Distribution network operation and maintenance grid division method for intelligent precise operation and maintenance

Abstract

The invention discloses a distribution network operation and maintenance grid division method for intelligent precise operation and maintenance, which improves typicality and high efficiency of operation and maintenance grids, reduces operation and maintenance time cost and path redundancy, and comprises a distribution network operation and maintenance grid preliminary division method and a distribution network operation and maintenance grid division optimization method, the distribution network operation and maintenance grid preliminary division method comprises the following steps that S1.1, a target distribution network solution forms different types of plot units, and a unique identification label is distributed to each plot unit; s1.2, on the basis of the global shortest path principle, a topological relation in all plot unit identification labels in the target power distribution network is considered, and a preliminary minimum distribution network operation and maintenance grid optimization model is constructed; s1.3, solving the preliminary minimum distribution network operation and maintenance grid optimization model to obtain a preliminary grid set, and counting remaining units which are not included in the preliminary grid; according to the distribution network operation and maintenance grid division optimization method, the compromise similarity is firstly calculated, so that the optimal combination of the remaining units is realized according to the compromise similarity.

Description

Technical Field

[0001] This invention belongs to the field of power engineering technology, specifically relating to a distribution network operation and maintenance grid division method for intelligent and precise operation and maintenance. Background Technology

[0002] Against the backdrop of the power system's development towards multi-element collaborative and intelligent operation and maintenance (O&M) of "source-grid-load-storage-charging," the distribution network O&M grid, as the core carrier for resource scheduling, fault repair, and cost control, directly determines O&M efficiency and power supply reliability through its rational division. Influenced by factors such as the expansion of the distribution network and increased coupling of elements, traditional grid division easily faces the dual problems of "spatial looseness" and "lack of coordination": relying solely on a single equipment type for division leads to a scattered distribution of "source-grid-load-storage-charging" units, resulting in excessive travel distances for O&M personnel across units; forcibly merging units while ignoring differences in unit operating characteristics can trigger new problems such as load fluctuation imbalances and energy storage regulation failures within the same grid. If these problems are not properly addressed, they will not only increase O&M time costs but may also lead to the escalation of faults due to delayed response. Therefore, scientific and precise distribution network O&M grid division has become a crucial link in supporting intelligent power grid O&M. Current distribution network systems have accumulated massive amounts of operational data across the "source-grid-load-storage-charging" chain, encompassing multi-dimensional information such as unit attributes and topological relationships. This provides abundant material for grid partitioning methods based on multi-factor collaboration. However, balancing grid spatial compactness with operational synergy and achieving optimal merging of remaining units remains a key challenge that urgently needs to be addressed in this field. Conducting research on multi-dimensional collaborative distribution network operation and maintenance grid partitioning is of profound significance for improving distribution network operation and maintenance efficiency, reducing trial-and-error costs, and meeting the precise operation and maintenance needs under complex operating conditions.

[0003] However, existing methods for dividing distribution network operation and maintenance grids generally have the following limitations: First, existing methods mostly construct grids using a single dimension, failing to fully deconstruct the topological relationships of all elements (source-grid-load-storage-charging) and failing to optimize the spatial distribution of units based on the global shortest path. This results in a loose initial grid space and scattered unit distribution. Given the multi-element coupling characteristics of distribution networks, it is impossible to form a basic grid that is "functionally complete and spatially compact." The distance that operation and maintenance personnel travel across units generally exceeds the optimal threshold, increasing time costs and hindering the continuous improvement of operation and maintenance efficiency, seriously affecting the practicality and economy of grid division. Second, existing methods often directly merge the remaining units after the initial division through simple topological relationships, lacking a trade-off analysis mechanism between "operational characteristics and operational convenience." They fail to cover the long-term, short-term, short-term, and long-term operational status of units, as well as large / small-scale changes, and do not quantify operational similarity through the shortest path. This leads to difficulties in collaborative management of merged grids due to significant differences in operational characteristics. This makes it impossible for existing methods to achieve optimal merging based on the actual correlation of units, making it difficult to balance grid operation coordination and maintenance efficiency, ultimately hindering the realization of the goal of intelligent and precise operation and maintenance of distribution networks. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the technical problem to be solved by this invention is to provide a distribution network operation and maintenance grid division method for intelligent and precise operation and maintenance, which solves the problem that the existing distribution network operation and maintenance grid division does not fully consider the topological association of all elements of "source, network, load, storage and charging", resulting in a loose grid space; and solves the problem that "operational coordination and operation and maintenance convenience cannot be taken into account" when merging the remaining units after the initial division.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0006] A distribution network operation and maintenance grid partitioning method for intelligent and precise operation and maintenance includes a preliminary distribution network operation and maintenance grid partitioning method and a distribution network operation and maintenance grid partitioning optimization method.

