A power distribution network self-healing capability evaluation method and system based on power grid fusion feedback

Abstract

The application discloses a power distribution network self-healing capability evaluation method and system based on power grid fusion feedback, relates to the technical field of power systems, and comprises the following steps: acquiring power distribution network information, obtaining an evaluation basic data set based on data preprocessing according to the power distribution network information, and the like.The application realizes the quantitative characterization of the monitoring coverage degree of secondary equipment on primary equipment by calculating the power distribution section coverage coefficient through the node mapping fusion of primary topology and secondary topology of the power distribution network, realizes the accurate quantification of the electrical coupling and communication correlation characteristics between nodes of the power distribution network by constructing a node correlation coefficient model, realizes the adaptive division of the power distribution network sub-regions through the benchmark coverage coefficient and the comprehensive coverage coefficient, and dynamically determines the power distribution evaluation section quantity of each sub-region by calculating and evaluating the demand coefficient through the regional balance coefficient, so that the evaluation calculation efficiency is greatly improved while the evaluation comprehensiveness is ensured.

Description

Technical Field

[0001] This invention relates to the field of power system technology, specifically to a method and system for assessing the self-healing capability of distribution networks based on power grid integration feedback. Background Technology

[0002] The distribution network is a crucial link in the power system connecting the generation side and the user side, and its power supply reliability directly affects the stable operation of the economy and the guarantee of electricity supply. Self-healing capability, as a core feature of smart distribution networks, enables automatic fault location, isolation, and restoration of power to non-faulty areas after a fault occurs, minimizing the duration and scope of power outages. Accurate and comprehensive assessment of the distribution network's self-healing capability is a core basis for scientifically guiding distribution network planning and construction, optimizing operation and maintenance strategies, and improving the grid's disaster resilience and emergency response capabilities.

[0003] Currently, the assessment of the self-healing capability of distribution networks suffers from several limitations. It cannot accurately analyze the distribution network structure, nor can it adapt to issues such as insufficient flexibility of the primary network structure, relatively lagging secondary system construction, and imperfect coordination mechanisms between primary and secondary systems. Existing technologies focus only on the electrical characteristics and fault isolation capabilities of primary equipment, completely ignoring the crucial supporting role of secondary protection equipment and communication networks in the self-healing process. They only construct the primary topology of the distribution network without integrating the topology of secondary equipment and the mapping relationship between nodes, leading to discrepancies between the assessment model and the actual operating structure of the power grid. Furthermore, they fail to differentiate based on the electrical characteristics, communication quality, and equipment configuration of different regions, often requiring the testing and assessment of each distribution node, resulting in a large data volume and impacting the efficiency of the distribution network self-healing capability assessment. Summary of the Invention

[0004] To address the aforementioned technical problems, this paper provides a method and system for assessing the self-healing capability of distribution networks based on power grid integration feedback. This technical solution solves the problems mentioned in the background technology, such as the inability to accurately analyze the distribution network structure, the inability to adapt to the insufficient flexibility of the primary network structure, the relatively lagging construction of secondary systems, and the imperfect coordination mechanism between primary and secondary systems in traditional distribution networks. Existing technologies only focus on the electrical characteristics and fault isolation capabilities of primary equipment, completely ignoring the key supporting role of secondary protection equipment and communication networks in the self-healing process. They only construct the primary topology of the distribution network without integrating the topology of secondary equipment and the mapping relationship between nodes, resulting in deviations between the assessment model and the actual operating structure of the power grid. Furthermore, they fail to differentiate based on the electrical characteristics, communication quality, and equipment configuration of different regions, often requiring detection and assessment of each distribution node, resulting in a large amount of data and affecting the efficiency of the distribution network self-healing capability assessment.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for assessing the self-healing capability of a distribution network based on power grid integration feedback includes: Obtain distribution network information, which includes distribution network coverage area information and distribution network equipment information; Based on the distribution network information and data preprocessing, a basic dataset for evaluation is obtained; Based on the primary and secondary construction information in the evaluation dataset, a comprehensive topology map is constructed using collaborative analysis. Based on the comprehensive topology diagram, the distribution network is divided into regions to obtain sub-region information; Based on the sub-region information, obtain the calibration node information corresponding to each sub-region; Based on the comprehensive topology map, the topology connection segment between any two adjacent labeled nodes in the sub-region is taken as the power distribution segment, and the power distribution segment information is obtained. Based on the distribution section information, the average value of the coverage coefficient corresponding to the distribution section is used as the coverage coefficient of the sub-region. The self-healing capability of the distribution network is assessed based on sub-regional information and coverage coefficients.

[0006] Preferably, the step of obtaining the evaluation basis dataset based on distribution network information and data preprocessing specifically includes: Based on the distribution network information, primary construction information and secondary construction information are obtained, where primary construction and secondary construction represent different construction stages of the distribution network. Based on the construction information, obtain the corresponding primary equipment, equipment load, switch status and equipment action execution time; Based on the secondary construction information, obtain the protection action signals, fault recording data, terminal online rate and communication delay data of the secondary equipment; Based on the distribution network information, historical fault data is obtained, including fault occurrence time, fault type, fault location, and fault handling time. Based on 3 The criterion is to remove outliers from all collected raw data. The processed data is time-aligned, with a sampling interval of 1 minute, to unify all data onto the same time axis; The time-aligned data is standardized to obtain the basic dataset for evaluation.

[0007] Preferably, the step of constructing a comprehensive topology map based on primary and secondary construction information in the evaluation dataset and through collaborative analysis specifically includes: Based on the primary equipment information in the primary construction information, the equipment is taken as a topology node, and the corresponding first node information and node connection relationship are obtained. Based on the extracted first node information and node connection relationships, a primary topology diagram of the distribution network is constructed. Based on the secondary equipment information in the secondary construction information, the equipment is used as a topology node to extract the second node information and node connection relationship corresponding to the secondary construction. Based on the extracted second node information and node connection relationships, a secondary topology diagram of the distribution network is constructed; Based on the distribution network information, obtain the node mapping relationship between the first node and the second node. The node mapping relationship represents the node connection relationship between the second node and the first node when the distribution network is undergoing secondary construction. Based on the node mapping relationship, the primary topology map and the secondary topology map of the distribution network are merged to obtain a comprehensive topology map.

