Multi-objective based distribution line sectionalization configuration method and apparatus

By constructing a multi-scenario operation dataset and screening topological weakness indicators, combined with a comprehensive scoring model and multi-objective optimization solution, the problem of multi-objective collaborative optimization in the segmented configuration of power distribution lines was solved, achieving a balance between power supply reliability, economy and low-carbon operation, and improving the adaptability and accuracy of the solution.

CN123073918APending Publication Date: 2026-11-10STATE GRID JIANGSU ELECTRIC POWER CO LTD TAIZHOU POWER SUPPLY BRANCH +2
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202611344834.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-01
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing technologies struggle to achieve multi-objective collaborative optimization in the segmented configuration of power distribution lines, failing to effectively balance the needs for power supply reliability, economy, and low-carbon operation. In particular, the uncertainty of the power distribution network's operating status increases after the integration of new energy sources such as distributed photovoltaic and wind power.

Method used

By constructing a multi-scenario running dataset, candidate segmentation configuration schemes are screened using topological weakness index and comprehensive scoring model. The final power distribution line segmentation configuration scheme is determined by combining multi-objective optimization solution and Pareto optimal solution set identification.

Benefits of technology

It achieves multi-objective collaborative optimization of segmented configuration of power distribution lines under a unified framework, taking into account the requirements of power supply reliability, economy and low-carbon operation, improving the pertinence and adaptability of the solution, and avoiding the decision-making bias of traditional methods.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123073918A_ABST
    Figure CN123073918A_ABST
Patent Text Reader

Abstract

This invention relates to the field of power distribution technology, and discloses a method and apparatus for segmented configuration of power distribution lines based on multi-objectives. The method includes: converting the power distribution lines into an undirected topology graph; constructing a topology weakness index based on multi-objective evaluation data; generating candidate segmentation points in the undirected topology graph according to the topology weakness index; arranging and combining the candidate segmentation points to generate a set of candidate segmentation configuration schemes; using a comprehensive scoring model to perform preliminary screening of each candidate segmentation configuration scheme to obtain a subset of effective candidate segmentation configuration schemes; and, under the condition of satisfying the power distribution network operation constraints, performing multi-objective collaborative optimization on the subset of effective candidate segmentation configuration schemes to obtain a Pareto optimal solution set of the candidate segmentation configuration schemes, and identifying the knee point of the Pareto optimal solution set to determine the final power distribution line segmentation configuration scheme. This method can achieve multi-objective collaborative optimization of power distribution line segmentation configuration.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of power distribution technology, and for example to a method and apparatus for segmented configuration of power distribution lines based on multiple objectives. Background Technology

[0002] Segmented configuration of power distribution lines is an important technical means to improve the reliability of power supply systems. By rationally setting sectionalizing switches along the line, the entire distribution feeder is divided into several relatively independent power supply sections, which can effectively reduce the scope of power outages due to faults, reduce power loss for users, and improve power supply reliability. However, the number and installation location of sectionalizing switches directly affect the degree to which these effects are achieved: too few switches make it difficult to effectively reduce the scope of power outages due to faults; too many switches significantly increase equipment investment and maintenance costs. Therefore, scientifically determining the number and installation location of sectionalizing switches is a key issue in achieving a balance between reliability improvement and cost control.

[0003] Most existing methods for optimizing medium-voltage distribution lines employ static planning or single-scenario analysis, determining line segmentation schemes based on typical load levels or fixed operating conditions. This implicitly assumes that the distribution network's operating state remains relatively stable within the planning period. However, with the large-scale integration of new power resources such as distributed photovoltaics, wind power, energy storage, and electric vehicles into the distribution network, the network's operation has shifted from traditional unidirectional power supply to bidirectional power flow and multi-source coordinated power supply. The power flow distribution of distribution lines is simultaneously driven by both load changes and fluctuations in renewable energy output, significantly increasing operational uncertainty. Furthermore, continuous load growth, time-varying user electricity consumption behavior, and random fluctuations in renewable energy output further exacerbate the dynamic evolution of the operating state. Against this backdrop, the segmentation configuration of medium-voltage distribution lines must not only meet power supply reliability requirements but also consider multiple objectives such as construction investment, operational economy, and low-carbon operation. Therefore, traditional static planning or single-objective optimization methods are insufficient to meet the multi-objective comprehensive optimization needs of modern distribution networks.

[0004] In existing technologies, multi-objective optimization methods have been introduced into the field of distribution network planning. For example, patent application CN 202610271588.7 discloses a distributed photovoltaic (PV) distribution network planning method based on a green certificate trading mechanism. This method constructs a multi-objective programming model with the objectives of minimizing the total annual investment and operating cost, optimizing the distribution network voltage stability, and minimizing the system's active power loss. It also incorporates constraints such as node power balance, node voltage, PV installation capacity, green certificate quotas, and trading. A swarm intelligence optimization algorithm is used to solve the multi-objective programming model, yielding the optimal installation nodes, installation capacity, and corresponding green certificate trading strategies and quota fulfillment schemes for distributed PV. This method achieves a balance between the economic efficiency and technical reliability of distributed PV configuration through multi-objective optimization.

[0005] In the process of implementing the embodiments of the present invention, at least the following problems were found in the related art: While existing technologies have obtained distributed photovoltaic (PV) distribution network planning schemes that meet multi-objective requirements by constructing and solving multi-objective programming models, their optimization objects are the installation nodes and installed capacity of distributed PV, and do not involve the location and quantity configuration of distribution line sectionalizing switches. Therefore, they cannot be directly transferred to the scenario of distribution line sectionalizing. Thus, a multi-objective-based method for distribution line sectionalizing configuration is needed to achieve multi-objective collaborative optimization of distribution line sectionalizing. Summary of the Invention

[0006] To address the shortcomings of the existing technology, the present invention provides a multi-objective method and apparatus for segmented configuration of power distribution lines, so as to achieve multi-objective collaborative optimization of segmented configuration of power distribution lines.

[0007] According to one aspect of the present invention, the present invention provides a multi-objective-based method for segmented configuration of power distribution lines, comprising: obtaining multi-source operation data of the power distribution system in which the power distribution line is located; identifying and correcting anomalies in the multi-source operation data; constructing a multi-scenario operation dataset; and determining multi-objective evaluation data based on the multi-scenario operation dataset; converting the power distribution line into an undirected topology graph; constructing a topology weakness index based on the multi-objective evaluation data; generating candidate segmentation points in the undirected topology graph according to the topology weakness index; and arranging and combining the candidate segmentation points to generate a set of candidate segmentation configuration schemes; based on multi-scenario... The system calculates multi-objective evaluation indicators for each candidate segmented configuration scheme in different operating scenarios using a scenario-based dataset. A comprehensive scoring model is then used to initially screen each candidate segmented configuration scheme, obtaining a subset of effective candidate segmented configuration schemes. The comprehensive scoring model is established based on the multi-objective evaluation indicators. Under the condition of satisfying the distribution network operation constraints, multi-objective collaborative optimization is performed on the subset of effective candidate segmented configuration schemes to obtain the Pareto optimal solution set of the candidate segmented configuration schemes. Knee point identification is then performed on the Pareto optimal solution set to determine the final distribution line segmented configuration scheme.

[0008] Optionally, the multi-source operation data includes static data of distribution lines, historical operation data of the distribution system, and forecast information; obtaining multi-source operation data of the distribution system where the distribution lines are located, identifying and correcting anomalies in the multi-source operation data, and constructing a multi-scenario operation dataset, including: identifying anomalies in historical operation data, determining the abnormal data and the corresponding anomaly types, correcting the abnormal data according to the correction methods corresponding to the anomaly types, and obtaining corrected historical operation data; based on static data, combined with the corrected historical operation data and forecast information, constructing long-term load growth scenarios and short-term renewable energy fluctuation scenarios respectively; and... In the long-term load growth scenario, each load growth state is paired with the output fluctuation state of each distributed power source in the short-term renewable energy fluctuation scenario to obtain multiple operating scenarios. In each operating scenario, the node load demand of each node is taken as the node load value under the corresponding load growth state, and the distributed power output of each node is taken as the distributed power output value under the corresponding output fluctuation state. The net load of each node under the operating scenario is calculated based on the node load demand and the distributed power output. The node load demand, distributed power output, net load of each node under all operating scenarios, as well as the multi-source operating data, are combined to form a multi-scenario operating dataset.

[0009] Optionally, the multi-objective evaluation data serves as intermediate data in the process of generating reliability evaluation indicators, economic evaluation indicators, and carbon emission evaluation indicators, including line load rate data and affected load distribution data. Based on the multi-objective evaluation data, a topological weakness index is constructed, including: converting the distribution lines into an undirected topology graph, where the undirected topology graph uses distribution network nodes as the vertex set and line edges as the edge set; calculating the edge betweenness number of each line edge, where the edge betweenness number is the frequency with which all nodes pass through the line edge for the shortest electrical path; determining the normalized value of the line load rate for each line edge based on the line load rate data, and determining the normalized scale of the important loads affected by a fault at each line edge based on the affected load distribution data; and weighted summing the edge betweenness number, the normalized value of the line load rate, and the normalized scale of the important loads affected by the fault to obtain the topological weakness index for each line edge.

[0010] Optionally, candidate segmentation points are generated directionally in the undirected topology graph based on the topology weakness index. These candidate segmentation points are then arranged and combined to generate a set of candidate segmentation configuration schemes. This includes: identifying line segments in the undirected topology graph with topology weakness index values ​​higher than a preset level as priority candidate segments, and directionally generating candidate segmentation points within these priority candidate segments; performing minimum cut set analysis on the undirected topology graph to determine the cut set edge set; wherein, disconnection of line edges in the cut set edge set results in some nodes in the undirected topology graph losing connectivity with power supply nodes; and performing feeder critical path analysis on the undirected topology graph to determine critical... The path edge set is defined as follows: the line edges in the critical path edge set are located on the power supply path from the power source node to the end load node, and the number of nodes affected after a fault exceeds a preset value. The line segments corresponding to the line edges in the cut set edge set and the critical path edge set are supplemented as candidate segmentation points, and the directionally generated candidate segmentation points are merged with the supplemented candidate segmentation points to obtain a candidate segmentation point set. Combining line length constraints, feeder radiation structure constraints, and segmented switch constraints, the candidate segmentation points in the candidate segmentation point set are combined and arranged to generate a candidate segmentation configuration scheme set.

[0011] Optionally, a minimum cut set analysis is performed on the undirected topological graph to determine the cut set edge set, including: calculating the edge connectivity contribution value of each path edge in the undirected topological graph. ,in, To remove the edge of the line e forward, and The number of connected paths between them After removing line edge e, and The number of connected paths between them For power nodes, For the i-th load node, The set of load nodes; the line edges with a connectivity contribution value greater than zero are defined as cut set edges, and all cut set edges constitute the cut set edge set; Critical path analysis is performed on the undirected topology graph to determine the set of critical path edges, including: for each power supply path from the power source node to the end load node, calculating the fault impact range index of each line edge on the power supply path. ,in, For load nodes The load weighting factor, For load nodes The load demand, The baseline failure rate for line edge e is... For load nodes Power restoration time The set of load nodes downstream that lose power when a fault occurs at line edge e; the line edges whose fault impact range index exceeds a preset threshold are identified as critical path edges, and the set of critical path edges is composed of all critical path edges.