[0007] The preliminary grid division method for distribution network operation and maintenance includes the following steps:

[0008] S1.1 The target distribution network is decomposed into different types of land parcel units, and a unique identifier label is assigned to each land parcel unit;

[0009] S1.2 Based on the principle of global shortest path, and taking into account the topological relationships in the identifiers of all plot units in the target distribution network, a preliminary minimum distribution network operation and maintenance grid optimization model is constructed.

[0010] S1.3 Solve the above preliminary minimum distribution network operation and maintenance grid optimization model to obtain the preliminary grid set, and count the remaining units not included in the preliminary grid;

[0011] The proposed distribution network operation and maintenance grid partitioning optimization method classifies and processes the topological associations of the remaining units, and integrates the similarity index of operation characteristics and the similarity index of operation and maintenance convenience to calculate the compromise similarity, thereby achieving the optimal merging of the remaining units based on the compromise similarity.

[0012] Preferably, in step S1.1, the target distribution network is decomposed into five types of land parcel units: "source, grid, load, storage, and charging," represented by the letters S, N, L, E, and C, respectively. A unique identifier is assigned to each land parcel unit, and the set of land parcel units is represented as follows: ,in Indicates the first Each plot of land is labeled with three core pieces of information: plot attributes, plot geometric center coordinates, and plot unit attributes. Topological relationships ;

[0013] in, Indicates the first The plot unit and the first Each plot of land has a grid-load interaction relationship or an energy storage service relationship. Indicates the first The plot unit and the first There is no grid-load interaction or energy storage service relationship between the individual plots, so they do not need to be assigned to the same grid.

[0014] Preferably, step S1.2 includes the following steps:

[0015] Path length calculation: First, calculate the spatial distance between cells using the straight-line distance. Let any two plots within the grid be... , The coordinates of their geometric centers are respectively , The straight-line distance is calculated using the Euclidean distance formula as follows: Take the sum of the shortest connection paths between the plot units in the grid, construct an undirected graph with the plot units in the grid as nodes and the sum of the Euclidean distances to the existing plot units in the grid as edge weights, and select the set of edges with the smallest weights to form a minimum spanning tree. The distances corresponding to the edges in the tree are the shortest connection distances between the plot units. ,

[0016] The formula for calculating the total path length of a single grid cell is as follows:

[0017]

[0018] In the formula, For the first The total path length of each initial grid. For the first The minimum spanning tree edge set of a grid;

[0019] Spanning tree edge set constraints:

[0020] Constraint 1: Mesh attribute constraint, that is, each initial mesh must contain 1 element each of S / N / L / E / C;

[0021] Constraint 2: Membership unity constraint, which means that there must be no overlapping plot units in each grid;

[0022] A plot of land that satisfies both constraints 1 and 2 can be included in the minimum spanning tree edge set.

[0023] Preferably, step S1.3 uses a genetic algorithm to solve the aforementioned preliminary minimum distribution network operation and maintenance grid optimization model to obtain a preliminary grid set. It is necessary to count the remaining units not included in the preliminary grid, calculated as follows:

[0024]

[0025] In the formula, For the remaining land parcel units, This refers to the set of all land parcel units in the distribution network. This is a set of plot units for initial grid coverage.

[0026] Preferably, the method for optimizing the distribution network operation and maintenance grid includes the following steps:

[0027] When the number of topological intersections between the remaining parcel units in the S2.1 set and the initial grid is 1, the parcel unit is merged with the initial grid.

[0028] When there are multiple topological intersections between the remaining plot units in the S2.2 plot unit set and the initial grid, first calculate the operational similarity between the plot unit and the plot units with the same attributes in the initial grid;

[0029] S2.3 Calculate the operational similarity of the initial grid and combine it with the operational similarity to calculate the compromise similarity;

[0030] S2.4 Select the initial grid with the highest compromise similarity as the merging target for merging;

[0031] Repeat steps S2.1-S2.4 until all units in the remaining land parcel unit set have been merged.