[0008] Preferably, the step of dividing the distribution network into regions based on the comprehensive topology map and obtaining sub-region information specifically includes: Based on the node mapping relationship, the first node in the integrated topology graph that has a node mapping relationship is taken as the first feature node, and the second node corresponding to the first feature node in the node mapping relationship is taken as the second feature node. Based on the primary equipment information, obtain the switching equipment information therein, wherein the switching equipment is used to disconnect the power distribution network; Based on the switching equipment information, the first node corresponding to the switching equipment is used as the calibration node; Based on the comprehensive topology graph, the topology nodes between any two adjacent calibration nodes are obtained. Based on the topological nodes between the calibrated nodes, the ratio of the number of first feature nodes to the total number of topological nodes between them is used as the coverage coefficient. Based on the calibration nodes, the average of the coverage coefficients of all adjacent calibration nodes is used as the baseline coverage coefficient. The baseline coverage coefficient is used as the weighting coefficient for communication distance, and the weighting coefficient for electrical distance is obtained based on normalization constraints. Based on the power distribution network information, obtain the electrical distance and communication distance between any two topology nodes; Based on the electrical distance, communication distance and corresponding weight coefficient of any two topological nodes, obtain the correlation coefficient between the two topological nodes; Based on the correlation coefficient, the distribution network is divided into regions to obtain sub-region information.

[0009] Preferably, the step of dividing the distribution network into regions based on the correlation coefficient and obtaining sub-region information specifically includes: Take any calibration node as the target node, and take the two adjacent calibration nodes of the target node as the associated calibration nodes of the target node; Based on the coverage coefficients of all adjacent labeled nodes in the integrated topology graph, obtain the standard deviation of the coverage coefficients; The ratio of the standard deviation of the coverage factor to the baseline coverage factor is used as the correlation deviation factor; Set the basic correlation threshold; The product of the difference between 1 and the correlation deviation coefficient and the basic correlation threshold is used as the correlation threshold; Based on the difference in the correlation coefficient between the target node and the associated labeled nodes and the correlation threshold, it is determined whether the associated labeled nodes are of the same type, and node classification information is obtained. If the difference between the correlation coefficients of the target node and the associated calibration node does not exceed the correlation threshold, then the two associated calibration nodes corresponding to the target node are classified as nodes of the same type. Based on the node classification information, the distribution network areas corresponding to the same type of calibration nodes are divided into the same area, and the average correlation coefficient of the same type of calibration nodes is used as the benchmark correlation coefficient of the area to obtain sub-area information.

[0010] Preferably, the assessment of the distribution network's self-healing capability based on sub-regional information and coverage coefficient specifically includes: The regional balance coefficient is obtained by multiplying the baseline correlation coefficient and the coverage coefficient in the sub-region information. The absolute value of the difference between the regional balance coefficient corresponding to each sub-region and 1 is used as the evaluation demand coefficient of that sub-region; Based on the self-healing assessment requirements of the distribution network, a threshold for the number of distribution segments is set, which represents the minimum number of distribution segments allowed during the self-healing assessment of the distribution network. Based on the distribution segment information corresponding to the sub-region, the product of the total number of distribution segments in each sub-region and the assessment demand coefficient is taken as the number of distribution assessment segments in that sub-region. If the number of distribution assessment segments in a sub-region is less than the distribution segment number threshold, then the distribution segment number threshold will be used as the number of distribution assessment segments in that sub-region. Based on the correlation coefficients corresponding to the distribution sections, the distribution sections of the sub-region are sorted in ascending order of correlation coefficients to obtain the distribution section order information. Based on the number of distribution segment assessments and the order of distribution segments, the distribution assessment segments corresponding to each sub-region are selected. Based on the distribution assessment section, obtain the distribution network self-healing score for each sub-region.

[0011] Preferably, the step of obtaining the distribution network self-healing score for each sub-region based on the distribution assessment section specifically includes: Based on the power distribution assessment section, obtain the corresponding power distribution load for the power distribution assessment section; Based on the power distribution network information and the design of the switching equipment, the protection action time, communication delay time, and action execution time corresponding to the switching equipment are obtained. The sum of the protection action time, communication delay time, and action execution time is used as the fault isolation time; Based on the assessment dataset, obtain the historical fault data corresponding to the power distribution assessment section; Based on the historical fault data corresponding to each power distribution assessment section, the proportion of recoverable load corresponding to that power distribution assessment section is obtained; Based on historical fault data, the ratio of the number of historical faults detected to the total number of historical faults for each power distribution assessment section is used as the fault detection rate. Based on the self-healing assessment requirements of the distribution network, weighting coefficients are set for fault isolation time, recoverable load ratio, and fault detection rate. Based on the fault isolation time, recoverable load ratio, fault detection rate and corresponding weighting coefficients, the self-healing score of the distribution network corresponding to each distribution assessment section is obtained by weighted summation. Based on the distribution load corresponding to the distribution assessment section, obtain the distribution network self-healing score for each sub-region.

[0012] Furthermore, a distribution network self-healing capability assessment system based on power grid integration feedback is proposed to implement the assessment method described above, including: The main control module is used to select the topological connection segment between any two adjacent labeled nodes in a sub-region as a power distribution segment based on the comprehensive topology map, obtain power distribution segment information, select the power distribution evaluation segment corresponding to each sub-region based on the power distribution segment evaluation number and power distribution segment order information, and obtain the power distribution network self-healing score of each sub-region based on the power distribution evaluation segment. The information acquisition module is used to acquire distribution network information, which includes distribution network coverage area information and distribution network equipment information. Based on the distribution network information and data preprocessing, the module acquires an evaluation basic dataset. The topology analysis module is used to construct a primary topology map of the distribution network based on the extracted first node information and node connection relationships, construct a secondary topology map of the distribution network based on the extracted second node information and node connection relationships, and merge the primary topology map and the secondary topology map of the distribution network according to the node mapping relationship to obtain a comprehensive topology map. The display module interacts with the main control module and is used to output and display distribution network information, comprehensive topology map, sub-region information, distribution section information, and distribution network self-healing score.