[0012] Optionally, based on a multi-scenario operation dataset, multi-objective evaluation indicators are calculated for each candidate segmented configuration scheme in the candidate segmented configuration scheme set under different operation scenarios. This includes: obtaining the segmented switch configuration information corresponding to each candidate segmented configuration scheme in the candidate segmented configuration scheme set; performing power flow calculations on the candidate segmented configuration schemes based on the node load demand and distributed power output of each node in the target operation scenario in the multi-scenario operation dataset, combined with the segmented switch configuration information, to obtain the active power loss and line load rate of each line segment in the target operation scenario; performing fault simulations on the candidate segmented configuration schemes based on the line load rate of each line segment in the target operation scenario, combined with the historical fault data corresponding to the target operation scenario, to determine the affected load capacity and outage duration after a fault in each line segment in the target operation scenario; and calculating the reliability evaluation index, economic evaluation index, and carbon emission evaluation index of the candidate segmented configuration schemes in the target operation scenario based on the affected load capacity, outage duration, and active power loss of each line segment in the target operation scenario.

[0013] Optionally, a comprehensive scoring model is used to initially screen each candidate segment configuration scheme to obtain a subset of effective candidate segment configuration schemes. This includes: obtaining a comprehensive evaluation value for each candidate segment configuration scheme based on reliability evaluation indicators, economic evaluation indicators, and carbon emission evaluation indicators using a comprehensive scoring model; screening each candidate segment configuration scheme in the candidate segment configuration scheme set based on the comprehensive evaluation value to obtain a subset of intermediate candidate segment configuration schemes; during the application of each candidate segment configuration scheme in the subset of intermediate candidate segment configuration schemes, real-time acquisition of the operating status data of the power distribution system, and determining the degree of change in line operating status based on the operating status data; and dynamically correcting and incrementally updating the candidate segment configuration schemes in the subset of intermediate candidate segment configuration schemes based on the degree of change in line operating status to obtain a subset of effective candidate segment configuration schemes.

[0014] Optionally, under the condition of satisfying the distribution network operation constraints, a multi-objective collaborative optimization solution is performed on the subset of effective candidate segmented configuration schemes to obtain the Pareto optimal solution set of the candidate segmented configuration schemes. This includes: establishing a multi-objective optimization model with the minimum of reliability evaluation index, economic evaluation index, and carbon emission evaluation index as the optimization objectives; establishing distribution network operation constraints in the multi-objective optimization model; wherein, the distribution network operation constraints include flexible constraints, and the flexible constraints are fuzzified. When the degree of index exceeding the limit is within the tolerance range, the corresponding candidate segmented configuration scheme is not directly eliminated. Instead, a penalty amount is determined according to the degree of index exceeding the limit, and the penalty amount is superimposed on the objective component of the optimization objective; based on the Pareto optimality principle, the objective component values ​​of each candidate segmented configuration scheme in the subset of effective candidate segmented configuration schemes are non-dominated and sorted to obtain the Pareto optimal solution set composed of mutually non-dominated candidate segmented configuration schemes.

[0015] Optionally, the Pareto optimal solution set is subjected to knee point identification to determine the final distribution line segmentation configuration scheme. This includes: determining the target points corresponding to each candidate segmentation configuration scheme in the target space, and constructing a Pareto front from the target points corresponding to all candidate segmentation configuration schemes in the Pareto optimal solution set; wherein the target space consists of three dimensions: reliability evaluation index, economic evaluation index, and carbon emission evaluation index; searching for inflection knee points with diminishing marginal benefits along the Pareto front, and using the candidate segmentation configuration schemes corresponding to the inflection knee points as the final distribution line segmentation configuration schemes; wherein the candidate segmentation configuration schemes corresponding to the inflection knee points are determined to be the distribution line segmentation configuration schemes with the best comprehensive cost-effectiveness among the reliability evaluation index, economic evaluation index, and carbon emission evaluation index.

[0016] According to another aspect of the present invention, the present invention also provides a multi-objective-based distribution line segmentation configuration device, comprising: a multi-objective evaluation vector construction module, configured to obtain multi-source operation data of the distribution system in which the distribution line is located, identify and correct anomalies in the multi-source operation data, construct a multi-scenario operation dataset, and determine multi-objective evaluation data based on the multi-scenario operation dataset; a candidate segmentation configuration scheme generation module, connected to the multi-objective evaluation vector construction module, configured to convert the distribution line into an undirected topology graph, construct a topology weakness index based on the multi-objective evaluation data, generate candidate segmentation points in the undirected topology graph according to the topology weakness index, and arrange and combine the candidate segmentation points to generate a candidate segmentation configuration scheme set; and a candidate segmentation configuration scheme filtering module. The module is connected to the candidate segmented configuration scheme generation module and is configured to calculate the multi-objective evaluation index of each candidate segmented configuration scheme in the candidate segmented configuration scheme set under different operating scenarios based on a multi-scenario running dataset. It also uses a comprehensive scoring model to perform preliminary screening of each candidate segmented configuration scheme to obtain a subset of effective candidate segmented configuration schemes. The comprehensive scoring model is established based on the multi-objective evaluation index. The multi-objective collaborative optimization and decision module is connected to the candidate segmented configuration scheme screening module and is configured to perform multi-objective collaborative optimization on the subset of effective candidate segmented configuration schemes under the condition of satisfying the distribution network operation constraints, obtain the Pareto optimal solution set of the candidate segmented configuration schemes, and perform knee point identification on the Pareto optimal solution set to determine the final distribution line segmented configuration scheme.

[0017] The multi-objective-based distribution line segmentation configuration method and apparatus provided by the present invention have at least the following beneficial technical effects: 1. Through a progressive architecture, multi-objective evaluation elements are integrated throughout the entire process of candidate segment configuration scheme generation, preliminary screening, optimization solution and final decision-making; Pareto non-dominated sorting is used to fully preserve the conflict relationship between multiple objectives, and knee point identification is used to objectively determine the candidate segment configuration scheme with the best overall cost performance from the Pareto front, thereby realizing multi-objective collaborative optimization of power distribution line segment configuration under a unified framework.

[0018] 2. By combining topological weakness index screening with minimum cut set analysis and feeder critical path analysis, the candidate segmentation point set can cover both topologically weak sections and critical locations with a large fault impact range. This effectively reduces the candidate solution space while ensuring the quality of candidate solutions, and improves the relevance and completeness of the generated candidate segmentation configuration solutions.

[0019] 3. The comprehensive scoring model is used to quickly eliminate schemes with poor overall performance, thereby narrowing the search scale. Then, by monitoring the degree of change in the operating status in real time, only the candidate schemes affected by the changes are dynamically corrected and incrementally updated, avoiding full recalculation and balancing screening efficiency with the dynamic adaptability of the candidate scheme set to changes in the operating environment.

[0020] 4. By mapping the Pareto optimal solution set to a three-dimensional target space to form a Pareto front, and automatically searching for the turning knee point along the front to determine the final solution, the solution with the best overall cost-effectiveness is objectively identified based on the geometric shape of the front, thereby improving the objectivity and credibility of multi-objective decision-making. Attached Figure Description

[0021] Figure 1 A flowchart illustrating a multi-objective-based power distribution line segmentation configuration method provided in an embodiment of the present invention; Figure 2 A flowchart illustrating another multi-objective-based distribution line segmentation configuration method provided in an embodiment of the present invention; Figure 3 A flowchart illustrating another multi-objective-based distribution line segmentation configuration method provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of an undirected topological graph provided in an embodiment of the present invention; Figure 5 A schematic diagram of a multi-objective power distribution line segmentation configuration device provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of another multi-objective power distribution line segmentation configuration device provided in an embodiment of the present invention. Detailed Implementation

[0022] To better understand the above technical solutions, a detailed description of the solutions will be provided below in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0023] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0024] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.

[0025] In this embodiment of the invention, the power distribution system in which the distribution line is located refers to the end link of the power network that receives electrical energy from the transmission network or local power source and distributes it to the end users through the distribution line. Its voltage level covers multiple levels such as medium voltage and low voltage. The network structure mostly adopts a radial or weak ring network operation mode. The feeder is divided into several sections by sectionalizing switches, and the load transfer after fault isolation is realized by tie switches.

[0026] In some practical applications, the aforementioned power distribution system is a medium-voltage power distribution system. Medium-voltage power distribution systems typically refer to power distribution networks with voltage levels ranging from 10kV to 35kV, serving as a crucial intermediate link connecting high-voltage transmission networks and low-voltage users. Medium-voltage power distribution systems are the primary access level for distributed power sources such as distributed photovoltaics and wind power. The randomness and intermittency of distributed power output are superimposed at the medium-voltage level, resulting in significant time-varying and uncertain power flow distribution and operating status in medium-voltage power distribution systems. Therefore, the segmented configuration scheme places greater emphasis on comprehensive evaluation and adaptive correction across multiple scenarios and time domains.

[0027] Medium-voltage power distribution systems are a typical and preferred application scenario for the methods of the embodiments of the present invention. However, it should be understood that the methods of the embodiments of the present invention are also applicable to other voltage level power distribution systems with similar segmented configuration requirements. The embodiments of the present invention do not specifically limit this.

[0028] Combination Figure 1 As shown, this embodiment of the invention provides a multi-objective-based distribution line segmentation configuration method, including the following steps: S10: Obtain multi-source operation data of the power distribution system where the power distribution line is located, identify and correct anomalies in the multi-source operation data, construct a multi-scenario operation dataset, and determine multi-objective evaluation data based on the multi-scenario operation dataset.

[0029] Here, multi-source operational data includes static data of distribution lines, historical operational data of the distribution system, and forecast information. Specifically: static data of distribution lines refers to data describing the distribution line's topology and inherent electrical parameters, which remain largely unchanged or change slowly with operational status during the planning period. This includes inherent parameters related to line operation carbon emission calculations, such as distribution network topology, line parameters, sectionalizing switch locations, and carbon emission factors per unit of electricity. Historical operational data of the distribution system refers to data collected and recorded during the operation of the distribution system, reflecting the actual operational status and its changing patterns at different time points. This includes historical load data for each node, historical output data of distributed generation sources, historical fault data, and historical line power flow data. Forecast information of the distribution system refers to information obtained based on historical operational data and external forecasting methods, reflecting future operational status trends and the evolution direction of connected resources. This includes regional load growth forecast information, user access planning information, distributed generation output forecast information, and distributed generation grid connection planning information.

[0030] A multi-scenario operation dataset refers to a standardized dataset covering multiple operation scenarios, formed by time alignment, scenario construction, and pairing of multi-source operation data from a power distribution system. As a unified data foundation, the multi-scenario operation dataset serves two main purposes: firstly, it provides input for determining multi-objective evaluation data, supports the construction of topological weakness indicators, and thus provides multi-objective guidance for generating candidate segmented configuration schemes; secondly, it directly serves as the operation scenario input for calculating multi-objective evaluation indicators for each candidate segmented configuration scheme, providing computational conditions covering multi-timescale operation characteristics for candidate scheme evaluation. This ensures that the generation and evaluation of segmented configuration schemes no longer rely on a single fixed operating condition but are based on multi-timescale operation conditions covering long-term load growth and short-term renewable energy fluctuations.

[0031] Multi-objective evaluation data refers to intermediate data generated or used in the process of generating reliability evaluation indicators, economic evaluation indicators, and carbon emission evaluation indicators. It includes at least line load rate data and affected load distribution data. The line load rate data is obtained by performing power flow calculations on each operating scenario in the multi-scenario operation data set, reflecting the load level of each line segment under different operating scenarios. The affected load distribution data is obtained by analyzing the fault impact range, reflecting the downstream load capacity affected by each line fault and its spatial distribution.

[0032] S20 transforms the power distribution lines into an undirected topology graph, constructs a topology weakness index based on multi-objective evaluation data, generates candidate segmentation points in the undirected topology graph according to the topology weakness index, and arranges and combines the candidate segmentation points to generate a set of candidate segmentation configuration schemes.

[0033] Here, the topology vulnerability index refers to an evaluation index used to quantitatively characterize the comprehensive vulnerability of each line edge in an undirected topology graph across three dimensions: topological importance, operating load level, and severity of fault consequences. A line edge with a higher topology vulnerability index value means that the line edge occupies a more critical hub position in the topology, has a higher load level under current operating conditions, and will have a larger scale of significant loads affected after a fault. Therefore, the marginal improvement benefit of installing a sectionalizing switch at that location is also greater.