[0032] Preferably, in S2.1, the preliminary mesh set is represented as follows: , Indicates the first Each grid, for the remaining set of plot units Any plot unit in First, determine the number of topological intersections between it and the initial mesh: if If there is only a topological association with one initial grid, it is directly incorporated into that grid.

[0033] Preferably, in S2.1, if If a plot cell has topological associations with multiple initial grids, it is necessary to first calculate the similarity between the operational status of the plot cell and plot cells with the same attributes within the grid. The steps are as follows:

[0034] 1) Constructing runtime status feature vectors: Selecting multiple core runtime characteristics to form... The running vector;

[0035] 2) Calculate the average operational similarity: Calculate the similarity between the plot unit and the units with the same attribute in each grid. First, calculate the similarity of the operational characteristic curves at the same time scale. Then, take the average of the operational similarity at multiple time scales. Finally, perform normalization processing.

[0036] Preferably, step S2.3 includes the following steps:

[0037] First, determine the geometric center of the initial mesh, then calculate... To grid The geometric center is represented as Euclidean geometric distance After normalizing the distance, it is converted into operational similarity, i.e.

[0038]

[0039] In the formula, For operational similarity, for The minimum distance to all associated grids. for The maximum distance to all associated grids, where the associated grids are... A grid whose Euclidean geometric distance from the grid meets the threshold set by the power grid operation and maintenance specifications;

[0040] Introducing weighting coefficients By weighted fusion of operational similarity and maintenance similarity, a compromise similarity is obtained, calculated as follows:

[0041]

[0042] In the formula, For the sake of operation and maintenance trade-off similarity, This is the average running similarity calculated in S2.2.

[0043] Preferably, S2.4

[0044] Select the grid with the highest similarity between operation and maintenance as... The merger objective, namely

[0045]

[0046] In the formula, To and The grid with the highest compromise similarity.

[0047] The technical solution of this invention includes a preliminary grid partitioning method for distribution network operation and maintenance based on the shortest path and an optimized grid partitioning method for distribution network operation and maintenance based on the similarity of operation-maintenance trade-offs, which has the following beneficial effects:

[0048] To address the problem of loose grid spatial distribution caused by the insufficient consideration of the topological relationships of all elements (source, grid, load, storage, and charging) in existing distribution network operation and maintenance grid partitioning, this invention proposes an initial partitioning method for distribution network operation and maintenance grids based on the shortest path. First, the target distribution network is deconstructed into five types of land parcels: source, grid, load, storage, and charging, and each parcel is assigned a unique identifier. The identifier contains three core pieces of information: unit attributes, geometric center coordinates, and topological relationships. The topological relationships are determined based on factors such as grid operation specifications and power supply radius to determine whether there are grid-load interactions or energy storage services between units, thus ensuring that the partitioning meets grid operation standards. Second, a preliminary minimum operation and maintenance grid based on the global shortest path is constructed, requiring each grid to contain one unit from each of the five types to ensure functional integrity. Spatial distances between units are calculated using Euclidean distance, constructing an undirected graph with units as nodes and inter-unit distances as weights. The sum of the shortest connection paths within the grid is solved using the minimum spanning tree algorithm, with the overall shortest path across all grids as the optimization objective. Simultaneously, grid attribute constraints and single affiliation constraints are set to prevent duplicate unit affiliation and ensure clear operation and maintenance responsibilities. Finally, a genetic algorithm is used to solve the above model, forming a spatially compact initial grid partitioning scheme. Set operations are then used to statistically analyze the remaining units not included in any grid, providing a basis for subsequent full-coverage optimization. This method achieves a systematic analysis of all elements of the distribution network (source-grid-load-storage-charging) and the compact construction of the initial grid, helping to reduce cross-regional operation and maintenance travel and improve the scientific nature of grid partitioning and the efficiency of operation and maintenance work.