[0013] Optionally, the main control module specifically includes: The control unit is used to select the distribution assessment section corresponding to each sub-region based on the number of distribution section assessments and the distribution section order information, and to obtain the distribution network self-healing score of each sub-region based on the distribution assessment section. An information receiving unit interacts with an information acquisition module and a topology analysis module to receive data and transmit it to the power distribution section unit. The power distribution section unit is used to obtain power distribution section information by taking any two adjacent topological connection segments between labeled nodes in a sub-region as power distribution sections according to the comprehensive topology map.

[0014] Optionally, the information acquisition module specifically includes: The first acquisition unit is used to acquire power distribution network information, which includes power distribution network coverage area information and power distribution network equipment information. The second acquisition unit is used to acquire the evaluation basic dataset based on the distribution network information and data preprocessing.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention proposes a method and system for assessing the self-healing capability of distribution networks based on power grid fusion feedback. By fusing node mapping between the primary and secondary topologies of the distribution network, the coverage coefficient of distribution sections is calculated, enabling a quantitative representation of the monitoring coverage of primary equipment by secondary equipment. By constructing a node correlation coefficient model, the electrical coupling and communication correlation characteristics between distribution network nodes are accurately quantified. Through the benchmark coverage coefficient and the comprehensive coverage coefficient, the adaptive division of distribution network sub-regions is achieved. By calculating the assessment demand coefficient through the regional balance coefficient, the number of distribution assessment sections in each sub-region is dynamically determined, significantly improving the assessment calculation efficiency while ensuring the comprehensiveness of the assessment. Attached Figure Description

[0016] Figure 1 This is a flowchart of a method for assessing the self-healing capability of a distribution network based on power grid fusion feedback, as proposed in this invention. Figure 2 This is a flowchart illustrating the process of obtaining the basic dataset for evaluation in this invention; Figure 3 This is a flowchart of the process for obtaining the comprehensive topology graph in this invention; Figure 4 This is a flowchart of the sub-region information acquisition process in this invention; Figure 5 This is a block diagram of a distribution network self-healing capability assessment system based on power grid fusion feedback proposed in this invention. Detailed Implementation

[0017] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0018] Reference Figure 1 - Figure 4 As shown in the figure, an embodiment of the present invention provides a method for assessing the self-healing capability of a distribution network based on power grid fusion feedback, comprising: Obtain distribution network information, which includes distribution network coverage area information and distribution network equipment information; Based on the distribution network information and data preprocessing, a basic dataset for evaluation is obtained; Specifically, based on distribution network information and data preprocessing, a basic dataset for evaluation is obtained, including: Based on the distribution network information, primary construction information and secondary construction information are obtained, where primary construction and secondary construction represent different construction stages of the distribution network. Based on the construction information, obtain the corresponding primary equipment, equipment load, switch status and equipment action execution time; Based on the secondary construction information, obtain the protection action signals, fault recording data, terminal online rate and communication delay data of the secondary equipment; Based on the distribution network information, historical fault data is obtained, including fault occurrence time, fault type, fault location, and fault handling time. Based on 3 The criterion is to remove outliers from all collected raw data. The processed data is time-aligned, with a sampling interval of 1 minute, to unify all data onto the same time axis; The time-aligned data is standardized to obtain the basic dataset for evaluation.

[0019] This solution comprehensively covers three core dimensions involved in the distribution network's self-healing process: physical layer electrical characteristics, control layer protection communication capabilities, and historical fault evolution patterns. By simultaneously acquiring primary and secondary construction information and historical fault data, it provides comprehensive and systematic data support for subsequent integrated assessments. The 3σ criterion from statistics is used to detect and remove outliers from all collected raw data, automatically identifying and filtering abnormal data points caused by sensor faults, transmission interference, and human input errors. This eliminates the interference of noise data on the assessment results at the source, avoiding assessment bias caused by individual outliers and ensuring the authenticity and reliability of the assessment dataset. Using a 1-minute sampling interval, primary equipment data, secondary equipment data, and historical fault data from different sources and sampling frequencies are unified onto the same timeline, ensuring accurate correspondence between protection action signals, fault waveform data, and electrical quantity changes in the time dimension. This provides an accurate time-series benchmark for subsequent distribution network self-healing assessments. The system standardizes time-aligned data of various types, converting electrical quantities such as voltage, current, and power, communication quantities such as terminal online rate and communication latency, and time quantities such as fault handling time into dimensionless standardized values. This effectively eliminates the differences in dimensions and magnitudes between different types of data, enabling data from different dimensions to be compared, calculated, and fused for analysis on the same scale.

[0020] In this embodiment, the time-aligned data is mapped to the [0, 1] interval using the max-min normalization method.

[0021] Based on the primary and secondary construction information in the evaluation dataset, a comprehensive topology map is constructed using collaborative analysis. Specifically, based on the primary and secondary construction information in the evaluation dataset, a comprehensive topology map is constructed using collaborative analysis, including: Based on the primary equipment information in the primary construction information, the equipment is taken as a topology node, and the corresponding first node information and node connection relationship are obtained. Based on the extracted first node information and node connection relationships, a primary topology diagram of the distribution network is constructed. Based on the secondary equipment information in the secondary construction information, the equipment is used as a topology node to extract the second node information and node connection relationship corresponding to the secondary construction. Based on the extracted second node information and node connection relationships, a secondary topology diagram of the distribution network is constructed; Based on the distribution network information, obtain the node mapping relationship between the first node and the second node. The node mapping relationship represents the node connection relationship between the second node and the first node when the distribution network is undergoing secondary construction. Based on the node mapping relationship, the primary topology map and the secondary topology map of the distribution network are merged to obtain a comprehensive topology map.