[0034] S30, based on the multi-scenario running dataset, calculate the multi-objective evaluation index of each candidate segment configuration scheme in the candidate segment configuration scheme set under different running scenarios, and use the comprehensive scoring model to perform preliminary screening of each candidate segment configuration scheme to obtain a subset of effective candidate segment configuration schemes; wherein, the comprehensive scoring model is established based on the multi-objective evaluation index.

[0035] S40, under the condition of satisfying the distribution network operation constraints, performs multi-objective collaborative optimization on the subset of effective candidate segmented configuration schemes to obtain the Pareto optimal solution set of the candidate segmented configuration schemes, and identifies the knee point of the Pareto optimal solution set to determine the final distribution line segmented configuration scheme.

[0036] The multi-objective-based distribution line segmentation configuration method provided in this invention constructs a multi-scenario operation dataset and determines multi-objective evaluation data, providing a unified evaluation basis covering the operational characteristics of multiple time scales for segmentation configuration. By integrating the multi-objective evaluation data into the topological weakness index, the generation process of candidate segmentation points itself becomes multi-objective-oriented, effectively reducing the candidate scheme space and improving the quality of candidate schemes. Through a hierarchical optimization architecture combining preliminary screening with multi-objective collaborative optimization, a Pareto optimal solution set is obtained under the premise of satisfying the distribution network operation constraints, ensuring that the conflict relationships between multiple objectives are fully preserved. By identifying the knee point, the comprehensive optimal scheme is objectively determined from the Pareto front, avoiding decision-making bias caused by subjective weighting. Thus, this method realizes multi-objective collaborative optimization of distribution line segmentation configuration under a unified framework. The resulting segmentation configuration scheme can take into account the operational needs of multiple objectives and can be dynamically updated to adapt to changes in the operating status of the distribution system.

[0037] Optionally, in step S10, multi-source operation data of the power distribution system where the power distribution line is located is obtained, anomaly identification and correction are performed on the multi-source operation data, and a multi-scenario operation dataset is constructed, including: obtaining multi-source operation data of the power distribution system after time alignment; identifying anomalies in historical operation data in the multi-source operation data, determining the abnormal data and the anomaly type corresponding to the abnormal data, correcting the abnormal data according to the correction method corresponding to the anomaly type, and obtaining corrected historical operation data; based on static data, combined with the corrected historical operation data and prediction information, long-term load growth scenario and short-term renewable energy fluctuation scenario are constructed respectively; In the long-term load growth scenario, each load growth state is paired with the output fluctuation state of each distributed power source in the short-term renewable energy fluctuation scenario to obtain multiple operating scenarios. In each operating scenario, the node load demand of each node is taken as the node load value under the corresponding load growth state, and the distributed power output of each node is taken as the distributed power output value under the corresponding output fluctuation state. The net load of each node under the operating scenario is calculated based on the node load demand and the distributed power output. The node load demand, distributed power output, net load of each node under all operating scenarios, as well as the multi-source operating data, are combined to form a multi-scenario operating dataset.

[0038] In some possible implementations, multi-source runtime data is time-aligned according to the following formula: in, The alignment value of the k-th class of data at a unified time t after time-weighted mapping. For the k-th type of data at the original sampling time The sampled values ​​on For the j-th original sampling time of the k-th data type, is the weight coefficient corresponding to the j-th original sampling point, t is the target time on the unified time series after time alignment, and k is the index number of the multi-source running data.

[0039] Weighting coefficient w j Based on the original sampling time t j The time distance between the target time t and the target time is determined. The closer the time distance, the larger the weight coefficient; the farther the time distance, the smaller the weight coefficient. Through this weighted mapping method, data from different sampling periods can be unified under the same time coordinate, providing a temporally consistent data foundation for subsequent scene construction and power flow calculation.

[0040] In some possible implementations, based on static data, combined with corrected historical operating data and forecast information, long-term load growth scenarios and short-term renewable energy fluctuation scenarios are constructed respectively. These include: determining the topology and line parameters of distribution lines based on static data; predicting the load growth trend of each node within the planning period based on historical load data of each node in the historical operating data, combined with regional load growth forecast information and user access planning information in the forecast information, obtaining multiple load growth states; each load growth state corresponds to a set of node load demand values ​​for each node; multiple load growth states constitute a long-term load growth scenario; simulating the fluctuation characteristics of distributed power generation output based on historical output data of distributed power sources in the historical operating data, combined with distributed power generation output forecast information in the forecast information, obtaining multiple distributed power generation output fluctuation states; each distributed power generation output fluctuation state corresponds to a set of output values ​​for each distributed power source; multiple distributed power generation output fluctuation states constitute a short-term renewable energy fluctuation scenario.

[0041] In some possible implementations, the net load of each node in the operating scenario is calculated based on the node load demand and distributed power output, including calculating the net load of each node according to the following formula: in, Let i be the net load of node i in scenario s at time t. Let be the load demand of node i at time t in scenario s. The distributed power output of node i in scenario s at time t is given, where s is the scenario number, i is the node number, and t is the time identifier.

[0042] By pairing long-term load growth scenarios with short-term renewable energy fluctuation scenarios, multiple operating scenarios covering operating characteristics at multiple time scales are generated. This enables the multi-scenario operating dataset to comprehensively reflect the operating conditions of the power distribution system under different load growth levels and renewable energy output fluctuations, providing a data foundation that takes into account both long-term trends and short-term fluctuations for the generation and evaluation of subsequent candidate segmented configuration schemes.

[0043] In some possible implementations, anomaly identification is performed on historical operational data in multi-source operational data to determine the anomalous data and the corresponding anomaly type. The anomalous data is then corrected according to the correction method corresponding to the anomaly type. This includes: when the anomaly type is data missing, identifying the historical operational data to be corrected that is missing for a continuous long period in the historical operational data; under the same operational scenario, calculating the coupling similarity between node load and distributed power generation output based on the correlation between node load data and distributed power generation output data in the historical operational data for periods without missing data; based on the coupling similarity, determining the reference period that is closest to the load-output coupling pattern of the missing period, and supplementing the historical operational data to be corrected with historical operating condition data within the reference period in the historical operational data.

[0044] Furthermore, the coupling similarity between node load and distributed generation output is calculated using the following formula: in, Coupling similarity represents the degree of similarity between the missing period and the reference period in terms of the coupling pattern between node load and distributed generation output. Let be the load deviation of node i in scenario s (the difference between the load demand at node i during the missing period and the reference period). The distributed power output deviation of node i in scenario s (distributed power output deviation of node i in scenario s). Let L2 be the L2 norm of the vector.

[0045] By calculating the L2 norm of the two-dimensional vector formed by node load deviation and distributed generation output deviation, the combined deviation between the missing period and the reference period in terms of both load and output is quantified. The smaller the deviation, the smaller the L2 norm, and the closer the exponential function value is to 1, indicating a more similar coupling pattern between the two. Therefore, the coupling similarity considers the joint variation characteristics of both node load and distributed generation output variables, and can be used to select the reference period that is closest to the operating conditions of the missing period from historical operating data to support the completion of missing data for continuous long periods.

[0046] By identifying continuous long-term missing data and matching reference time periods based on the coupling similarity between node load and distributed power output, the completed data maintains consistency with the missing time periods in terms of load-output linkage. This avoids deviations in subsequent power flow calculations and evaluation index calculations due to missing data, and improves the accuracy of multi-scenario operation datasets in reproducing actual operating conditions.

[0047] In other possible implementations, the anomaly types include instantaneous device anomalies, scene transition anomalies, and cross-scene temporal drift anomalies. Anomaly identification is performed on historical operational data from multi-source operational data to determine the abnormal data and its corresponding anomaly type. The abnormal data is then corrected according to the correction method corresponding to the anomaly type. This also includes: correcting abnormal data of the instantaneous device anomaly type using a neighborhood weighted average correction method; correcting abnormal data of the scene transition anomaly type using a multi-scene boundary threshold truncation correction method; and correcting abnormal data of the cross-scene temporal drift anomaly type using a neighborhood interpolation completion method.

[0048] Furthermore, for abnormal data with the anomaly type of instantaneous equipment anomaly, a neighborhood weighted mean correction method is used, including correction according to the following formula: in, The corrected data values. For the set of samples in the neighborhood of the outlier, Let be the weight coefficient corresponding to the p-th sample in the neighborhood sample set Ψ. Let p be the original sample value of the p-th sample in the neighborhood sample set Ψ, where Ψ is the neighborhood sample set of the outlier, and p is the index number of the sample in the neighborhood sample set Ψ. By replacing the outlier value with the weighted average of the normal samples in the neighborhood of the outlier, the local abrupt changes caused by the instantaneous equipment anomaly to the historical operating data can be effectively eliminated, maintaining the local smoothness and continuity of the data in the time series.

[0049] By classifying abnormal data in historical operation data into instantaneous device anomalies, scene transition anomalies, and cross-scene temporal drift anomalies, and then applying targeted corrections using neighborhood weighted mean correction, multi-scene boundary threshold truncation correction, and neighborhood interpolation completion methods respectively, the impact of different types of anomalies on the quality of multi-source operation data can be effectively eliminated, improving the accuracy and reliability of multi-scene operation datasets and providing a high-quality data foundation for the subsequent generation of candidate segmentation configuration schemes and the calculation of multi-objective evaluation indicators.

[0050] Optionally, in step S20, a topological weakness index is constructed based on multi-objective evaluation data, including: converting the power distribution lines into an undirected topological graph, wherein the undirected topological graph has power distribution network nodes as the vertex set and line edges as the edge set; calculating the edge betweenness number of each line edge, wherein the edge betweenness number is the frequency of all nodes passing through the line edge for the shortest electrical path; determining the normalized value of the line load rate of each line edge based on the line load rate data, and determining the normalized scale of the important loads affected by a fault at each line edge based on the affected load distribution data; and weighted summing the edge betweenness number, the normalized value of the line load rate, and the normalized scale of the important loads affected by the fault to obtain the topological weakness index of each line edge.

[0051] In some possible implementations, a weighted sum is taken of the edge betweenness, the normalized value of the line load rate, and the normalized scale of the fault affecting critical loads to obtain the topology weakness index for each line edge, including calculating the topology weakness index according to the following formula: in, Let e ​​be the topological weakness of the line edge. Let e ​​be the edge betweenness of the line edge. This is the normalized value of the line load rate. This represents the normalized scale of the critical loads affected by a fault at line edge e. These are the weighting coefficients.

[0052] By integrating the line load rate data and affected load distribution data generated during the multi-objective evaluation data generation process into the construction of the topology weakness index, the directional generation of candidate segment locations not only depends on topology characteristics, but also takes into account multi-objective evaluation factors such as operating status and fault consequences. Thus, multi-objective guidance is embedded in the candidate scheme generation stage, improving the pertinence and quality of the candidate segment configuration scheme set.

[0053] Optionally, combined Figure 2 As shown, in step S20, candidate segmentation points are generated in the undirected topology graph according to the topology weakness index, and the candidate segmentation points are arranged and combined to generate a set of candidate segmentation configuration schemes, including the following steps: S201, the line segments in the undirected topology graph whose topological weakness index value is higher than the preset level are identified as priority candidate segments, and candidate segment points are generated in a directional manner within the priority candidate segments.

[0054] S202, perform minimum cut set analysis on the undirected topology graph to determine the cut set edge set; where, when the line edge in the cut set edge set is broken, some nodes in the undirected topology graph lose their connection with the power supply node.