[0049] To address the challenge of balancing operational synergy and maintenance convenience when merging remaining units after initial partitioning, this invention proposes an optimization method for distribution network operation and maintenance grid partitioning based on a trade-off similarity. First, for the remaining units after initial partitioning, the number of topological associations with each initial grid is determined: if a unit has only one topological intersection with an initial grid, it is directly merged into that grid to maintain the integrity of topological associations. Second, for the remaining units with topological intersections with multiple initial grids, multi-timescale operational state feature vectors are constructed, and their operational similarity with units of the same attribute in each grid is calculated. The results from multiple time periods are averaged and normalized. Simultaneously, based on the Euclidean distance from the unit to the geometric center of each grid, combined with distance thresholds set by power grid specifications, an operation and maintenance similarity reflecting ease of maintenance is calculated. Then, a weighted coefficient is introduced to weight and fuse operational similarity and operation and maintenance similarity to obtain an operational-maintenance trade-off similarity, which serves as the core criterion for unit merging. Finally, the remaining cells are merged one by one into the grid with the highest compromise similarity. This process is repeated iteratively until all cells are partitioned, forming an optimized grid that covers the entire domain and balances operational collaboration with ease of maintenance. This method achieves a comprehensive integration of the operational characteristics and operational efficiency of the remaining cells in grid partitioning, effectively improving the typicality of grid partitioning and the efficiency of operation and maintenance.

[0050] These features and advantages of the present invention will be disclosed in detail in the following specific embodiments and accompanying drawings. Attached Figure Description

[0051] The invention will be further described below with reference to the accompanying drawings: Figure 1 This is a flowchart of a distribution network operation and maintenance grid division method for intelligent and precise operation and maintenance according to the present invention. Detailed Implementation

[0052] The technical solutions of the embodiments of the present invention will be explained and described below with reference to the accompanying drawings. However, the following embodiments are only preferred embodiments of the present invention and not all of them. Other embodiments obtained by those skilled in the art based on the embodiments in the implementation methods without creative effort are all within the protection scope of the present invention.

[0053] Those skilled in the art will understand that, without conflict, the features in the following embodiments and implementations can be combined with each other.

[0054] This invention proposes a distribution network operation and maintenance (O&M) grid partitioning method for intelligent and precise O&M, including a preliminary O&M grid partitioning method based on the shortest path and an optimized O&M grid partitioning method based on the similarity of operation-maintenance trade-offs. The specific process is as follows: Figure 1 As shown.

[0055] S1 Preliminary Method for Distribution Network Operation and Maintenance Mesh Division Based on Shortest Path

[0056] To address the problem that existing distribution network operation and maintenance grid divisions do not fully consider the complete combination relationship of "source, network, load, storage, and charging" and the topological relationship, resulting in insufficient grid typicality and compactness, this invention proposes a preliminary grid division method for distribution network operation and maintenance based on the shortest path, thereby constructing a preliminary grid division scheme that covers the typical elements of "source, network, load, storage, and charging" and the compact distribution of these typical elements.

[0057] S1.1 Deconstruction and Label Definition of Distribution Network Plot Units

[0058] The target distribution network is decomposed into five types of land parcel units: "Source (S, such as distributed power sources), Grid (N, such as distribution lines / substations), Load (L, such as industrial / residential loads), Storage (E, such as energy storage power stations), and Charging (C, such as charging pile clusters)". Each land parcel unit is assigned a unique identifier. The set of land parcel units is represented as follows: ,in Indicates the first The first plot of land. (The rest of the text appears to be a fragment and requires further context for accurate translation.) Taking a single land parcel unit as an example, the label of each land parcel unit contains three core pieces of information: unit attributes (clearly specifying the S / N / L / E / C type), unit geometric center coordinates, and other relevant information. Topological relationships Among them, the topology relationship refers to the grid-load interaction relationship and energy storage service range relationship established according to grid operation specifications, power supply radius, etc. Indicates the first The plot unit and the first Each plot of land has a grid-load interaction relationship or an energy storage service relationship. This indicates that there is no grid-load interaction or energy storage service relationship between the two, and they do not need to be assigned to the same grid.