[0022] This scheme achieves, for the first time, a complete digital modeling of the "physical layer-control layer" two-layer architecture of the distribution network by constructing independent primary and secondary topology diagrams with primary and secondary equipment as topology nodes respectively. It accurately represents the electrical connection relationships of primary equipment such as transformers, circuit breakers, and lines, and clearly restores the topological connection logic of secondary systems such as protection devices, terminal equipment, and communication networks, providing a complete structural foundation for comprehensively evaluating the self-healing capability of the distribution network. Based on the distribution network construction information, the correspondence between the first and second nodes is extracted, establishing a precise mapping between primary equipment and secondary protection and monitoring equipment. The protection coverage and monitoring objects of each secondary device are clearly defined. This mapping relationship is the core basis for subsequent calculation of the distribution section coverage coefficient and evaluation of the secondary system's support for the self-healing process, enabling accurate quantitative evaluation of the rationality of the secondary system configuration. The independent primary and secondary topologies are organically integrated through node mapping to form a comprehensive topology diagram including electrical connections, protection configurations, and communication associations, ensuring the accuracy of subsequent node analysis.

[0023] Understandably, power distribution networks are often not built in one go, but rather in batches. The first phase of construction primarily involves installing electrical equipment such as transformers, circuit breakers, and lines to establish the basic framework of the distribution network. The second phase, building upon this first phase, selects some of the electrical equipment from the first phase and connects it to protection and monitoring devices. It also connects these protection and monitoring systems to form a protection and monitoring network, enabling the protection and monitoring of the distribution network. Therefore, based on the primary equipment information from the first phase of construction, treating the equipment as topology nodes, and obtaining the corresponding first node information and node connection relationships, essentially represents the information of the electrical equipment during the first phase of construction. The connection relationship between the devices is determined by the secondary equipment information in the secondary construction information. The devices are treated as topology nodes, and the second node information and node connection relationship corresponding to the secondary construction are extracted. This actually represents the connection relationship between the protection and monitoring devices set up during the secondary construction. The node mapping relationship between the first node and the second node is obtained based on the distribution network information. This actually represents the connection relationship between the protection and monitoring devices and the primary equipment during the secondary construction. For example, for electrical equipment A in the primary construction, if secondary equipment a is set up to connect with electrical equipment A during the secondary construction to realize protection and monitoring functions, then there is a mapping relationship between electrical equipment A and secondary equipment a.

[0024] Based on the comprehensive topology diagram, the distribution network is divided into regions to obtain sub-region information; Specifically, based on the comprehensive topology diagram, the distribution network is divided into regions to obtain sub-region information, including: Based on the node mapping relationship, the first node in the integrated topology graph that has a node mapping relationship is taken as the first feature node, and the second node corresponding to the first feature node in the node mapping relationship is taken as the second feature node. Based on the primary equipment information, obtain the switching equipment information therein, wherein the switching equipment is used to disconnect the power distribution network; Based on the switching equipment information, the first node corresponding to the switching equipment is used as the calibration node; Based on the comprehensive topology graph, the topology nodes between any two adjacent calibration nodes are obtained. Based on the topological nodes between the calibrated nodes, the ratio of the number of first feature nodes to the total number of topological nodes between them is used as the coverage coefficient. Based on the calibration nodes, the average of the coverage coefficients of all adjacent calibration nodes is used as the baseline coverage coefficient. The baseline coverage coefficient is used as the weighting coefficient for communication distance, and the weighting coefficient for electrical distance is obtained based on normalization constraints. Based on the power distribution network information, obtain the electrical distance and communication distance between any two topology nodes; Based on the electrical distance, communication distance and corresponding weight coefficient of any two topological nodes, obtain the correlation coefficient between the two topological nodes; Based on the correlation coefficient, the distribution network is divided into regions to obtain sub-region information.

[0025] In this scheme, the first and second characteristic nodes associated with primary and secondary equipment are accurately identified through node mapping relationships. Switching equipment, which performs the core function of fault isolation, is used as the calibration node, ensuring that the divided distribution sections naturally possess the physical basis for independent fault isolation and power restoration, providing a logically consistent basic unit for subsequent evaluation. The proportion of the first characteristic node within the distribution section is defined as the coverage coefficient, intuitively reflecting the configuration density and coverage of secondary equipment in different sections, providing a core quantitative basis for evaluating the supporting role of the communication and protection system in the self-healing process. The benchmark coverage coefficient is used as the weighting coefficient for communication distance, achieving dynamic adaptive adjustment of the weight according to the secondary configuration level of the area: in areas with good secondary coverage, the weight of communication distance automatically increases; in areas with weak secondary coverage, the weight of electrical distance increases accordingly. This allows the correlation coefficient calculation to accurately reflect the actual relative importance of electrical coupling and communication support in different areas, significantly improving the rationality of area division. By integrating the electrical distance and communication distance between topology nodes, a multi-dimensional correlation coefficient model is constructed, ensuring that the divided sub-regions not only have close electrical connections but also reliable communication guarantees, enabling them to independently complete the entire process of fault self-healing.

[0026] Understandably, node mapping relationship refers to the monitoring / protection correspondence between primary and secondary equipment in a distribution network. That is, it refers to the relationship between the primary equipment objects monitored, protected, or controlled by each secondary device (such as protection devices, monitoring and control terminals, FTUs / DTUs, etc.). Therefore, the first node (i.e., the first characteristic node) with a node mapping relationship refers to a primary equipment node equipped with corresponding secondary monitoring / protection equipment. The baseline coverage coefficient reflects the overall configuration level of all secondary equipment, and the secondary configuration level determines the relative importance of the communication network in the self-healing process. When the secondary configuration level is high (large baseline coverage coefficient), the communication network is generally complete, and the communication distance has a more significant impact on node association; the association analysis should focus more on the communication dimension. When the secondary configuration level is low (small baseline coverage coefficient), the communication network is weak, and electrical coupling characteristics remain the dominant factor determining node association; the association analysis should focus more on the electrical dimension. Based on normalization constraints, the sum of the communication distance weight coefficient and the electrical distance weight coefficient is 1. The difference between 1 and the baseline coverage coefficient is used as the electrical distance weight coefficient. It is important to note that secondary equipment (FTU, DTU, protection devices, etc.) is itself a terminal node in the communication network. Each secondary device needs to connect to the communication network to upload data and receive control commands. Therefore, a higher coverage density of secondary equipment (a larger baseline coverage coefficient) means a denser network of communication nodes, a more complete communication network topology, and a more reliable communication link quality. Conversely, areas with low secondary equipment coverage density also have relatively weak communication network infrastructure. In distribution network engineering practice, the construction of secondary systems and communication networks are planned and implemented simultaneously; that is, where there is secondary equipment, there will inevitably be communication coverage, and where there is no secondary equipment, communication nodes will not be built separately. Therefore, the baseline coverage coefficient is essentially an indirect representation of the completeness of the communication network. Using the baseline coverage coefficient as a weighting coefficient for communication distance enables adaptive and dynamic adjustment of the weight. For example, for urban core area distribution networks (high secondary coverage, such as 0.8~0.9), the communication weight is high, the correlation coefficient reflects the communication correlation characteristics more, and the communication guarantee of the divided sub-areas is more reliable; for rural distribution networks (low secondary coverage, such as 0.3~0.5), the electrical weight is high, the correlation coefficient reflects the electrical coupling characteristics more, and the electrical connection of the divided sub-areas is closer.