[0055] In some possible implementations, a minimum cut set analysis is performed on the undirected topological graph to determine the cut set edge set, including: calculating the edge connectivity contribution value of each path edge in the undirected topological graph. ,in, To remove the edge of the line e forward, and The number of connected paths between them After removing line edge e, and The number of connected paths between them For power nodes, For the i-th load node, The set of load nodes; the line edges with a connectivity contribution value greater than zero are identified as cut set edges, and the cut set edge set is composed of all cut set edges.

[0056] Based on the accurate calculation of the changes in connectivity paths, all critical line edges on the necessary power supply path between the power source and the load node can be identified objectively and systematically, avoiding omissions or misjudgments. This provides a complete and reliable topological basis for supplementing candidate segmentation points, improving the comprehensiveness and accuracy of candidate segmentation point generation.

[0057] S203, perform feeder critical path analysis on the undirected topology graph to determine the critical path edge set; wherein, the line edges in the critical path edge set are located on the power supply path from the power node to the end load node, and the number of nodes affected after a fault exceeds a preset value.

[0058] In some possible implementations, feeder critical path analysis is performed on the undirected topology graph to determine the set of critical path edges, including: for each power supply path from the power source node to the end load node, calculating the fault impact range index of each line edge on the power supply path. ,in, For load nodes The load weighting factor, For load nodes The load demand, The baseline failure rate for line edge e is... For load nodes Power restoration time The set of load nodes downstream that lose power when a fault occurs at line edge e; the line edges whose fault impact range index exceeds a preset threshold are identified as critical path edges, and the set of critical path edges is composed of all critical path edges.

[0059] By simultaneously incorporating the weighting coefficients of downstream power-outage load nodes, load demand, baseline failure rate, and power restoration time into the fault impact range index, and by compound weighting the outage capacity and expected outage loss, the determination of critical path edges not only reflects the static outage range after a fault, but also integrates the impact of fault occurrence probability and restoration time on the consequences of the fault. This allows for a more accurate identification of the line edges with the most severe fault consequences, providing a more quantitative basis that aligns with actual operational risk characteristics for supplementing candidate segmentation points at critical locations, and improving the coverage and targeting of candidate segmentation configuration schemes for high-risk sections.

[0060] S204, supplement the line segments corresponding to the line edges in the cut set edge set and the critical path edge set as candidate segmentation points, and merge the directional generated candidate segmentation points with the supplemented candidate segmentation points to obtain a candidate segmentation point set.

[0061] S205 combines line length constraints, feeder radiation structure constraints, and segmented switch constraints to combine and arrange the candidate segmented points in the candidate segmented point set to generate a candidate segmented configuration scheme set.

[0062] In some possible implementations, combining line length constraints, feeder radiation structure constraints, and sectionalizing switch constraints, the candidate segmentation points in the candidate segmentation point set are combined and arranged to generate a candidate segmentation configuration scheme set. This includes: combining the candidate segmentation points in the candidate segmentation point set according to a preset upper limit for the number of newly added sectionalizing switches to obtain multiple candidate segmentation point combinations; eliminating combinations that do not meet the line length constraints, which include the line length between adjacent sectionalizing switches not being less than a preset minimum length; and filtering the combinations after line length constraint selection. Among the candidate segmentation point combinations, those that do not meet the feeder radiation structure constraints are eliminated. The feeder radiation structure constraints include the distribution line still maintaining a radial operation and not forming a ring network after the addition of segment switches. Among the candidate segmentation point combinations screened by the feeder radiation structure constraints, those that conflict with the existing segment switch layout are eliminated. Conflicts include the candidate segmentation point coinciding with the existing segment switch position or the adjacent distance being less than the preset minimum spacing. The candidate segmentation point combinations retained after constraint screening are determined as candidate segmentation configuration schemes, and all the retained candidate segmentation configuration schemes constitute the candidate segmentation configuration scheme set.

[0063] By sequentially screening candidate segmentation point combinations for line length constraints, feeder radiation structure constraints, and conflicts with existing segmentation switch layouts, it is ensured that each scheme in the generated candidate segmentation configuration scheme set is feasible in engineering implementation, does not conflict with existing configurations, and does not change the radiation operation mode of the distribution network. This avoids invalid calculations for infeasible schemes in subsequent multi-objective evaluation and optimization stages, thereby improving optimization solution efficiency.

[0064] Combination Figure 4 As shown, taking a 10kV single-radial medium-voltage feeder as an example, the specific construction process of the candidate segment configuration scheme set is explained.

[0065] This 10kV single-radial medium-voltage feeder includes bus node 0, node 1 (residential load), node 2 (residential + commercial load), node 3 (commercial + industrial load), and node 4 (important industrial load). Node 2 is connected to distributed photovoltaic (PV) systems. The distribution line is abstracted as an undirected topology graph G=(V, E), where V={0,1,2,3,4} and E={e1(0-1), e2(1-2), e3(2-3), e4(3-4)}.

[0066] First, calculate the topological weakness index T for each line edge. e =α×B e +β×L e +γ×F e Let α=0.3, β=0.4, and γ=0.3; the frequency of the shortest electrical path passing through the line edge for all node pairs is used as the basis. In a single-radial topology, this is equal to the number of downstream nodes of the line edge, and after normalization, B... e1 =1.00、B e2 =0.75, B e3 =0.50、B e4 =0.25; Take the larger load rate value under each operating scenario and truncate it to 1.0 to get L. e1 =0.52, L e2 =0.70, L e3 =1.00、L e4 =1.00; Let the node load weights be Node 1 (Residential) = 0.5, Node 2 (Residential + Commercial) = 0.5, Node 3 (Commercial + Industrial) = 1.0, and Node 4 (Important Industrial) = 2.0. The sum of the weights of all downstream nodes affected by a fault on each edge is normalized to F. e1 =1.00、F e2 =0.875, F e3 =0.75, F e4 =0.50; The weighted summation according to the above weights yields: T e1 =0.808, T e2 =0.768, T e3 =0.775, T e4 =0.625.

[0067] Assuming there is already a segmented switch at node 1 (the beginning of e1), e1 is excluded. Setting the topological weakness threshold to 0.700, then e2 (T) e2 =0.768) and e3(T e3=0.775) is higher than the threshold, so the line segments containing e2 and e3 are determined as priority candidate segments. Candidate segmentation point p1 (between node 1 and node 2) is generated on e2, and candidate segmentation point p2 (between node 2 and node 3) is generated on e3. Minimum cut set analysis and feeder critical path analysis are performed on the undirected topology graph: Although e4 has a topology weakness T e4 =0.625 is below the threshold, but downstream node 4 of e4 is a top-level critical industrial load (weight 2.0), and e4 is located on the critical path at the end of the feeder. Once e4 fails, node 4 will lose all power and have no relay path. Although the impact range of the failure is small, the consequences are serious. Therefore, candidate segmentation point p3 (between node 3 and node 4) is added to e4. The candidate points generated and added by the directional generation are merged to obtain the candidate segmentation point set {p1, p2, p3}.

[0068] Finally, considering the constraints of line length (the line length between adjacent sectionalizing switches is not less than the preset minimum length), feeder radial structure (the radial operation is maintained after adding sectionalizing switches), and sectionalizing switch constraints (a sectionalizing switch already exists at node 1, so it is not repeated), and limiting the number of newly added sectionalizing switches to no more than 2, the {p1, p2, p3} are combined and arranged as follows: single-point scheme {p1}, {p2}, {p3}; two-point scheme {p1, p2}, {p1, p3}, {p2, p3}; three-point scheme {p1, p2, p3} is eliminated because it exceeds the upper limit of 2 newly added switches. This generates a candidate sectionalizing configuration scheme set Ω={S1, S2, S3, S4, S5, S6}, corresponding to the above six combinations. Each candidate sectionalizing configuration scheme in this set has passed topology weakness screening, critical path supplementation, and engineering constraint verification, serving as the input scheme set for subsequent multi-objective evaluation and optimization solutions.

[0069] By combining topology weakness index screening with minimum cut set analysis and feeder critical path analysis, high-value candidate segmentation points are first generated in a targeted manner. Then, cut set locations and critical path locations that are likely to cause large-scale power outages after a fault are added. This allows the candidate segmentation point set to cover both topology weak sections and critical locations with a large fault impact range. Thus, while ensuring the quality of candidate solutions, the candidate solution space is effectively reduced, and the relevance and completeness of the candidate segmentation configuration solution generation are improved.

[0070] Optionally, in step S30, based on the multi-scenario operation dataset, the multi-objective evaluation index of each candidate segmented configuration scheme in the candidate segmented configuration scheme set under different operation scenarios is calculated, including: obtaining the segmented switch configuration information corresponding to each candidate segmented configuration scheme in the candidate segmented configuration scheme set; based on the node load demand and distributed power output of each node in the target operation scenario in the multi-scenario operation dataset, combined with the segmented switch configuration information, performing power flow calculation on the candidate segmented configuration scheme to obtain the active power loss and line load rate of each line segment in the target operation scenario; based on the line load rate of each line segment in the target operation scenario, combined with the historical fault data corresponding to the target operation scenario, performing fault simulation on the candidate segmented configuration scheme to determine the affected load capacity and power outage duration after the fault of each line segment in the target operation scenario; and calculating the reliability evaluation index, economic evaluation index, and carbon emission evaluation index of the candidate segmented configuration scheme in the target operation scenario according to the affected load capacity, power outage duration, and active power loss of each line segment in the target operation scenario.

[0071] In some possible implementations, based on multi-scenario operational data, the node load demand and distributed power output of each node in the target operational scenario are centralized. Combined with sectional switch configuration information, power flow calculations are performed on candidate sectional configuration schemes to obtain the active power loss and line load rate of each line segment in the target operational scenario. This includes: calculating the net load of each node based on its load demand and distributed power output; determining the connection status of each line segment based on the sectional switch configuration information corresponding to the candidate sectional configuration schemes; solving the AC power flow equations based on the net load of each node, the connection status of each line segment, and the line parameters of the distribution lines to obtain the active power and line load rate of each line segment; and calculating the active power loss of each line segment based on its active power and line resistance. By combining sectional switch configuration information to solve the AC power flow equations, the active power loss and line load rate of each line segment in the target operational scenario can be accurately obtained, providing a reliable power flow data foundation for subsequent fault simulation and multi-objective evaluation index calculations.

[0072] In some possible implementations, based on the line load rate of each line segment under the target operating scenario and combined with the historical fault data corresponding to the target operating scenario, fault simulation is performed on the candidate segment configuration schemes to determine the affected load capacity and power outage duration after a fault in each line segment under the target operating scenario. This includes: determining the corrected fault rate of each line segment based on the baseline fault rate of each line segment in the historical fault data and combined with the line load rate of each line segment under the target operating scenario; determining the set of affected load nodes after fault isolation based on the segment switch configuration information corresponding to the candidate segment configuration scheme for the fault occurrence scenario of each line segment; accumulating the affected load capacity based on the load demand of each node in the set of affected load nodes; and determining the power outage duration of the set of affected load nodes based on the fault location time, fault isolation operation time, and fault recovery time.

[0073] Here, fault location time is the time required from the occurrence of a fault to the determination of the faulty section, fault isolation operation time is the time required to isolate the faulty section by operating the sectionalizing switch, and fault recovery time is the time required to restore power to the non-faulty section through the tie switch or the original power supply path.

[0074] Furthermore, based on the fault location time, fault isolation operation time, and fault recovery time, the outage duration of the affected load node set is determined, including calculating the outage duration according to the following formula: T i = t location + t isolation + t restoration Among them, T i Let t be the duration of the power outage for the i-th affected load node. location t is the fault location time. isolation For the fault isolation operation time, t restoration This refers to the fault recovery time.

[0075] Understandably, for non-faulty sections upstream of the fault, power is restored through the original power supply path, and the power outage duration is the sum of the fault location time and the fault isolation operation time; for non-faulty sections downstream of the fault, if there is a tie switch, the power outage duration must also include the power restoration time through the tie switch; for the faulty section itself, the power outage duration must also include the fault repair time.