[0059] S1.2 Construction of the Preliminary Minimal Distribution Network Operation and Maintenance Mesh Based on Global Shortest Path

[0060] Based on the principle of global shortest path and considering the topological relationships in the unit plot labels, a preliminary minimum distribution network operation and maintenance grid is constructed, which is a grid containing only one source, network, load, storage, and charging plot unit respectively. Global shortest path refers to minimizing the sum of the shortest path lengths formed by the center points of all plot units in all grids, ensuring a compact grid space. The specific steps include:

[0061] 1) Path length calculation: First, calculate the spatial distance between units using the straight-line distance. Let any two plots within the grid be... , The coordinates of their geometric centers are respectively , The straight-line distance is calculated using the Euclidean distance formula as follows: Take the shortest connection path sum of the plot cells in the grid, construct an undirected graph with plot cells as nodes and the sum of Euclidean distances to existing plot cells in the same grid as edge weights. Select the set of edges with the smallest weights to form a minimum spanning tree. The distances corresponding to the edges in the tree are the shortest connection distances between plot cells. The formula for calculating the total path length of a single grid cell is as follows:

[0062] (1)

[0063] In the formula, For the first The total path length of each initial grid. For the first The minimum spanning tree edge set of each grid.

[0064] 2) Spanning tree edge set constraints: Constraint 1: Mesh attribute constraints, that is, each initial mesh must contain one element each of S / N / L / E / C to avoid missing mesh functionality;

[0065] Constraint 2: Membership unity constraint, which requires that no plots overlap in any grid to avoid confusion in operation and maintenance responsibilities. Only plots that satisfy both constraints 1 and 2 can be included in the minimum spanning tree edge set.

[0066] This method can create a more compact operations and maintenance network, reducing the time cost of operations and maintenance personnel's work.

[0067] S1.3 Preliminary Grid Formation and Remaining Cell Statistics

[0068] The above optimization model is solved using a genetic algorithm to obtain a preliminary mesh set. It is necessary to count the remaining cells not included in the preliminary mesh; the calculation formula is as follows.

[0069] (2)

[0070] In the formula, For the remaining land parcel units, This refers to the set of all land parcel units in the distribution network. This is a set of plot units for initial grid coverage, ensuring that subsequent optimization can achieve full coverage of the distribution network.

[0071] S2 Optimization Method for Distribution Network Operation and Maintenance Mesh Partitioning Based on Operation-Maintenance Trade-off Similarity

[0072] To address the issue of "inability to balance operational synergy and maintenance convenience" when merging remaining units after initial partitioning, this invention proposes a distribution network maintenance grid partitioning optimization method based on operational-maintenance trade-off similarity. By classifying and processing the topological associations of remaining units, and integrating multi-dimensional operational characteristics and maintenance convenience indicators to calculate trade-off similarity, the optimal merging of remaining units is achieved, thereby improving the overall grid synergy.

[0073] S2.1 Direct merging of remaining cells in a single topological intersection

[0074] The initial mesh set is represented as , Indicates the first One grid. For the remaining set of parcel units. land parcel units, such as First, determine the number of topological intersections between it and the initial mesh: if With only one initial grid, for example If a topological relationship exists, then directly... Merge into this grid; the merging logic is as follows:

[0075] (3)

[0076] S2.2 Calculation of similarity of remaining cells in multiple topological intersections

[0077] like If a plot cell has topological associations with multiple initial grids, it is necessary to first calculate the similarity between the operational status of the plot cell and plot cells with the same attributes within the grid. The steps are as follows:

[0078] 1) Constructing runtime status feature vectors: Selecting multiple core runtime characteristics to form... The operating vectors, such as the operating characteristic curves for the past year, quarter, week, and day, and the operating characteristic curves for the next year, quarter, week, and day.

[0079] 2) Calculate the average running similarity : Calculate land parcel units With units of the same attribute within each grid, for example Cell grid in The similarity of the running curves is calculated first, then the similarity across multiple time scales is averaged, and finally normalization is performed. The DTW algorithm can be used for similarity calculation; no specific method is specified here.

[0080] S2.3 Calculation based on operation and maintenance similarity and compromise similarity

[0081] Operational similarity reflects operational convenience through spatial distance (the closer the distance, the shorter the travel time for operations personnel, and the higher the convenience): First, determine the geometric center of the initial grid (take the average of the coordinates of the centers of all cells within the grid as the grid's "reference point"), for example, the grid The geometric center is represented as .calculate To grid The geometric center is represented as Euclidean geometric distance After normalizing the distance, it is converted into operational similarity, i.e.

