrate that in large-scale network scenarios, this strategy can reduce the algorithm's time complexity by 30%-50%, effectively alleviating the performance bottleneck of the traditional A algorithm under large data volumes, allowing the algorithm to output path planning results within a reasonable timeframe.

[0053] The capacity demand model is used to calculate the minimum current-carrying capacity that the recommended temporary bypass cable laying path must have to ensure safe operation even under the worst operating conditions. The inputs of the capacity demand model include the historical or predicted maximum load current of all load nodes in the operation influence domain D, the impedance parameters of each path in the distribution network topology logic model G, and the set safety margin coefficient.

[0054] The expression for the capacity demand model is: ; In the formula, Minimum current-carrying capacity requirements for temporary bypass cable laying paths. The actual current value flowing through the temporary bypass cable laying path. The historical or predicted maximum load current for all load nodes within the operation's influence domain D.

[0055] The equipment selection model matches and generates a complete list of temporary bypass system equipment and its connection topology that can be directly procured and deployed from the material management data of the multi-source heterogeneous data, based on the calculation results of the bypass path model and the capacity requirement model. The inputs to the equipment selection model are the recommended temporary bypass cable laying path, the minimum current carrying capacity requirement of the temporary bypass cable laying path, the operating voltage level, and a standardized equipment database. The standardized equipment database contains various types of bypass switches, cables, connectors, mobile transformer substations, etc., and their technical parameters.

[0056] The equipment selection model includes a bypass cable selection submodule, a bypass switch cart selection submodule, a connection and protection equipment selection submodule, and a topology model generation submodule. The bypass cable selection submodule selects the conductor cross-sectional area of ​​the cable based on the minimum current carrying capacity requirement of the temporary bypass cable laying path. The rated current carrying capacity of the selected cable must be greater than or equal to the minimum current carrying capacity requirement of the temporary bypass cable laying path. The selection logic of the bypass switch cart submodule is as follows: a switch cart with a rated voltage that meets the working voltage level and a rated current that is greater than the minimum current carrying capacity requirement of the temporary bypass cable laying path must be selected. It should have complete relay protection functions and mechanical interlocking devices to ensure operational safety; the connection and protection equipment selection submodule is used to select matching straight-through or T-type intermediate joints according to the cable type and specifications, to make special terminal heads for the cable connection in the bypass switch, and if the recommended temporary bypass cable laying path passes through areas of vehicle crushing or sharp objects, the cable should be equipped with corresponding anti-crushing protective pipes or ground protection troughs; the topology model generation submodule is used to combine all the selected equipment according to their physical connection relationship into a complete temporary bypass system key parameter model, the temporary bypass system key parameter model clearly defines the energy flow direction, the interface type and quantity between each device.

[0057] This embodiment constructs a key parameter model of the temporary bypass system, which physically isolates the work area from the operating power grid, ensuring that users receive a continuous power supply throughout the entire operation. This embodiment introduces graph theory, optimization algorithms, and other methods to establish an accurate mathematical model, making key aspects such as bypass path selection and capacity calculation more scientific and objective, reducing reliance on human experience, and improving decision-making efficiency and the optimality of the solution.

[0058] S4: Perform uninterrupted power supply work based on the calculation results of the key parameter model of the temporary bypass system.

[0059] Example 2: The present invention also provides a live-line working system for cable-type distribution networks. The system employs a live-line working method for cable-type distribution networks as described in Example 1. The system includes: The data acquisition and preprocessing module is used to acquire multi-source heterogeneous data from cable-type distribution networks and perform data preprocessing operations on the multi-source heterogeneous data. The distribution network topology logic model building module is used to build the distribution network topology logic model of cable-type distribution networks. The module for establishing the key parameter model of the operation impact domain and temporary bypass system is used to establish the key parameter model of the operation impact domain and temporary bypass system based on the distribution network topology logic model. The execution module is used to perform uninterrupted power supply operations based on the calculation results of the key parameter model of the temporary bypass system.

[0060] Example 3: The present invention also provides an electronic device, including one or more processors and a memory.

[0061] A processor can be a central processing unit (CPU) or other form of processing unit with data processing and / or instruction execution capabilities, and can control other components in an electronic device to perform desired functions.