[0076] By combining the line load rate to correct the failure rate and simulating the fault isolation process under the action of sectionalizing switches, the affected load capacity and power outage duration after a fault in each line segment can be accurately determined, providing fault consequence data that conforms to actual operating conditions for reliability and economic evaluation.

[0077] In some possible implementations, the reliability evaluation index of the candidate segment configuration scheme under the target operating scenario is calculated based on the affected load capacity of each line segment under the target operating scenario. This includes multiplying the corrected fault rate, the affected load capacity after the fault, and the corresponding power outage duration of each line segment under the target operating scenario, and summing them over all line segments to obtain the reliability evaluation index of the candidate segment configuration scheme under the target operating scenario.

[0078] Furthermore, the reliability evaluation index of the candidate segmentation configuration scheme under the target operating scenario is calculated according to the following formula: in, The reliability evaluation value of candidate segmentation configuration scheme m. Let be the corrected fault rate of the i-th line segment. Let be the duration of the power outage affecting the loads after a fault in the i-th line segment. Let N be the load capacity affected after the fault in the i-th line segment, N be the total number of line segments in the distribution line, and i be the line segment index number, i = 1, 2, …, N.

[0079] By accumulating the product of the failure rate, affected load capacity, and outage duration of each line segment, the comprehensive impact of the candidate segment configuration scheme on the power outage under the target operating scenario can be fully reflected, providing an accurate quantitative basis for reliability for subsequent multi-objective optimization.

[0080] In some possible implementations, the economic evaluation index of the candidate segmented configuration scheme under the target operating scenario is calculated based on the outage duration of each line segment. This includes: determining the equipment investment cost and operation and maintenance cost based on the segmented switch configuration information corresponding to the candidate segmented configuration scheme; determining the line network loss cost based on the active power loss and unit electricity price of each line segment under the target operating scenario; determining the fault outage loss cost based on the affected load capacity and outage duration of each line segment under the target operating scenario, combined with the unit outage loss coefficient; and summing the equipment investment cost, operation and maintenance cost, line network loss cost, and fault outage loss cost to obtain the economic evaluation index of the candidate segmented configuration scheme under the target operating scenario.

[0081] Furthermore, the economic evaluation index of the candidate segmented configuration scheme under the target operating scenario is calculated according to the following formula: in, Let m be the economic evaluation value of the candidate segmentation configuration scheme. For equipment investment costs, For operation and maintenance costs, For network loss costs, Costs incurred due to power outages.

[0082] By uniformly converting and summing various costs such as equipment investment, operation and maintenance, network loss and power outage losses, the economic efficiency of candidate segmented configuration schemes throughout the entire life cycle under the target operating scenario can be fully reflected, providing accurate quantitative input for the economic objective in multi-objective optimization.

[0083] In some possible implementations, the carbon emission evaluation index of the candidate segmented configuration scheme under the target operating scenario is calculated based on the active power loss of each line segment under the target operating scenario. This includes multiplying the active power loss of each line segment under the target operating scenario by the carbon emission factor per unit of electricity and summing the results over all line segments to obtain the carbon emission evaluation index of the candidate segmented configuration scheme under the target operating scenario.

[0084] Furthermore, the carbon emission evaluation index of the candidate segmented configuration scheme under the target operating scenario is calculated according to the following formula: in, The carbon emission evaluation value for candidate segment configuration scheme m. Let t be the active power loss of the line. denoted as the carbon emission factor per unit of electricity, T represents the set of runtime segments under the target operating scenario, and t represents the time index.

[0085] By multiplying the active power loss of each line segment by the carbon emission factor per unit of electricity and summing the results, the indirect carbon emission level caused by line losses under the target operating scenario of the candidate segment configuration scheme can be directly quantified, providing a clear calculation basis for the evaluation of low-carbon targets.

[0086] By combining the segmented switch configuration information corresponding to the candidate segmented configuration schemes with the node load demand and distributed power output in the multi-scenario operation dataset, power flow calculation and fault simulation are performed sequentially. This allows the calculation of reliability evaluation indicators, economic evaluation indicators, and carbon emission evaluation indicators to reflect the actual impact of segmented switch configuration on power flow distribution and fault isolation range, while also covering the load level and renewable energy output status under different operating scenarios. This results in multi-objective evaluation indicators that can truly characterize the comprehensive performance of each candidate segmented configuration scheme under multiple scenario conditions, providing accurate and reliable quantitative basis for subsequent scheme selection and multi-objective optimization.

[0087] Optionally, combined Figure 3 As shown, in step S30, the comprehensive scoring model is used to initially screen each candidate segmentation configuration scheme to obtain a subset of effective candidate segmentation configuration schemes, including the following steps: S301 uses a comprehensive scoring model to obtain the comprehensive evaluation value of each candidate segment configuration scheme based on reliability evaluation index, economic evaluation index and carbon emission evaluation index.

[0088] In some possible implementations, the comprehensive scoring model is constructed as follows: Reliability evaluation indicators, economic evaluation indicators, and carbon emission evaluation indicators of candidate segmented configuration schemes under at least one operating scenario are obtained; the reliability evaluation indicators, economic evaluation indicators, and carbon emission evaluation indicators are normalized respectively to obtain normalized reliability evaluation indicators, normalized economic evaluation indicators, and normalized carbon emission evaluation indicators; preset weight coefficients are assigned to the normalized reliability evaluation indicators, normalized economic evaluation indicators, and normalized carbon emission evaluation indicators respectively; the normalized evaluation indicators are multiplied by their corresponding weight coefficients and then summed to obtain the comprehensive evaluation value of the candidate segmented configuration scheme.

[0089] Furthermore, the comprehensive scoring model is expressed as: Among them, Q m R is the comprehensive evaluation value of candidate line segmentation scheme m. m C m E m These are the normalized reliability evaluation index, economic evaluation index, and low-carbon evaluation index, respectively. ω1, ω2, and ω3 are the weight coefficients corresponding to the normalized reliability evaluation index, economic evaluation index, and low-carbon evaluation index, respectively, and ω1+ω2+ω3=1.

[0090] By normalizing the three types of evaluation indicators and then weighting and summing them to construct a comprehensive scoring model, the multi-objective evaluation results can be unified into a single comparable comprehensive evaluation value, which facilitates the rapid sorting and preliminary screening of candidate segmentation configuration schemes and improves the screening efficiency and operability in the initial screening stage.

[0091] S302, based on the comprehensive evaluation value, each candidate segment configuration scheme in the candidate segment configuration scheme set is screened to obtain an intermediate candidate segment configuration scheme subset.

[0092] In some possible implementations, the candidate segmented configuration schemes in the candidate segmented configuration scheme set are filtered based on the comprehensive evaluation value to obtain an intermediate subset of candidate segmented configuration schemes. This includes: comprehensively processing the comprehensive evaluation values ​​obtained by each candidate segmented configuration scheme in various running scenarios in the multi-scenario running dataset to obtain the final comprehensive evaluation value of the candidate segmented configuration scheme; sorting the candidate segmented configuration schemes in the candidate segmented configuration scheme set according to the final comprehensive evaluation value in ascending order; removing candidate segmented configuration schemes whose final comprehensive evaluation value is higher than a preset threshold from the sorted results; and determining the remaining candidate segmented configuration schemes after removal as the intermediate subset of candidate segmented configuration schemes. By sorting based on the final comprehensive evaluation value and combining it with a preset threshold for filtering, candidate segmented configuration schemes with poor comprehensive performance can be quickly removed while ensuring the basic feasibility of the schemes, effectively reducing the size of candidate schemes and improving the computational efficiency of subsequent multi-objective optimization solutions. S303: During the application of each candidate segment configuration scheme in the intermediate candidate segment configuration scheme subset, the operating status data of the power distribution system is acquired in real time, and the degree of change in the line operating status is determined based on the operating status data.

[0093] In some possible implementations, the degree of change in the line's operating status is determined based on the operating status data, including: obtaining the values ​​of each operating characteristic in the power distribution system before and after the update, where the operating characteristics include node load, line load rate, distributed power output, and failure rate; calculating the absolute value of the rate of change of each operating characteristic before and after the update; and summing the absolute values ​​of the rates of change of each operating characteristic to obtain the degree of change in the line's operating status.

[0094] Furthermore, the degree of change in the line's operating status is calculated using the following formula: in, The degree of change in the line's operating status. The value of the i-th running feature before the update. Let be the updated value of the i-th running feature, N be the total number of running features, and i be the running feature index number, i = 1,2, …, N.

[0095] By quantifying the degree of change in line operating status by accumulating the absolute value of the rate of change of each operating characteristic, the extent to which the operating status of the power distribution system deviates from the original assessment benchmark can be objectively reflected, providing a clear and calculable basis for determining whether dynamic correction and incremental updates should be triggered.

[0096] S304: Based on the degree of change in line operation, the candidate segment configuration schemes in the intermediate candidate segment configuration scheme subset are dynamically corrected and incrementally updated to obtain a subset of effective candidate segment configuration schemes.

[0097] In some possible implementations, the candidate segment configuration schemes in the intermediate candidate segment configuration scheme subset are dynamically corrected and incrementally updated according to the degree of change in line operation to obtain a subset of effective candidate segment configuration schemes. This includes: identifying areas where the degree of change in line operation exceeds a preset threshold as change areas; performing local mutation operations only on candidate segment configuration schemes in the intermediate candidate segment configuration scheme subset that involve change areas and adjacent topology segments of change areas, where mutation operations include adding segment switches in the target topology segment, adjusting the installation position of existing segment switches, and removing redundant segment switches; recalculating and sorting the comprehensive evaluation value only on the mutated local candidate scheme subset, while retaining the candidate segment configuration schemes in the intermediate candidate segment configuration scheme subset that do not involve change areas; and merging the updated local candidate scheme subset with the retained candidate segment configuration schemes that do not involve change areas to obtain a subset of effective candidate segment configuration schemes. By performing local mutations and re-evaluations only on candidate segment configuration schemes for the changed region and its adjacent topology segments, the computational overhead of regenerating and evaluating candidate schemes across the entire line is avoided while ensuring that the scheme subsets dynamically respond to changes in the operating environment. This achieves lightweight incremental updates of candidate scheme subsets.

[0098] The constructed candidate segmentation configuration scheme set For example, S1={p1}, S2={p2}, S3={p3}, S4={p1, p2}, S5={p1, p3}, and S6={p2, p3} illustrate the specific implementation process of using the comprehensive scoring model to perform preliminary screening of each candidate segment configuration scheme and obtain a subset of effective candidate segment configuration schemes.

[0099] Taking a target operating scenario obtained by pairing load growth state L1 (current load level) with distributed power generation output fluctuation state D3 (full output at midday on a sunny day) in a multi-scenario operating dataset as an example, power flow calculations and fault simulations are performed on six candidate schemes to obtain the normalized reliability evaluation index R. m Economic evaluation index C m And carbon emission assessment index E m Taking weighting coefficients ω1=0.4, ω2=0.3, and ω3=0.3, the comprehensive evaluation value of each scheme is calculated according to the comprehensive scoring model, and the results are as follows: S1{0.85(R m ), 0.42 (C m ), 0.35 (E m ), 0.571 (Q) mS2{0.55, 0.50, 0.40, 0.490}; S3{0.48, 0.55, 0.45, 0.492}; S4{0.40, 0.68, 0.48, 0.508}; S5{0.38, 0.72, 0.52, 0.524}; S6{0.35, 0.75, 0.55, 0.530}. Set the preset threshold to Q. th =0.55. Q of S1 m =0.571>0.55, so it is directly eliminated; the remaining schemes have Q m All are below the threshold. Therefore, the subset of intermediate candidate segmentation configuration schemes retained after initial screening is: .