[0027] Specifically, based on the correlation coefficient, the distribution network is divided into regions to obtain sub-region information, including: Take any calibration node as the target node, and take the two adjacent calibration nodes of the target node as the associated calibration nodes of the target node; Based on the coverage coefficients of all adjacent labeled nodes in the integrated topology graph, obtain the standard deviation of the coverage coefficients; The ratio of the standard deviation of the coverage factor to the baseline coverage factor is used as the correlation deviation factor; Set the basic correlation threshold; The product of the difference between 1 and the correlation deviation coefficient and the basic correlation threshold is used as the correlation threshold; Based on the difference in the correlation coefficient between the target node and the associated labeled nodes and the correlation threshold, it is determined whether the associated labeled nodes are of the same type, and node classification information is obtained. If the difference between the correlation coefficients of the target node and the associated calibration node does not exceed the correlation threshold, then the two associated calibration nodes corresponding to the target node are classified as nodes of the same type. Based on the node classification information, the distribution network areas corresponding to the same type of calibration nodes are divided into the same area, and the average correlation coefficient of the same type of calibration nodes is used as the benchmark correlation coefficient of the area to obtain sub-area information.

[0028] In this scheme, by calculating the baseline coverage coefficient and the standard deviation of the coverage coefficient, an association threshold adapted to the characteristics of different distribution networks is generated. It can automatically adjust the division standard according to the secondary equipment configuration level, communication network quality and electrical topology of different regions, ensuring that reasonable division results can be obtained in both urban distribution networks with good secondary coverage and rural distribution networks with poor secondary coverage. By calculating the association deviation coefficient, the degree of deviation between the secondary system configuration of different sections and the global average level is accurately quantified, so that the regional division results can accurately reflect the spatial distribution differences of the synergy between primary and secondary systems, providing a core quantitative basis for subsequent accurate location of self-healing weak links. In this embodiment, the basic association threshold is 0.6.

[0029] Based on the sub-region information, obtain the calibration node information corresponding to each sub-region; Based on the comprehensive topology map, the topology connection segment between any two adjacent labeled nodes in the sub-region is taken as the power distribution segment, and the power distribution segment information is obtained. Based on the distribution section information, the average value of the coverage coefficient corresponding to the distribution section is used as the coverage coefficient of the sub-region. The self-healing capability of the distribution network is assessed based on sub-regional information and coverage coefficients.

[0030] Specifically, based on sub-regional information and coverage coefficients, an assessment of the distribution network's self-healing capability is conducted, including: The regional balance coefficient is obtained by multiplying the baseline correlation coefficient and the coverage coefficient in the sub-region information. The absolute value of the difference between the regional balance coefficient corresponding to each sub-region and 1 is used as the evaluation demand coefficient of that sub-region; Based on the self-healing assessment requirements of the distribution network, a threshold for the number of distribution segments is set, which represents the minimum number of distribution segments allowed during the self-healing assessment of the distribution network. Based on the distribution segment information corresponding to the sub-region, the product of the total number of distribution segments in each sub-region and the assessment demand coefficient is taken as the number of distribution assessment segments in that sub-region. If the number of distribution assessment segments in a sub-region is less than the distribution segment number threshold, then the distribution segment number threshold will be used as the number of distribution assessment segments in that sub-region. Based on the correlation coefficients corresponding to the distribution sections, the distribution sections of the sub-region are sorted in ascending order of correlation coefficients to obtain the distribution section order information. Based on the number of distribution segment assessments and the order of distribution segments, the distribution assessment segments corresponding to each sub-region are selected. Based on the distribution assessment section, obtain the distribution network self-healing score for each sub-region.

[0031] In this scheme, a regional balance coefficient is constructed by multiplying the baseline correlation coefficient and the coverage coefficient. This deeply couples the electrical-communication coordination characteristics of the distribution network with the coverage of secondary equipment, forming a comprehensive quantitative index that fully reflects the matching level of the primary and secondary systems in a sub-region. The absolute value of the difference between the regional balance coefficient and the ideal value of 1 is used as the assessment demand coefficient, establishing a positive correlation mechanism between assessment resource input and regional weakness. Sub-regions with poorer self-healing foundations and lower primary-secondary coordination receive higher assessment demand coefficients and more assessment resources; conversely, assessment input is appropriately reduced. This significantly improves assessment efficiency while ensuring comprehensiveness.

[0032] Specifically, based on the distribution assessment section, the self-healing score of the distribution network in each sub-region is obtained, including: Based on the power distribution assessment section, obtain the corresponding power distribution load for the power distribution assessment section; Based on the power distribution network information and the design of the switching equipment, the protection action time, communication delay time, and action execution time corresponding to the switching equipment are obtained. The sum of the protection action time, communication delay time, and action execution time is used as the fault isolation time; Based on the assessment dataset, obtain the historical fault data corresponding to the power distribution assessment section; Based on the historical fault data corresponding to each power distribution assessment section, the proportion of recoverable load corresponding to that power distribution assessment section is obtained; Based on historical fault data, the ratio of the number of historical faults detected to the total number of historical faults for each power distribution assessment section is used as the fault detection rate. Based on the self-healing assessment requirements of the distribution network, weighting coefficients are set for fault isolation time, recoverable load ratio, and fault detection rate. Based on the fault isolation time, recoverable load ratio, fault detection rate and corresponding weighting coefficients, the self-healing score of the distribution network corresponding to each distribution assessment section is obtained by weighted summation. Based on the distribution load corresponding to the distribution assessment section, obtain the distribution network self-healing score for each sub-region.