[0082] (4)

[0083] In the formula, For operational similarity (mapped to the range [0,1]). for The minimum distance to all associated grids. for The maximum distance to all associated grids, where the associated grids are... A grid whose Euclidean geometric distance from the grid meets the threshold set by the power grid operation and maintenance specifications.

[0084] To simultaneously satisfy operational collaboration and ease of maintenance, a weighting coefficient is introduced. By weighted fusion of operational similarity and maintenance similarity, a compromise similarity (as...) is obtained. The core criteria for merger determination), calculated as follows:

[0085] (5)

[0086] In the formula, For the sake of operation and maintenance trade-off similarity, This is the average running similarity calculated in S2.2.

[0087] S2.4 Optimization and Mesh Update of Multiple Intersection Remaining Cells

[0088] Select the grid with the highest similarity between operation and maintenance as... The merger objective, namely

[0089] (6)

[0090] In the formula, To and The grid with the highest similarity is selected as the compromise. Steps S2.1-S2.4 are repeated until the remaining set of plot cells is reached. All units were merged to form a typical operation and maintenance grid partitioning scheme that covers the entire distribution network, is highly efficient in operation and maintenance, and achieves the partitioning goal of "compact space, coordinated operation, and efficient operation and maintenance".

[0091] The technical solution adopted in this implementation method, based on the shortest path distribution network operation and maintenance grid preliminary division method, can generate a compact initial grid division result. It also extracts redundant units not included in any grid through set operations, laying the foundation for the next step of achieving full-area coverage optimization. This achieves a systematic analysis of all elements of the distribution network ("source-grid-load-storage-charging") and a compact initial grid construction, helping to reduce cross-regional operation and maintenance travel and improve the scientific nature of grid division and the efficiency of operation and maintenance work. Furthermore, the distribution network operation and maintenance grid division optimization method based on the similarity of operation and maintenance trade-offs achieves a comprehensive merging of the operational characteristics and operation and maintenance efficiency of the remaining units in the grid division, effectively improving the typicality of grid division and the efficiency of operation and maintenance.

[0092] The above description is merely a specific embodiment of the invention, but the scope of protection of the invention is not limited thereto. Those skilled in the art should understand that the invention includes, but is not limited to, the contents described in the accompanying drawings and the specific embodiments above. Any modifications that do not depart from the functional and structural principles of the invention will be included within the scope of the claims.

Claims

1. A distribution network operation grid partitioning method for intelligent and accurate operation, characterized in that, This includes methods for the initial division of distribution network operation and maintenance grids and methods for optimizing the division of distribution network operation and maintenance grids. The preliminary grid division method for distribution network operation and maintenance includes the following steps: S1.1 The target distribution network is decomposed into different types of land parcel units, and a unique identifier label is assigned to each land parcel unit; S1.2 Based on the principle of global shortest path, and taking into account the topological relationships in the identifiers of all plot units in the target distribution network, a preliminary minimum distribution network operation and maintenance grid optimization model is constructed. S1.3 Solve the above preliminary minimum distribution network operation and maintenance grid optimization model to obtain the preliminary grid set, and count the remaining units not included in the preliminary grid; The proposed distribution network operation and maintenance grid partitioning optimization method classifies and processes the topological associations of the remaining units, and integrates the similarity index of operation characteristics and the similarity index of operation and maintenance convenience to calculate the compromise similarity, thereby achieving the optimal merging of the remaining units based on the compromise similarity. 2.The distribution network operation grid division method for intelligent and precise operation of claim 1, wherein In S1.1, the target distribution network is decomposed into five types of land parcel units: "source, grid, load, storage, and charging," represented by the letters S, N, L, E, and C, respectively. A unique identifier is assigned to each land parcel unit, and the set of land parcel units is represented as follows: ,in Indicates the first Each plot of land is labeled with three core pieces of information: plot attributes, plot geometric center coordinates, and plot unit attributes. Topological relationships ; in, Indicates the first The plot unit and the first Each plot of land has a grid-load interaction relationship or an energy storage service relationship. Indicates the first The plot unit and the first There is no grid-load interaction or energy storage service relationship between the individual plots, so they do not need to be assigned to the same grid.