[0100] During the application of the aforementioned intermediate candidate scheme, the operating status data of the power distribution system was continuously monitored. Assuming that the load of node 4 (important industrial load) was 200 kW at the initial evaluation, and the line e4 load factor was 65%; after a period of operation, the load of node 4 was monitored to rise to 320 kW, the line e4 load factor rose to 95%, the photovoltaic output of node 3 did not change significantly, and the failure rate parameter remained unchanged. The absolute values ​​of the rates of change of each operating characteristic were calculated and summed to obtain the degree of change in the line operating status: ΔS = |(320−200) / 200| + |(95%−65%) / 65%| + 0 + 0 = 0.6 + 0.462 ≈ 1.062. The preset trigger threshold was ΔS. th =0.5. Since ΔS=1.062>0.5, the dynamic correction and incremental update process is triggered.

[0101] The load at node 4 increased significantly, and the corresponding line edge e4 and its adjacent topology segment (between node 3 and node 4) became the change area. Only the candidate segment configuration schemes involving the change area and its adjacent topology were subject to local mutation operations. The candidate segment configuration schemes involving e4 in Ω' are S3 (with p3 on e4), S5 (p1+p3), and S6 (p2+p3). These three schemes were mutated to generate local mutation schemes: Mutation scheme A adds one segment switch upstream of node 4 (on e4) and adjusts the original p3 position, forming S3'; Mutation scheme B removes the redundant candidate point p2 in S6, retaining p3, forming S6'. Only the mutated local candidate scheme subset {S3', S6'} was recalculated for power flow, fault simulation, and comprehensive scoring. The Q values ​​for S3' and S6' were found to be 0.461 and 0.488, respectively, both superior to the original schemes. Candidate schemes S2 and S4, which did not involve the change area, maintained their original scores. The updated subset of local candidate schemes {S3', S6'} is merged with the retained candidate schemes {S2, S4} that do not involve the changed area to obtain the effective candidate segmented configuration scheme subset: Ω''={S2, S4, S3', S6'}. This effective candidate segmented configuration scheme subset serves as input for subsequent multi-objective collaborative optimization. Under the condition of satisfying the distribution network operation constraints, Pareto optimization and knee point identification are performed to ultimately determine the optimal distribution line segmented configuration scheme.

[0102] By using a comprehensive scoring model to initially screen each candidate segment configuration scheme, schemes with poor overall performance are quickly eliminated based on the comprehensive evaluation value, thus narrowing down the scale of schemes entering the optimization stage. Then, during the application of the intermediate candidate segment configuration scheme subset, the operating status data of the power distribution system is acquired in real time, and the degree of change in the line operating status is determined. Only candidate segment configuration schemes affected by changes in operating status are dynamically corrected and incrementally updated. In this way, without re-traversing all candidate schemes, the effective candidate segment configuration scheme subset can adapt to changes in the operating status of the power distribution system in a timely manner, balancing screening efficiency and the dynamic adaptability of the scheme set to changes in the operating environment.

[0103] Optionally, in step S40, under the condition of satisfying the distribution network operation constraints, a multi-objective collaborative optimization solution is performed on the subset of effective candidate segmented configuration schemes to obtain the Pareto optimal solution set of the candidate segmented configuration schemes. This includes: establishing a multi-objective optimization model with the minimum of reliability evaluation index, economic evaluation index, and carbon emission evaluation index as the optimization objectives; establishing distribution network operation constraints in the multi-objective optimization model; wherein, the distribution network operation constraints include flexible constraints, and the flexible constraints are fuzzified. When the degree of index exceeding the limit is within the tolerance range, the corresponding candidate segmented configuration scheme is not directly eliminated. Instead, a penalty amount is determined according to the degree of index exceeding the limit, and the penalty amount is superimposed on the objective component of the optimization objective; based on the Pareto optimality principle, the objective component values ​​of each candidate segmented configuration scheme in the subset of effective candidate segmented configuration schemes are non-dominated and sorted to obtain the Pareto optimal solution set composed of mutually non-dominated candidate segmented configuration schemes.

[0104] In some possible implementations, a multi-objective optimization model is established with the minimum of reliability evaluation index, economic evaluation index, and carbon emission evaluation index as the optimization objective. This includes: obtaining the reliability evaluation index, economic evaluation index, and carbon emission evaluation index of each candidate segmented configuration scheme in the subset of effective candidate segmented configuration schemes; and using the reliability evaluation index, economic evaluation index, and carbon emission evaluation index as the three objective components of the multi-objective optimization model, and constructing an objective function with the simultaneous minimization of the three objective components as the objective.

[0105] Furthermore, the objective function is expressed as: Where F is the target vector, R is the reliability evaluation index, C is the economic evaluation index, and E is the carbon emission evaluation index.

[0106] The objective function represents the goal of simultaneously minimizing three objective components—reliability evaluation index R, economic evaluation index C, and carbon emission evaluation index E—during the multi-objective collaborative optimization process. Since these three objectives are interdependent, there is typically no single solution that minimizes all three components simultaneously. Therefore, solving this objective function yields a Pareto optimal solution set to preserve the trade-offs between the objectives and provide a candidate set for subsequent knee point identification and determination of the overall optimal solution.

[0107] In some possible implementations, flexible constraints include voltage operation constraints and line current carrying capacity constraints. Penalties are determined based on the degree of exceedance of the indicators, and these penalties are then superimposed on the target components of the optimization objective. This includes setting fully permissible, tolerable, and absolutely prohibited intervals for both voltage operation constraints and line current carrying capacity constraints. When the voltage operation indicator or line current carrying capacity indicator falls within the corresponding tolerable interval, the degree of exceedance of the voltage operation indicator or line current carrying capacity indicator is determined based on the deviation between the actual value of the voltage operation indicator or line current carrying capacity indicator and the boundary of the fully permissible interval. A penalty is then determined based on the degree of exceedance, with the penalty being positively correlated with the degree of exceedance. The penalty corresponding to the voltage operation constraint is superimposed on the target component of the reliability evaluation indicator, and the penalty corresponding to the line current carrying capacity constraint is superimposed on the target component of the economic evaluation indicator. By setting tolerable intervals for different flexible constraints and calculating corresponding penalties based on the degree of exceedance before superimposing them on the associated target components, it is possible to guide the optimization search towards convergence in the direction of satisfying the constraints while retaining solutions with slight exceedances, thus balancing the diversity of feasible solutions and the comprehensive performance of multi-objective optimization results.

[0108] In other possible implementations, distribution network operation constraints also include hard constraints, such as constraints on the number of sectionalizing switches, the installation location of sectionalizing switches, and the radial topology of the distribution network. Under the condition of satisfying the distribution network operation constraints, a multi-objective collaborative optimization solution is performed on a subset of valid candidate sectionalizing configuration schemes to obtain the Pareto optimal solution set of the candidate sectionalizing configuration schemes. This also includes directly removing candidate sectionalizing configuration schemes that violate the hard constraints from the subset of valid candidate sectionalizing configuration schemes. This ensures that all candidate sectionalizing configuration schemes entering the multi-objective collaborative optimization solution meet the basic conditions of engineering feasibility and safe operation of the distribution network, avoiding ineffective optimization calculations for infeasible schemes.

[0109] In some possible implementations, based on the Pareto optimality principle, the objective component values ​​of each candidate segmented configuration scheme in the subset of effective candidate segmented configuration schemes are non-dominated to obtain a Pareto optimal solution set composed of non-dominated candidate segmented configuration schemes. This includes: for any two candidate segmented configuration schemes in the subset of effective candidate segmented configuration schemes, if the first candidate segmented configuration scheme is not inferior to the second candidate segmented configuration scheme in all objective components and is superior to the second candidate segmented configuration scheme in at least one objective component, then the first candidate segmented configuration scheme is determined to dominate the second candidate segmented configuration scheme; candidate segmented configuration schemes in the subset of effective candidate segmented configuration schemes that are not dominated by any other candidate segmented configuration schemes are determined as non-dominated candidate segmented configuration schemes; and the Pareto optimal solution set is composed of all non-dominated candidate segmented configuration schemes. By comparing the target component values ​​of each candidate segment configuration scheme pairwise and identifying non-dominated candidate segment configuration schemes, we can objectively obtain a Pareto optimal solution set that is not inferior to other schemes in multi-objective trade-offs, providing a complete and non-redundant set of optimal candidate schemes for subsequent knee point identification and final decision-making.

[0110] By establishing a multi-objective optimization model with the minimum of reliability evaluation index, economic evaluation index, and carbon emission evaluation index as optimization objectives, and by fuzzifying the flexible constraints, candidate segmented configuration schemes that slightly exceed the limits are not directly eliminated due to rigid constraints. Instead, a penalty amount positively correlated with the degree of exceeding the limits is applied to the objective components, thereby guiding the optimization process to converge toward satisfying the constraints while preserving the diversity of feasible schemes. On this basis, the candidate segmented configuration schemes are non-dominatedly ranked according to the Pareto optimality principle, so that the conflict relationship between multiple objectives is fully preserved in the Pareto optimal solution set, rather than being artificially weighted and compressed into a single scalar. This provides complete multi-objective trade-off information for the subsequent selection of the comprehensive optimal scheme.

[0111] Optionally, in step S40, knee point identification is performed on the Pareto optimal solution set to determine the final distribution line segment configuration scheme, including: determining the target points corresponding to each candidate segment configuration scheme in the target space, and forming a Pareto front by the target points corresponding to all candidate segment configuration schemes in the Pareto optimal solution set; wherein, the target space consists of three dimensions: reliability evaluation index, economic evaluation index, and carbon emission evaluation index; searching for the turning knee point with diminishing marginal benefits along the Pareto front, and taking the candidate segment configuration scheme corresponding to the turning knee point as the final distribution line segment configuration scheme; wherein, the candidate segment configuration scheme corresponding to the turning knee point is the distribution line segment configuration scheme with the best comprehensive cost-effectiveness among the three factors of reliability evaluation index, economic evaluation index, and carbon emission evaluation index.

[0112] In some possible implementations, the target points corresponding to each candidate piecewise configuration scheme in the Pareto optimal solution set are determined in the target space, and the Pareto front is formed by the target points corresponding to all candidate piecewise configuration schemes in the Pareto optimal solution set. This includes: constructing a three-dimensional target space using reliability evaluation index, economic evaluation index, and carbon emission evaluation index as the first, second, and third coordinate axes, respectively; for each candidate piecewise configuration scheme in the Pareto optimal solution set, the reliability evaluation index value, economic evaluation index value, and carbon emission evaluation index value corresponding to the candidate piecewise configuration scheme are used as the first, second, and third coordinate values ​​of the candidate piecewise configuration scheme in the three-dimensional target space, respectively, to determine the target point corresponding to the candidate piecewise configuration scheme; and the Pareto front is formed by the target points corresponding to all candidate piecewise configuration schemes in the Pareto optimal solution set. By constructing a three-dimensional target space using reliability evaluation indicators, economic evaluation indicators, and carbon emission evaluation indicators as coordinate axes, and mapping each candidate segment configuration scheme as a target point, the distribution and trade-off relationships of candidate schemes in multiple target dimensions can be presented intuitively and without information loss, providing a geometric basis for knee point identification and objective determination of the comprehensive optimal scheme along the Pareto front.

[0113] The following uses the subset of effective candidate segmentation configuration schemes Ω''={S2, S4, S3', S6'} as an example to illustrate the multi-objective collaborative optimization solution and knee point identification process in step S40.