[0033] This scheme integrates three core indicators—fault isolation time, recoverable load ratio, and fault detection rate—to consider both the speed of the self-healing process and the accuracy of fault detection and the completeness of load restoration, thus comprehensively and objectively reflecting the actual self-healing capability of the distribution network. By using the load value of the distribution assessment section as a weight to calculate the comprehensive self-healing score of the sub-region, it achieves a precise match between the assessment results and the importance of regional power supply. Sections with larger loads and higher power supply reliability requirements have a greater impact on the overall score of the sub-region, avoiding the unreasonable phenomenon of "lightly loaded sections lowering the score of heavily loaded sections" in the traditional average scoring method. This mechanism enables the assessment results to accurately reflect the power supply risks of different regions, providing a scientific basis for priority upgrades under resource-constrained conditions.

[0034] In this embodiment, the weighting coefficient for fault isolation time is 0.45, the weighting coefficient for recoverable load ratio is 0.35, and the weighting coefficient for fault detection rate is 0.2. It can be understood that the primary goal of distribution network self-healing is to quickly restore power supply. Fault isolation time directly determines the outage duration in non-faulty areas and is the most critical factor affecting users' power supply reliability experience. The recoverable load ratio directly reflects the number of users and load scale that can be restored after a fault occurs, determining the scope of the outage impact. One of the core advantages of primary and secondary integrated distribution networks is the ability to restore more loads through flexible network reconfiguration, and this indicator effectively reflects this advantage. The fault detection rate is a prerequisite for initiating the self-healing process; only by accurately detecting faults can subsequent location, isolation, and restoration operations be performed.

[0035] The specific self-healing score of the distribution network in each sub-region is as follows: In the formula, Assess the self-healing rating of the distribution network in the sub-region. Indicates the first The comprehensive weight of each power distribution assessment section, Indicates the first Distribution network self-healing score for each distribution assessment section Indicates the first The correlation characteristic coefficients of each power distribution assessment section Indicates the first Load weight of each power distribution assessment section Indicates the first The correlation coefficient of each power distribution assessment section Represents the baseline correlation coefficient of the sub-region. Indicates the first The load of each power distribution assessment section Indicates the first The load of each power distribution section The total number of distribution assessment sections, This represents the total number of power distribution sections in the sub-region.

[0036] It is important to note that the evaluation object of this invention is the self-healing capability of the distribution network, not the coverage rate of secondary equipment. The distribution and coverage of secondary equipment is only one of the many input parameters of this invention, just as a doctor needs to take a patient's temperature, but the temperature measurement itself is not a diagnosis. The final output of this invention is a distribution network self-healing score, which integrates three major indicators: fault isolation time, recoverable load ratio, and fault detection rate, and incorporates information from multiple dimensions such as electrical characteristics, communication quality, and historical faults, far exceeding what "secondary equipment coverage rate" can encompass. The self-healing capability of the distribution network is itself a complex, multi-dimensional system problem. The self-healing capability of the distribution network refers to the ability of the power grid to automatically complete fault location, isolation, and restoration of power supply to non-faulty areas after a fault occurs. This capability depends on topology, line impedance, switch configuration, load distribution, and the configuration and coverage of protection devices, monitoring and control terminals, fault indicators, etc. If only the distribution and coverage of secondary equipment are statistically analyzed, it can only answer "how many secondary devices there are," but cannot answer "how strong is the self-healing capability." For example, consider two distribution networks with the same secondary coverage rate. One network has a robust structure, excellent communication, and few historical faults, while the other has a weak structure, unstable communication, and frequent faults. Their self-healing capabilities will obviously differ drastically. Simple coverage rate statistics cannot distinguish this difference. The technical contribution of this invention lies precisely in transcending simple coverage rate statistics. In existing technologies, distribution network self-healing assessment does indeed have a simplistic tendency to "only look at secondary coverage rate," leading to a significant deviation between the assessment results and the actual self-healing capabilities. One of the core improvements of this invention is to overcome the limitations of a single coverage rate indicator. Through adaptive regional division and differentiated assessment granularity, it achieves optimized allocation of assessment resources, resulting in overall efficiency higher than simpler methods.

[0037] Weak areas (high assessment demand coefficient): Allocate more assessment sections to ensure accuracy. High-quality areas (low assessment demand coefficient): allocate fewer assessment segments to save computational load. This is precisely a technological improvement to address the distortion in evaluation caused by "simplifying simple problems," rather than complicating simple problems.

[0038] Reference Figure 5 As shown, further, combining the above-mentioned distribution network self-healing capability assessment method based on power grid fusion feedback, a distribution network self-healing capability assessment system based on power grid fusion feedback is proposed, including: The main control module is used to select the topological connection segment between any two adjacent labeled nodes in a sub-region as a power distribution segment based on the comprehensive topology map, obtain power distribution segment information, select the power distribution evaluation segment corresponding to each sub-region based on the power distribution segment evaluation number and power distribution segment order information, and obtain the power distribution network self-healing score of each sub-region based on the power distribution evaluation segment. The information acquisition module is used to acquire distribution network information, which includes distribution network coverage area information and distribution network equipment information. Based on the distribution network information and data preprocessing, the module acquires an evaluation basic dataset. The topology analysis module is used to construct a primary topology map of the distribution network based on the extracted first node information and node connection relationships, construct a secondary topology map of the distribution network based on the extracted second node information and node connection relationships, and merge the primary topology map and the secondary topology map of the distribution network according to the node mapping relationship to obtain a comprehensive topology map. The display module interacts with the main control module and is used to output and display distribution network information, comprehensive topology map, sub-region information, distribution section information, and distribution network self-healing score.