[0062] The memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and a processor may execute the program instructions to implement a live-line working method for a cable-type distribution network according to any embodiment of this application described above, and / or other desired functions. Various contents such as initial external parameters and thresholds may also be stored in the computer-readable storage medium.

[0063] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A method for uninterrupted power supply operation in a cable-type distribution network, characterized in that, The method includes the following steps: S1: Collect multi-source heterogeneous data of cable-type distribution networks and perform data preprocessing operations on the multi-source heterogeneous data; S2: Establish the topology logic model of the cable-type distribution network; S3: Establish the operation influence domain and key parameter models of the temporary bypass system based on the distribution network topology logic model; the key parameter model of the temporary bypass system includes a bypass path model, a capacity demand model, and an equipment selection model; the bypass path model includes a subgraph extraction module, a path search module, and an optimal path output module; the subgraph extraction module extracts a graph neural network model from the distribution network topology logic model that contains only nodes within the influence domain D and the determined power supply connection point. The determined load lead-out point The relevant key nodes are connected subgraphs G′; the path calculation module evaluates each edge in the connected subgraph G′ using an improved A algorithm to obtain a recommended temporary bypass cable laying path; the optimal path output module outputs the recommended temporary bypass cable laying path. An improved A algorithm is used to evaluate each edge in the connected subgraph G′ to obtain the recommended temporary bypass cable laying path. Specifically: Sa: Risk factor parameters are added to each edge in the connected subgraph G′; Sb: The actual cumulative cost function, heuristic function value, and total cost function of the node are defined and initialized; Sc: List initialization and node affiliation settings are performed; Sd: In the main loop search phase, node expansion and cost updates are performed, and optimal node selection and termination judgment are performed; Se: In the path backtracking phase, a reverse path is constructed to obtain the recommended temporary bypass cable laying path. S4: Perform uninterrupted power supply work based on the calculation results of the key parameter model of the temporary bypass system.

2. The method for uninterrupted power supply operation in a cable-type distribution network according to claim 1, characterized in that, In Sb, let g(n) represent the actual cumulative cost function from the starting node star to node n. The initialization rule for the actual cumulative cost function of the node is as follows: In the initial stage of the improved A algorithm, g(n) of all other nodes in the connected subgraph G′ except the starting node start is initialized to infinity; the starting node start is the starting point of the path, and its actual cumulative cost to itself is naturally 0, that is, g(start) = 0. set up The optimized heuristic function value, wherein the heuristic function value The calculation formula is: ； In the formula, This represents the geometric distance from node n to the target node goal; This represents the average impedance of the edges in the entire path network of the connected subgraph G′. Let α, β, and γ be the average risk factor of the edges in the entire path network of the connected subgraph G′, where α, β, and γ are the weight coefficients of the three factors of distance, impedance, and risk, respectively, and satisfy α+β+γ=1, and v is the velocity constant. The heuristic function value The initialization rule is as follows: for each node n in the connected subgraph G′, its initial heuristic function value must be calculated according to the above formula. ; Let the total cost function f(n) be the total cost of node n, obtained by adding the actual cumulative cost parameter of node n and the heuristic function value. That is, the expression of the total cost function is: ； The initialization rule for the total cost function is: combining g(n) and The initialization result, the total cost function of the starting node start is: For the other nodes of the connected subgraph G′, the total cost function is: 。 3. The method for uninterrupted power supply operation in a cable-type distribution network according to claim 2, characterized in that, In the Sc, the list initialization includes the initialization of the Open list and the initialization of the Closed list. The Open list is used to store nodes to be expanded in the improved A algorithm. During the initialization phase, the starting node is added to the Open list, and at this time, the Open list contains only this one node. The Closed list is used to store nodes in the improved A algorithm that have already completed the expansion operation, that is, all the adjacent nodes of these nodes have been explored, and the nodes will not be expanded again. During the initialization phase, the Closed list is empty. The node attribution setting is specifically as follows: after the initialization of the open list and the closed list is completed, the attribution of the starting node (start) needs to be confirmed again: that is, the starting node is only in the open list and not in the closed list.