[0114] Assuming that distribution network operation constraints are established in the multi-objective optimization model, the flexible constraints include voltage operation constraints and line current carrying capacity constraints, while the hard constraints include constraints on the number of sectionalizing switches, the installation location of sectionalizing switches, and the radial topology constraints of the distribution network. After fuzzification processing, candidate sectionalizing configuration scheme S2 was detected to have a slight over-limit of line current carrying capacity, but the degree of over-limit falls within the tolerable range, and the penalty is determined to be 0.05. This penalty is added to the economic evaluation index C component of S2. The other schemes S4, S3', and S6' did not exceed the limits and no penalty is required. After hard constraint verification, all candidate sectionalizing configuration schemes meet the constraints on the number of sectionalizing switches, installation location, and radial topology, and no scheme is eliminated. The target component values ​​of each candidate segment configuration scheme after processing are: S2{0.60(R), 0.73(C), 0.52(E)}; S4{0.40, 0.68, 0.48}; S3'{0.45, 0.62, 0.44}; S6'{0.38, 0.72, 0.50}.

[0115] Based on the Pareto optimality principle, each candidate segmented configuration is compared pairwise: Comparing S3' and S2, R (0.45 < 0.60), C (0.62 < 0.73), and E (0.44 < 0.52), S3' is superior to S2 in all three objective components; therefore, S3' dominates S2, and S2 is eliminated. Comparing S4, S3', and S6' pairwise: none of them is superior to the others in all three objectives; they are not mutually dominant. Therefore, the Pareto optimal solution set P = {S3', S4, S6'} is formed by these non-dominant candidate configurations.

[0116] The three candidate segmented configuration schemes in the Pareto optimal solution set P are mapped to a three-dimensional target space with reliability evaluation index R, economic evaluation index C, and carbon emission evaluation index E as coordinate axes, forming a Pareto front. The Pareto fronts are arranged in ascending order of R value: S6' (R=0.38, C=0.72, E=0.50), S4 (R=0.40, C=0.68, E=0.48), and S3' (R=0.45, C=0.62, E=0.44). The marginal benefit changes between adjacent candidate segmented configuration schemes are calculated: from S6' to S4, R sacrifices 0.02, C improves 0.04, and E improves 0.02; a unit sacrifice of R can be exchanged for an improvement of 2.0 in C and 1.0 in E. From S4 to S3': R sacrifices 0.05, C improves 0.06, and E improves 0.04; a unit sacrifice of R can only be exchanged for an improvement of 1.2 in C and 0.8 in E. It is evident that the marginal benefits from S6' to S4 are relatively high, while the marginal benefits from S4 to S3' decrease significantly. This means that the cost of further sacrificing reliability for improved economy and low carbon emissions increases significantly. Therefore, S4 is the inflection point on the Pareto front, and S4 is determined to be the distribution line segmentation configuration scheme with the best overall cost-effectiveness in terms of reliability, economy, and low carbon emissions.

[0117] By mapping each candidate segmented configuration scheme in the Pareto optimal solution set to a target space composed of three dimensions—reliability evaluation index, economic evaluation index, and carbon emission evaluation index—to form a Pareto front, and automatically searching for the inflection point of diminishing marginal benefits along the Pareto front, the determination of the final scheme no longer depends on manually set weight coefficients. Instead, it objectively identifies the candidate segmented configuration scheme with the best overall cost-effectiveness based on the geometric shape of the Pareto front. This avoids the bias of subjective weighting on the multi-objective decision-making results and improves the objectivity and credibility of the final segmented configuration scheme in the sense of multi-objective trade-offs.

[0118] In some possible embodiments, the multi-objective-based distribution line segmentation configuration method further includes performing rolling optimization: applying the final distribution line segmentation configuration scheme to the distribution line and completing the segmentation switch configuration; collecting real-time operating status data of the distribution line, including line power flow, node voltage, line load rate, distributed power output, and line fault status, and constructing a line operating status vector; comparing the actual operation evaluation index after the implementation of the final distribution line segmentation configuration scheme with the corresponding optimization objective evaluation index based on the line operating status vector, and calculating the comprehensive operation deviation; when the comprehensive operation deviation meets the preset rolling optimization triggering condition, updating the multi-scenario operation dataset with the latest collected operating status data, and re-executing the steps of multi-objective evaluation data determination, candidate segmentation configuration scheme generation, preliminary screening, multi-objective collaborative optimization solution, and knee point identification to obtain the updated distribution line segmentation configuration scheme.

[0119] Optionally, the line operating state vector at time t is represented as: in, For line flow, For node voltage, For line load rate, To contribute to distributed power sources This indicates a line fault condition.

[0120] Optionally, based on the line operation state vector, the actual operation evaluation indicators after the implementation of the final distribution line segment configuration scheme are compared with the corresponding optimization target evaluation indicators to calculate the comprehensive operation deviation. This includes: calculating the reliability evaluation indicators, economic evaluation indicators, and carbon emission evaluation indicators of the actual operation after the implementation of the final distribution line segment configuration scheme based on the line operation state vector, and using these as the actual operation evaluation indicators; obtaining the optimization target evaluation indicators corresponding to the final distribution line segment configuration scheme, which are the reliability evaluation indicators, economic evaluation indicators, and carbon emission evaluation indicators corresponding to the final distribution line segment configuration scheme in the multi-objective optimization solution; and multiplying the absolute values ​​of the differences between each actual operation evaluation indicator and the corresponding optimization target evaluation indicator by preset weighting coefficients and then summing them to obtain the comprehensive operation deviation.

[0121] In some possible implementations, the overall operating deviation is calculated according to the following formula: in, To account for overall operational deviations, For the k-th actual operation evaluation indicator, For the k-th optimization objective evaluation index, Let be the weight coefficient corresponding to the k-th evaluation indicator, where k is the index number of the evaluation indicator, k = 1, 2, 3, corresponding to the reliability evaluation indicator, economic evaluation indicator, and carbon emission evaluation indicator, respectively. When the overall operational deviation is small, it indicates that the current line segment configuration scheme can meet the operational requirements; conversely, it indicates that the line operation status has changed significantly, and further adjustments to the line segment configuration scheme are needed.

[0122] Optionally, the preset rolling optimization trigger conditions include: the overall operating deviation being greater than or equal to a preset overall operating deviation threshold; or, the monitored regional load growth exceeding a preset load growth threshold; or, the monitored change in distributed power supply access capacity exceeding a preset capacity change threshold; or, the monitored line failure rate exceeding a preset failure rate threshold. For example, the preset rolling optimization trigger conditions are expressed as follows: in, To account for overall operational deviations, This is the threshold for comprehensive operational deviation.

[0123] By continuously monitoring the line operation status and calculating the comprehensive operation deviation after the optimal segmented configuration scheme is applied, and updating the optimization results with the latest operation data when the trigger conditions are met, the segmented configuration scheme can adaptively adjust to the evolution of the operating environment, such as load growth, changes in new energy access, and changes in failure rate, and maintain the continuous effectiveness of the scheme in the long-term operation process.

[0124] Optionally, the multi-objective-based distribution line segmentation configuration method further includes: after completing rolling optimization, outputting an updated distribution line segmentation configuration scheme, wherein the updated distribution line segmentation configuration scheme includes the installation location of sectionalizing switches, the number of line segments, the operating status of each line segment, and the corresponding reliability evaluation indicators, economic evaluation indicators, and carbon emission evaluation indicators; and generating a line segmentation optimization configuration report based on the updated distribution line segmentation configuration scheme. By outputting an updated scheme including the installation location of sectionalizing switches, the number of line segments, the operating status, and the three types of evaluation indicators, and forming an optimization configuration report, a complete and executable basis for line segmentation adjustment is provided to operation and maintenance personnel. Simultaneously, a benchmark configuration is provided for the implementation of the scheme and continuous monitoring in the next operating cycle, supporting the continuous operation of closed-loop management.

[0125] It should be noted that the specific implementation methods of the above steps are merely illustrative examples of the embodiments of the present invention and are not intended to limit the entirety of the technical solution of the present invention. Those skilled in the art can make adaptive adjustments, substitutions, or combinations of the specific implementation methods in each step based on the above embodiments and according to actual application scenarios and engineering conditions.

[0126] Combination Figure 5 As shown, this embodiment of the invention provides a multi-objective-based distribution line segmentation configuration device 50, including a multi-scenario operation dataset construction module 501, a candidate segmentation configuration scheme generation module 502, a candidate segmentation configuration scheme screening module 503, and a multi-objective collaborative optimization and decision-making module 504. The multi-scenario operation dataset construction module 501 is configured to obtain multi-source operation data of the distribution system where the distribution line is located, identify and correct anomalies in the multi-source operation data, construct a multi-scenario operation dataset, and determine multi-objective evaluation data based on the multi-scenario operation dataset. The candidate segmentation configuration scheme generation module 502 is connected to the multi-objective evaluation vector construction module 501 and is configured to convert the distribution line into an undirected topology graph, construct a topology weakness index based on the multi-objective evaluation data, and generate candidate segmentation points in the undirected topology graph according to the topology weakness index, and then sort the candidate segmentation points. The module generates a set of candidate segmented configuration schemes by combining columns. The candidate segmented configuration scheme filtering module 503 is connected to the candidate segmented configuration scheme generation module 502 and is configured to calculate the multi-objective evaluation index of each candidate segmented configuration scheme in the candidate segmented configuration scheme set under different operating scenarios based on a multi-scenario running dataset. It also uses a comprehensive scoring model to perform preliminary screening of each candidate segmented configuration scheme to obtain a subset of effective candidate segmented configuration schemes. The comprehensive scoring model is established based on the multi-objective evaluation index. The multi-objective collaborative optimization and decision-making module 504 is connected to the candidate segmented configuration scheme filtering module 503 and is configured to perform multi-objective collaborative optimization on the subset of effective candidate segmented configuration schemes under the condition of satisfying the distribution network operation constraints. It obtains the Pareto optimal solution set of the candidate segmented configuration schemes and identifies the knee point of the Pareto optimal solution set to determine the final distribution line segmented configuration scheme.

[0127] Combination Figure 6 As shown, this embodiment of the invention provides a multi-objective power distribution line segmentation configuration device 60, including a processor 601 and a memory 602, and may also include a communication interface 603 and a bus 604. The processor 601, communication interface 603, and memory 602 can communicate with each other via the bus 604. The communication interface 603 can be used for information transmission. The processor 601 can call logical instructions in the memory 602 to execute the multi-objective power distribution line segmentation configuration method of the above embodiment.

[0128] Furthermore, the logic instructions in the aforementioned memory 602 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium.

[0129] The memory 602, as a computer-readable storage medium, can be used to store software programs and computer-executable programs, such as the program instructions / modules corresponding to the methods in the embodiments of the present invention. The processor 601 executes functional applications and data processing by running the program instructions / modules stored in the memory 602, thereby implementing the multi-objective power distribution line segmentation configuration method in the above method embodiments.

[0130] The memory 602 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the terminal device. Furthermore, the memory 602 may include high-speed random access memory and may also include non-volatile memory.

[0131] Processor 601 is a circuit with signal processing capabilities. In one implementation, processor 601 can be a circuit with instruction reading and execution capabilities, such as a central processing unit (CPU), microprocessor, graphics processing unit (GPU) (which can be understood as a type of microprocessor), or digital signal processor (DSP). In another implementation, processor 601 can implement certain functions through the logical relationships of hardware circuits. These logical relationships of hardware circuits are fixed or reconfigurable. For example, processor 601 can be a hardware circuit implemented as an ASIC or a programmable logic device (PLD), such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document and configuring the hardware circuit can be understood as the process of the processor loading instructions to implement the functions of some or all of the above modules. Furthermore, it can also be a hardware circuit designed for artificial intelligence, which can be understood as a type of ASIC, such as a neural network processing unit (NPU), tensor processing unit (TPU), or deep learning processing unit (DPU). The processor 601 is used to execute related programs to implement the functions required by the units in the multi-objective power distribution line segment configuration device of the present invention, or to execute the multi-objective power distribution line segment configuration method of the present invention.

[0132] As can be seen, each module in the above device can be one or more processors (or processing circuits) configured to implement the above methods, such as: CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor types.