[0039] The main control module specifically includes: The control unit is used to select the distribution assessment section corresponding to each sub-region based on the number of distribution section assessments and the distribution section order information, and to obtain the distribution network self-healing score of each sub-region based on the distribution assessment section. An information receiving unit interacts with an information acquisition module and a topology analysis module to receive data and transmit it to the power distribution section unit. The power distribution section unit is used to obtain power distribution section information by taking any two adjacent topological connection segments between labeled nodes in a sub-region as power distribution sections according to the comprehensive topology map.

[0040] The information acquisition module specifically includes: The first acquisition unit is used to acquire power distribution network information, which includes power distribution network coverage area information and power distribution network equipment information. The second acquisition unit is used to acquire the evaluation basic dataset based on the distribution network information and data preprocessing.

[0041] In summary, the advantages of this invention are as follows: By fusing node mapping between the primary and secondary topologies of the distribution network, a comprehensive topology map containing the relationships between primary and secondary equipment is constructed, achieving complete modeling of the physical structure and control system of the distribution network; by calculating the distribution section coverage coefficient through the proportion of the first characteristic node, a quantitative representation of the monitoring coverage of primary equipment by secondary equipment is achieved; by integrating the weighted calculation of electrical distance and communication distance, a node correlation coefficient model is constructed, achieving accurate quantification of the electrical coupling and communication correlation characteristics between nodes in the distribution network; by calculating the dynamic correlation threshold through the deviation between the benchmark coverage coefficient and the comprehensive coverage coefficient, adaptive division of sub-regions of the distribution network is achieved; by calculating the evaluation demand coefficient through the regional balance coefficient, the number of distribution evaluation sections in each sub-region is dynamically determined, achieving differentiated evaluation granularity allocation, and significantly improving the evaluation calculation efficiency while ensuring the comprehensiveness of the evaluation.

[0042] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A method for assessing the self-healing capability of a distribution network based on power grid fusion feedback, characterized in that, include: Obtain distribution network information, which includes distribution network coverage area information and distribution network equipment information; Based on the distribution network information and data preprocessing, a basic dataset for evaluation is obtained; Based on the primary and secondary construction information in the evaluation dataset, a comprehensive topology map is constructed using collaborative analysis. Based on the comprehensive topology diagram, the distribution network is divided into regions to obtain sub-region information; Based on the sub-region information, obtain the calibration node information corresponding to each sub-region; Based on the comprehensive topology map, the topology connection segment between any two adjacent labeled nodes in the sub-region is taken as the power distribution segment, and the power distribution segment information is obtained. Based on the distribution section information, the average value of the coverage coefficient corresponding to the distribution section is used as the coverage coefficient of the sub-region. The self-healing capability of the distribution network is assessed based on sub-regional information and coverage coefficients.

2. The method for assessing the self-healing capability of a distribution network based on power grid fusion feedback as described in claim 1, characterized in that, The step of obtaining the basic dataset for evaluation based on distribution network information and data preprocessing specifically includes: Based on the distribution network information, primary construction information and secondary construction information are obtained, where primary construction and secondary construction represent different construction stages of the distribution network. Based on the construction information, obtain the corresponding primary equipment, equipment load, switch status and equipment action execution time; Based on the secondary construction information, obtain the protection action signals, fault recording data, terminal online rate and communication delay data of the secondary equipment; Based on the distribution network information, historical fault data is obtained, including fault occurrence time, fault type, fault location, and fault handling time. Based on 3 The criterion is to remove outliers from all collected raw data. The processed data is time-aligned, with a sampling interval of 1 minute, to unify all data onto the same time axis; The time-aligned data is standardized to obtain the basic dataset for evaluation.

3. The method for assessing the self-healing capability of a distribution network based on power grid fusion feedback according to claim 2, characterized in that, The step of constructing a comprehensive topology map based on primary and secondary construction information in the evaluation dataset and through collaborative analysis includes: Based on the primary equipment information in the primary construction information, the equipment is taken as a topology node, and the corresponding first node information and node connection relationship are obtained. Based on the extracted first node information and node connection relationships, a primary topology diagram of the distribution network is constructed. Based on the secondary equipment information in the secondary construction information, the equipment is used as a topology node to extract the second node information and node connection relationship corresponding to the secondary construction. Based on the extracted second node information and node connection relationships, a secondary topology diagram of the distribution network is constructed; Based on the distribution network information, obtain the node mapping relationship between the first node and the second node. The node mapping relationship represents the node connection relationship between the second node and the first node when the distribution network is undergoing secondary construction. Based on the node mapping relationship, the primary topology map and the secondary topology map of the distribution network are merged to obtain a comprehensive topology map.

4. The method for assessing the self-healing capability of a distribution network based on power grid fusion feedback according to claim 3, characterized in that, The step of dividing the distribution network into regions based on the comprehensive topology map and obtaining sub-region information specifically includes: Based on the node mapping relationship, the first node in the integrated topology graph that has a node mapping relationship is taken as the first feature node, and the second node corresponding to the first feature node in the node mapping relationship is taken as the second feature node. Based on the primary equipment information, obtain the switching equipment information therein, wherein the switching equipment is used to disconnect the power distribution network; Based on the switching equipment information, the first node corresponding to the switching equipment is used as the calibration node; Based on the comprehensive topology graph, the topology nodes between any two adjacent calibration nodes are obtained. Based on the topological nodes between the calibrated nodes, the ratio of the number of first feature nodes to the total number of topological nodes between them is used as the coverage coefficient. Based on the calibration nodes, the average of the coverage coefficients of all adjacent calibration nodes is used as the baseline coverage coefficient. The baseline coverage coefficient is used as the weighting coefficient for communication distance, and the weighting coefficient for electrical distance is obtained based on normalization constraints. Based on the power distribution network information, obtain the electrical distance and communication distance between any two topology nodes; Based on the electrical distance, communication distance and corresponding weight coefficient of any two topological nodes, obtain the correlation coefficient between the two topological nodes; Based on the correlation coefficient, the distribution network is divided into regions to obtain sub-region information.