4. The method for uninterrupted power supply operation in a cable-type distribution network according to claim 3, characterized in that, In the Sd, the node with the smallest total cost function f(n) in the open list Open is selected as the current node current. It is then determined whether the current node current is the goal node: if yes, proceed to the next traversal operation; if no, move the current node current from the open list Open to the closed list Closed. Iterate through the set of all neighboring nodes of the current node (current), and calculate the new actual cumulative cost function for each neighboring node. The calculation formula is as follows: ； In the formula, The comprehensive cost function is calculated using the following formula: ； In the formula, e represents the edge from the current node to an adjacent node; in the formula, This represents the geometric distance from the current node to a certain adjacent node; This represents the average impedance of the edge formed by all adjacent nodes of the current node. Let α be the average risk factor of the edges formed by all adjacent nodes of the current node, where α, β, and γ are the weight coefficients of the three factors of distance, impedance, and risk, respectively. If the corresponding adjacent node is not in the open list, add it to the open list as a new node to be expanded, and then set... Calculate the heuristic function values ​​from the adjacent nodes to the target node as described above. Calculate the total cost function of the corresponding adjacent nodes; and record the parent node of the corresponding adjacent node as node current, so as to provide parent-child relationship links for subsequent path backtracking; If the corresponding neighboring nodes are in the open list, and the new actual cumulative cost function of the corresponding neighboring nodes is better; update the relevant cost information and parent node pointers of the neighboring nodes, specifically: update This ensures that the latest minimum actual cost to the neighbor node is recorded; the total cost function is recalculated and updated synchronously; the parent node of the corresponding adjacent node is reset to the current node to ensure that the path backtracking can point to a better parent node link.

5. The method for uninterrupted power supply operation in a cable-type distribution network according to claim 4, characterized in that, After traversing all adjacent nodes of the current node, a pre-filtering operation is performed on the nodes in the open list Open. The specific rules of the pre-filtering operation are as follows: a preset threshold is set, and nodes in the open list Open that satisfy the condition that the total cost function value exceeds the preset threshold are filtered out and discarded, and no further expansion processing is performed on them.

6. The method for uninterrupted power supply operation in a cable-type distribution network according to claim 5, characterized in that, The preset threshold is defined as the total cost function value exceeding 120% of the cost of the currently found optimal path, based on the cost of the optimal path.

7. A method for uninterrupted power supply operation in a cable-type distribution network according to claim 1, characterized in that, The capacity demand model is used to calculate the minimum current-carrying capacity that the recommended temporary bypass cable laying path must have; the inputs of the capacity demand model include the historical or predicted maximum load current of all load nodes within the operation influence domain D, the impedance parameters of each path in the distribution network topology logic model G, and the set safety margin coefficient.

8. A method for uninterrupted power supply operation in a cable-type distribution network according to claim 7, characterized in that, The expression for the capacity demand model is: ； In the formula, Minimum current-carrying capacity requirements for temporary bypass cable laying paths. The actual current value flowing through the temporary bypass cable laying path. The historical or predicted maximum load current for all load nodes within the operation's influence domain D.

9. A method for uninterrupted power supply operation in a cable-type distribution network according to claim 8, characterized in that, The equipment selection model matches and generates a complete list of temporary bypass system equipment and its connection topology that can be directly purchased and deployed from a standardized equipment library based on the calculation results of the bypass path model and the capacity requirement model. The inputs to the equipment selection model are the recommended temporary bypass cable laying path, the minimum current carrying capacity requirement of the temporary bypass cable laying path, the operating voltage level, and the standardized equipment database, which contains various types of bypass switches, cables, connectors, mobile transformers and their technical parameters. The equipment selection model includes a bypass cable selection submodule, a bypass switch car selection submodule, a connection and protection equipment selection submodule, and a topology model generation submodule.

10. A live-line working system for cable-type power distribution networks, characterized in that, The system employs a live-line working method for a cable-type distribution network as described in any one of claims 1-9, and the system comprises: The data acquisition and preprocessing module is used to acquire multi-source heterogeneous data from cable-type distribution networks and perform data preprocessing operations on the multi-source heterogeneous data. The distribution network topology logic model building module is used to build the distribution network topology logic model of cable-type distribution networks. The module for establishing the key parameter model of the operation impact domain and temporary bypass system is used to establish the key parameter model of the operation impact domain and temporary bypass system based on the distribution network topology logic model. The execution module is used to perform uninterrupted power supply operations based on the calculation results of the key parameter model of the temporary bypass system.

Images