[0133] Furthermore, the modules in the above devices can be integrated in whole or in part, or they can be implemented independently. In one implementation, these modules are integrated together as a system-on-a-chip (SOC). The SOC may include at least one processor for implementing any of the above methods or for implementing the functions of the modules of the device. The at least one processor may be of different types, such as CPU and FPGA, CPU and artificial intelligence processor, CPU and GPU, etc.

[0134] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if these modifications and modifications of the invention fall within the scope of the claims and their equivalents, the invention is also intended to include these modifications and modifications.

Claims

1. A method for segmented configuration of power distribution lines based on multiple objectives, characterized in that, include: Obtain multi-source operation data of the power distribution system where the power distribution line is located, identify and correct anomalies in the multi-source operation data, construct a multi-scenario operation dataset, and determine multi-objective evaluation data based on the multi-scenario operation dataset; The power distribution lines are transformed into an undirected topology graph. A topology weakness index is constructed based on multi-objective evaluation data. Candidate segmentation points are generated in the undirected topology graph according to the topology weakness index. The candidate segmentation points are arranged and combined to generate a set of candidate segmentation configuration schemes. Based on a multi-scenario running dataset, multi-objective evaluation indicators for each candidate segmented configuration scheme in the candidate segmented configuration scheme set under different running scenarios are calculated. A comprehensive scoring model is then used to perform preliminary screening of each candidate segmented configuration scheme to obtain a subset of effective candidate segmented configuration schemes. The comprehensive scoring model is established based on the multi-objective evaluation indicators. Under the condition of satisfying the operation constraints of the distribution network, a multi-objective collaborative optimization solution is performed on the subset of effective candidate segment configuration schemes to obtain the Pareto optimal solution set of the candidate segment configuration schemes. Then, the knee point of the Pareto optimal solution set is identified to determine the final distribution line segment configuration scheme.

2. The method for segmented configuration of power distribution lines according to claim 1, characterized in that, Multi-source operational data includes static data of power distribution lines, historical operational data of the power distribution system, and forecast information; Obtain multi-source operational data of the power distribution system where the power distribution lines are located, identify and correct anomalies in the multi-source operational data, and construct a multi-scenario operational dataset, including: Anomalies are identified in historical operational data to determine the abnormal data and the corresponding anomaly type. The abnormal data is then corrected according to the correction method corresponding to the anomaly type to obtain the corrected historical operational data. Based on static data, combined with corrected historical operating data and forecast information, long-term load growth scenario and short-term renewable energy fluctuation scenario are constructed respectively. By pairing the load growth states in the long-term load growth scenario with the output fluctuation states of the distributed power sources in the short-term renewable energy fluctuation scenario, multiple operating scenarios are obtained. In each operating scenario, the node load demand of each node is taken as the node load value under the corresponding load growth state, and the distributed power output of each node is taken as the distributed power output value under the corresponding output fluctuation state. The net load of each node in the operating scenario is calculated based on the node load demand and distributed power output. The node load demand, distributed power output, net load of each node in all operating scenarios, as well as multi-source operating data, are then combined to form a multi-scenario operating dataset.

3. The method for segmented configuration of power distribution lines according to claim 1, characterized in that, Multi-objective evaluation data serves as intermediate data in the process of generating reliability evaluation indicators, economic evaluation indicators, and carbon emission evaluation indicators, including line load rate data and affected load distribution data. A topological weakness index is constructed based on multi-objective evaluation data, including: Transform the power distribution lines into an undirected topology graph, where the undirected topology graph has the power distribution network nodes as the vertex set and the line edges as the edge set. Calculate the edge betweenness of each line edge, where the edge betweenness is the frequency with which all node pairs pass through the line edge in the shortest electrical path; Based on line load rate data, the normalized value of line load rate for each line side is determined, and based on the distribution data of affected loads, the normalized scale of important loads affected by a fault at each line side is determined. The topology weakness index of each line edge is obtained by weighted summing of the edge betweenness, the normalized value of the line load rate, and the normalized scale of the fault affecting important loads.

4. The method for segmented configuration of power distribution lines according to claim 3, characterized in that, Candidate segmentation points are generated directionally in the undirected topology graph based on the topological weakness index. These candidate segmentation points are then arranged and combined to generate a set of candidate segmentation configuration schemes, including: Line segments with topological weakness index values ​​higher than the preset level in the undirected topology graph are identified as priority candidate segments, and candidate segmentation points are generated in a directional manner within the priority candidate segments; Minimum cut set analysis is performed on the undirected topology graph to determine the cut set edge set; where the disconnection of the line edge in the cut set edge set causes some nodes in the undirected topology graph to lose their connection with the power node. Perform feeder critical path analysis on the undirected topology graph to determine the critical path edge set; wherein, the line edges in the critical path edge set are located on the power supply path from the power node to the end load node, and the number of nodes affected after a fault exceeds a preset value. The line segments corresponding to the line edges in the cut set edge set and the critical path edge set are supplemented as candidate segmentation points, and the directional generated candidate segmentation points are merged with the supplemented candidate segmentation points to obtain a candidate segmentation point set; By combining constraints on line length, feeder radiation structure, and segmented switch, the candidate segmented points in the candidate segmented point set are combined and arranged to generate a set of candidate segmented configuration schemes.

5. The method for segmented configuration of power distribution lines according to claim 4, characterized in that, Perform minimum cut set analysis on the undirected topological graph to determine the cut set edge set, including: calculating the edge connectivity contribution value of each path edge in the undirected topological graph. ,in, To remove the edge of the line e forward, and The number of connected paths between them After removing line edge e, and The number of connected paths between them For power nodes, For the i-th load node, The set of load nodes; the line edges with a connectivity contribution value greater than zero are defined as cut set edges, and all cut set edges constitute the cut set edge set; Critical path analysis is performed on the undirected topology graph to determine the set of critical path edges, including: for each power supply path from the power source node to the end load node, calculating the fault impact range index of each line edge on the power supply path. ,in, For load nodes The load weighting factor, For load nodes The load demand, The baseline failure rate for line edge e is... For load nodes Power restoration time The set of load nodes downstream that lose power when a fault occurs at line edge e; the line edges whose fault impact range index exceeds a preset threshold are identified as critical path edges, and the set of critical path edges is composed of all critical path edges.

6. The method for segmented configuration of power distribution lines according to claim 1, characterized in that, Based on a multi-scenario runtime dataset, multi-objective evaluation metrics are calculated for each candidate segmentation configuration scheme in the candidate segmentation configuration scheme set under different runtime scenarios, including: Retrieve the segment switch configuration information corresponding to each candidate segment configuration scheme in the candidate segment configuration scheme set; Based on the node load demand and distributed power output of each node in the target operating scenario, and combined with the segmented switch configuration information, power flow calculation is performed on the candidate segmented configuration schemes to obtain the active power loss and line load rate of each line segment in the target operating scenario. Based on the line load rate of each line segment under the target operating scenario, and combined with the historical fault data corresponding to the target operating scenario, fault simulation is performed on the candidate segment configuration scheme to determine the affected load capacity and power outage duration after the fault of each line segment under the target operating scenario. Based on the affected load capacity, outage duration, and active power loss of each line segment under the target operating scenario, the reliability evaluation index, economic evaluation index, and carbon emission evaluation index of the candidate segment configuration scheme under the target operating scenario are calculated respectively.

7. The method for segmented configuration of power distribution lines according to claim 6, characterized in that, A comprehensive scoring model is used to initially screen each candidate segmentation configuration scheme, obtaining a subset of valid candidate segmentation configuration schemes, including: Based on reliability evaluation indicators, economic evaluation indicators and carbon emission evaluation indicators, a comprehensive evaluation value for each candidate segment configuration scheme is obtained using a comprehensive scoring model. Based on the comprehensive evaluation value, each candidate segment configuration scheme in the candidate segment configuration scheme set is screened to obtain an intermediate candidate segment configuration scheme subset; During the application of each candidate segment configuration scheme in the intermediate candidate segment configuration scheme subset, the operating status data of the power distribution system is acquired in real time, and the degree of change in the line operating status is determined based on the operating status data. Based on the degree of changes in line operation, the candidate segment configuration schemes in the intermediate candidate segment configuration scheme subset are dynamically corrected and incrementally updated to obtain the effective candidate segment configuration scheme subset.

8. The method for segmented configuration of power distribution lines according to any one of claims 1 to 7, characterized in that, Under the constraints of distribution network operation, a multi-objective collaborative optimization solution is performed on a subset of effective candidate segmented configuration schemes to obtain the Pareto optimal solution set of the candidate segmented configuration schemes, including: A multi-objective optimization model is established with the minimum of reliability evaluation index, economic evaluation index and carbon emission evaluation index as the optimization objectives; In a multi-objective optimization model, distribution network operation constraints are established. Among them, distribution network operation constraints include flexible constraints. The flexible constraints are fuzzified. When the degree of exceeding the limit of the indicator is within the tolerance range, the corresponding candidate segment configuration scheme is not directly eliminated. Instead, the penalty amount is determined according to the degree of exceeding the limit of the indicator and then superimposed on the objective component of the optimization objective. Based on the Pareto optimality principle, the objective component values ​​of each candidate segment configuration scheme in the subset of effective candidate segment configuration schemes are sorted in a non-dominated manner to obtain a Pareto optimal solution set composed of mutually non-dominated candidate segment configuration schemes.

9. The method for segmented configuration of power distribution lines according to claim 8, characterized in that, Knee point identification is performed on the Pareto optimal solution set to determine the final distribution line segmentation configuration scheme, including: The target points corresponding to each candidate piecewise configuration scheme in the Pareto optimal solution set are determined in the target space, and the target points corresponding to all candidate piecewise configuration schemes in the Pareto optimal solution set constitute the Pareto front; wherein, the target space consists of three dimensions: reliability evaluation index, economic evaluation index, and carbon emission evaluation index. Search along the Pareto front for inflection knees where marginal benefits decrease, and use the candidate segmentation configuration schemes corresponding to the inflection knees as the final distribution line segmentation configuration schemes. Among them, the candidate segment configuration scheme corresponding to the turning knee point is the distribution line segment configuration scheme with the best comprehensive cost performance among reliability evaluation index, economic evaluation index and carbon emission evaluation index.

10. A multi-objective-based distribution line segmentation configuration device, characterized in that, include: The multi-objective evaluation vector construction module is configured to obtain multi-source operation data of the power distribution system where the power distribution line is located, identify and correct anomalies in the multi-source operation data, construct a multi-scenario operation dataset, and determine multi-objective evaluation data based on the multi-scenario operation dataset. The candidate segment configuration scheme generation module, connected to the multi-objective evaluation vector construction module, is configured to convert the power distribution line into an undirected topology graph, construct a topology weakness index based on multi-objective evaluation data, generate candidate segment points in the undirected topology graph according to the topology weakness index, and arrange and combine the candidate segment points to generate a set of candidate segment configuration schemes. The candidate segmentation configuration scheme filtering module, connected to the candidate segmentation configuration scheme generation module, is configured to calculate the multi-objective evaluation index of each candidate segmentation configuration scheme in the candidate segmentation configuration scheme set under different operating scenarios based on a multi-scenario running dataset, and use a comprehensive scoring model to perform preliminary screening of each candidate segmentation configuration scheme to obtain a subset of effective candidate segmentation configuration schemes; wherein, the comprehensive scoring model is established based on the multi-objective evaluation index; The multi-objective collaborative optimization and decision-making module, connected to the candidate segmented configuration scheme screening module, is configured to perform multi-objective collaborative optimization on a subset of effective candidate segmented configuration schemes under the condition of satisfying the distribution network operation constraints, obtain the Pareto optimal solution set of the candidate segmented configuration schemes, and identify the knee point of the Pareto optimal solution set to determine the final distribution line segmented configuration scheme.

Citation Information

Patent Citations

  • Distributed photovoltaic power distribution network planning method and system based on green certificate transaction mechanism

    CN122418692A