5. The method for assessing the self-healing capability of a distribution network based on power grid fusion feedback according to claim 4, characterized in that, The step of dividing the distribution network into regions based on correlation coefficients and obtaining sub-region information specifically includes: Take any calibration node as the target node, and take the two adjacent calibration nodes of the target node as the associated calibration nodes of the target node; Based on the coverage coefficients of all adjacent labeled nodes in the integrated topology graph, obtain the standard deviation of the coverage coefficients; The ratio of the standard deviation of the coverage factor to the baseline coverage factor is used as the correlation deviation factor; Set the basic correlation threshold; The product of the difference between 1 and the correlation deviation coefficient and the basic correlation threshold is used as the correlation threshold; Based on the difference in the correlation coefficient between the target node and the associated labeled nodes and the correlation threshold, it is determined whether the associated labeled nodes are of the same type, and node classification information is obtained. If the difference between the correlation coefficients of the target node and the associated calibration node does not exceed the correlation threshold, then the two associated calibration nodes corresponding to the target node are classified as nodes of the same type. Based on the node classification information, the distribution network areas corresponding to the same type of calibration nodes are divided into the same area, and the average correlation coefficient of the same type of calibration nodes is used as the benchmark correlation coefficient of the area to obtain sub-area information.

6. The method for assessing the self-healing capability of a distribution network based on power grid fusion feedback according to claim 5, characterized in that, The assessment of the distribution network's self-healing capability based on sub-regional information and coverage coefficients specifically includes: The regional balance coefficient is obtained by multiplying the baseline correlation coefficient and the coverage coefficient in the sub-region information. The absolute value of the difference between the regional balance coefficient corresponding to each sub-region and 1 is used as the evaluation demand coefficient of that sub-region; Based on the self-healing assessment requirements of the distribution network, a threshold for the number of distribution segments is set, which represents the minimum number of distribution segments allowed during the self-healing assessment of the distribution network. Based on the distribution segment information corresponding to the sub-region, the product of the total number of distribution segments in each sub-region and the assessment demand coefficient is taken as the number of distribution assessment segments in that sub-region. If the number of distribution assessment segments in a sub-region is less than the distribution segment number threshold, then the distribution segment number threshold will be used as the number of distribution assessment segments in that sub-region. Based on the correlation coefficients corresponding to the distribution sections, the distribution sections of the sub-region are sorted in ascending order of correlation coefficients to obtain the distribution section order information. Based on the number of distribution segment assessments and the order of distribution segments, the distribution assessment segments corresponding to each sub-region are selected. Based on the distribution assessment section, obtain the distribution network self-healing score for each sub-region.

7. The method for assessing the self-healing capability of a distribution network based on power grid fusion feedback according to claim 6, characterized in that, The process of obtaining the self-healing score of the distribution network for each sub-region based on the distribution assessment section specifically includes: Based on the power distribution assessment section, obtain the corresponding power distribution load for the power distribution assessment section; Based on the power distribution network information and the design of the switching equipment, the protection action time, communication delay time, and action execution time corresponding to the switching equipment are obtained. The sum of the protection action time, communication delay time, and action execution time is used as the fault isolation time; Based on the assessment dataset, obtain the historical fault data corresponding to the power distribution assessment section; Based on the historical fault data corresponding to each power distribution assessment section, the proportion of recoverable load corresponding to that power distribution assessment section is obtained; Based on historical fault data, the ratio of the number of historical faults detected to the total number of historical faults for each power distribution assessment section is used as the fault detection rate. Based on the self-healing assessment requirements of the distribution network, weighting coefficients are set for fault isolation time, recoverable load ratio, and fault detection rate. Based on the fault isolation time, recoverable load ratio, fault detection rate and corresponding weighting coefficients, the self-healing score of the distribution network corresponding to each distribution assessment section is obtained by weighted summation. Based on the distribution load corresponding to the distribution assessment section, obtain the distribution network self-healing score for each sub-region.

8. A distribution network self-healing capability assessment system based on power grid fusion feedback, used to implement the assessment method as described in any one of claims 1-7, characterized in that, include: The main control module is used to select the topological connection segment between any two adjacent labeled nodes in a sub-region as a power distribution segment based on the comprehensive topology map, obtain power distribution segment information, select the power distribution evaluation segment corresponding to each sub-region based on the power distribution segment evaluation number and power distribution segment order information, and obtain the power distribution network self-healing score of each sub-region based on the power distribution evaluation segment. The information acquisition module is used to acquire distribution network information, which includes distribution network coverage area information and distribution network equipment information. Based on the distribution network information and data preprocessing, the module acquires an evaluation basic dataset. The topology analysis module is used to construct a primary topology map of the distribution network based on the extracted first node information and node connection relationships, construct a secondary topology map of the distribution network based on the extracted second node information and node connection relationships, and merge the primary topology map and the secondary topology map of the distribution network according to the node mapping relationship to obtain a comprehensive topology map. The display module interacts with the main control module and is used to output and display distribution network information, comprehensive topology map, sub-region information, distribution section information, and distribution network self-healing score.

9. The distribution network self-healing capability assessment system based on power grid fusion feedback according to claim 8, characterized in that, The main control module specifically includes: The control unit is used to select the distribution assessment section corresponding to each sub-region based on the number of distribution section assessments and the distribution section order information, and to obtain the distribution network self-healing score of each sub-region based on the distribution assessment section. An information receiving unit interacts with an information acquisition module and a topology analysis module to receive data and transmit it to the power distribution section unit. The power distribution section unit is used to obtain power distribution section information by taking any two adjacent topological connection segments between labeled nodes in a sub-region as power distribution sections according to the comprehensive topology map.

10. The distribution network self-healing capability assessment system based on power grid fusion feedback according to claim 8, characterized in that, The information acquisition module specifically includes: The first acquisition unit is used to acquire power distribution network information, which includes power distribution network coverage area information and power distribution network equipment information. The second acquisition unit is used to acquire the evaluation basic dataset based on the distribution network information and data preprocessing.

Images