3. The distribution network operation grid division method for intelligent precision operation according to claim 2, characterized in that, S1.2 includes the following steps: Path length calculation: First, calculate the spatial distance between cells using the straight-line distance. Let any two plots within the grid be... , The coordinates of their geometric centers are respectively , The straight-line distance is calculated using the Euclidean distance formula as follows: Take the sum of the shortest connection paths between the plot units in the grid, construct an undirected graph with the plot units in the grid as nodes and the sum of the Euclidean distances to the existing plot units in the grid as edge weights, and select the set of edges with the smallest weights to form a minimum spanning tree. The distances corresponding to the edges in the tree are the shortest connection distances between the plot units. , The formula for calculating the total path length of a single grid cell is as follows: In the formula, For the first The total path length of each initial grid. For the first The minimum spanning tree edge set of a grid; Spanning tree edge set constraints: Constraint 1: Mesh attribute constraint, that is, each initial mesh must contain 1 element each of S / N / L / E / C; Constraint 2: Membership unity constraint, which means that there must be no overlapping plot units in each grid; A plot of land that satisfies both constraints 1 and 2 can be included in the minimum spanning tree edge set.

4. The distribution network operation grid division method for intelligent precision operation according to claim 3, characterized in that, S1.3 uses a genetic algorithm to solve the above preliminary minimum distribution network operation and maintenance grid optimization model to obtain a preliminary grid set. It is necessary to count the remaining units not included in the preliminary grid, and the calculation formula is as follows: wherein is the set of remaining patch cells, is the set of all patch cells of the distribution grid, is the set of patch cells covered by the preliminary grid.

5. The distribution network operation and maintenance grid division method for intelligent and precise operation and maintenance according to claim 1, characterized in that, The method for optimizing the grid division of the power distribution network operation and maintenance includes the following steps: When the number of topological intersections between the remaining parcel units in the S2.1 set and the initial grid is 1, the parcel unit is merged with the initial grid. When there are multiple topological intersections between the remaining plot units in the S2.2 plot unit set and the initial grid, first calculate the operational similarity between the plot unit and the plot units with the same attributes in the initial grid; S2.3 Calculate the operational similarity of the initial grid and combine it with the operational similarity to calculate the compromise similarity; S2.4 Select the initial grid with the highest compromise similarity as the merging target for merging; Repeat steps S2.1-S2.4 until all units in the remaining land parcel unit set have been merged.

6. The distribution network operation and maintenance grid division method for intelligent and precise operation and maintenance according to claim 5, characterized in that, In S2.1, the preliminary mesh set is represented as follows: , Indicates the first Each grid, for the remaining set of plot units Any plot unit in First, determine the number of topological intersections between it and the initial mesh: if If there is only a topological association with one initial grid, it is directly incorporated into that grid.

7. The distribution network operation and maintenance grid division method for intelligent and precise operation and maintenance according to claim 6, characterized in that, In S2.1, if If there is a topological association with multiple preliminary grids, the similarity between the operation of the plot unit and the plot unit with the same attributes in the grid needs to be calculated first. The steps are as follows: 1) Constructing the running state feature vector: select multiple core running characteristics to form the running vector; of the running vector; 2) Calculate the average operational similarity: Calculate the similarity between the plot unit and the units with the same attribute in each grid. First, calculate the similarity of the operational characteristic curves at the same time scale. Then, take the average of the operational similarity at multiple time scales. Finally, perform normalization processing.

8. The distribution network operation and maintenance grid division method for intelligent and precise operation and maintenance according to claim 5, characterized in that, S2.3 includes the following steps: First, determine the geometric center of the initial mesh, then calculate... To grid The geometric center is represented as Euclidean geometric distance After normalizing the distance, it is converted into operational similarity, i.e. In the formula, For operational similarity, for The minimum distance to all associated grids. for The maximum distance to all associated grids, where the associated grids are... A grid whose Euclidean geometric distance from the grid meets the threshold set by the power grid operation and maintenance specifications; Introducing weighting coefficients By weighted fusion of operational similarity and maintenance similarity, a compromise similarity is obtained, calculated as follows: In the formula, For the sake of operation and maintenance trade-off similarity, This is the average running similarity calculated in S2.

2.

9. The distribution network operation grid division method for intelligent precision operation according to claim 5, characterized in that, S2.4 The grid with the highest similarity between the operation and maintenance compromise is selected as the merging target of , that is In the formula, is the highest similarity with the grid with the highest compromise similarity.

Images