Centralized new energy power line path generation method, system, equipment and medium

Abstract

The invention discloses a centralized new energy power line path generation method, system and device, and a medium. The method comprises the steps of obtaining a plurality of historical line path cases and case parameters thereof in a preset area of a centralized new energy station; according to the historical line path cases and the case parameters thereof, a traversal method is adopted to quantify the land type weight and the terrain information weight of a target area corresponding to the centralized new energy station; based on the land type weight and the terrain information weight, with the shortest power line path of the target area as a target, generating an initial power line path through an A star algorithm and a reinforcement learning algorithm, and optimizing the initial power line path to obtain a power line optimized path; according to the environment data of the target area and the current voltage grade, tower arrangement is carried out on the power line optimization path, a final power line path arrangement strategy of the target area is generated, and the generation efficiency and accuracy of the centralized new energy station power transmission line are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power energy, and particularly to a method, a system, a device and a medium for generating a path of a centralized new energy power line. Background Art

[0002] With the continuous development of computer technology, using computer algorithms to assist in the design of transmission lines has become a research hotspot.

[0003] Currently, the Dijkstra algorithm and the ant colony algorithm are usually used for the design of the power transmission lines of centralized new energy power stations. However, the classic Dijkstra algorithm needs to traverse most of the areas of the map, with a high time complexity. The ant colony algorithm is too dependent on the tuning of hyperparameters, and different parameters need to be adapted to different geographical conditions, and it cannot guarantee the global optimum of the line selection. Therefore, the existing algorithms have limited efficiency and accuracy in the design of the power transmission lines of centralized new energy power stations.

[0004] Thus, how to improve the generation efficiency and accuracy of the power transmission lines of centralized new energy power stations has become a technical problem that needs to be urgently solved by those skilled in the art. Summary of the Invention

[0005] The present invention provides a method, a system, a device and a medium for generating a path of a centralized new energy power line, and solves the problem of how to improve the generation efficiency and accuracy of the power transmission lines of centralized new energy power stations.

[0006] To solve the above technical problem, a first aspect of the present invention provides a method for generating a path of a centralized new energy power line, including:

[0007] Obtaining a plurality of historical line path cases and their case parameters within a preset area of a centralized new energy power station;

[0008] Quantifying the land type weight and the terrain information weight of the target area corresponding to the centralized new energy power station by using a traversal method according to each of the historical line path cases and their case parameters;

[0009] Based on the land type weight and the terrain information weight, with the shortest power line path of the target area as the goal, generating and optimizing an initial power line path through the A* algorithm and the reinforcement learning algorithm to obtain an optimized power line path;

[0010] Arranging poles and towers for the optimized power line path according to the environmental data and the current voltage level of the target area to generate a final power line path arrangement strategy for the target area.

[0011] As one of the preferred solutions, quantifying the land type weight and terrain information weight of the target area corresponding to the centralized new energy power station by using the traversal method according to each of the historical line path cases and their case parameters includes:

[0012] Performing clustering analysis on each of the case parameters through a deep learning algorithm to obtain a clustering model, and inputting the design requirement information of the target area into the clustering model for processing to obtain a historical line path target case that matches the design requirement information;

[0013] Based on each of the historical line path target cases, quantifying the land type weight and terrain information weight of the target area by using the traversal method.

[0014] As one of the preferred solutions, the quantifying the land type weight and terrain information weight of the target area by using the traversal method based on each of the historical line path target cases includes:

[0015] Obtaining the actual curve of the transmission path and the actual land type data in each of the historical line path target cases;

[0016] Setting a first weight search space to traverse all land type weights for each of the historical line path target cases to obtain a first candidate weight combination;

[0017] Based on each of the first candidate weight combinations, using a path planning algorithm to generate a first virtual curve of the transmission path for each of the historical line path target cases under each of the first candidate weights to quantify the similarity distance from each of the actual curves of the transmission paths, and obtaining a first similarity calculation result;

[0018] According to the first similarity calculation result, taking the first candidate weight with the smallest similarity distance as the land type weight of the corresponding historical line path target case;

[0019] Obtaining the current land type data of the target area in the design requirement information to quantify the similarity distance from the actual land type data of each of the historical line path target cases, and obtaining a second similarity calculation result;

[0020] According to the second similarity calculation result, adjusting the land type weight of the historical line path target case with the smallest similarity distance to be used as the land type weight of the target area.

[0021] As one of the preferred solutions, the quantifying the land type weight and terrain information weight of the target area by using the traversal method based on each of the historical line path target cases further includes:

[0022] Obtain the actual data of the terrain information in each of the historical line path target cases;

[0023] Set a second weight search space, and traverse all terrain information weights for each of the historical line path target cases to obtain a second candidate weight combination;

[0024] Based on each of the second candidate weight combinations, use a path planning algorithm to generate a second virtual curve of the sending path for each of the historical line path target cases under each second candidate weight, so as to quantify the similarity distance between each of the sending path actual curves, and obtain a third similarity calculation result;

[0025] According to the third similarity calculation result, take the second candidate weight with the smallest similarity distance as the terrain information weight of the corresponding historical line path target case;

[0026] Obtain the current terrain information data of the target area in the design requirement information, so as to quantify the similarity distance between each of the historical line path target cases' terrain information actual data, and obtain a fourth similarity calculation result;

[0027] According to the fourth similarity calculation result, take the terrain information weight of the historical line path target case with the smallest similarity distance as the terrain information weight of the target area.

[0028] As one of the preferred solutions, based on the land type weight and the terrain information weight, with the goal of the shortest power line path in the target area, generate and optimize an initial power line path through the Astar algorithm and the reinforcement learning algorithm, including:

[0029] Perform grid division on the target area, and label the corresponding land type weight and terrain information weight for each divided grid based on the land type weight and terrain information weight of the target area;

[0030] Construct a cost function according to the land type weight and terrain information weight of the target area, and construct a heuristic function based on the distance between the current node and the target node in the target area;

[0031] Set the initial node and the termination node of the target area, with the goal of the shortest power line path in the target area, and based on the cost function and the heuristic function, calculate the cost and heuristic value of adjacent grids starting from the initial node, and expand the node with the smallest total cost until reaching the termination node;

[0032] Trace back the parent node from the termination node in reverse to generate an initial power line path.

[0033] As one of the preferred solutions, based on the land type weight and the terrain information weight, with the goal of the shortest power line path in the target area, an initial power line path is generated and optimized through the Astar algorithm and the reinforcement learning algorithm to obtain an optimized power line path, which further includes:

[0034] Taking the grid coordinates of the current node, the land type weight and the terrain information weight of the adjacent grid as the state space, and taking the moving direction and the preset range of path adjustment as the action space;

[0035] Constructing a reward function based on the path length change, the path turning times penalty, the path tortuosity coefficient, and the high-cost area crossing penalty;

[0036] Training the DQN model according to the state space, the action space and the reward function, and taking the initial power line path as the initial strategy to optimize by using the trained DQN model to obtain the optimized power line path.

[0037] As one of the preferred solutions, arranging the power line towers according to the environmental data of the target area and the current voltage level to generate the final power line path arrangement strategy for the target area, including:

[0038] Determining the positions of obstacles in each grid according to the environmental data to label each grid, and setting the span standard based on the current voltage level;

[0039] Based on the span standard, using the greedy strategy to initially arrange the power line towers for the optimized power line path, and optimizing the initial arrangement result through the improved genetic algorithm to obtain the final power line path arrangement strategy for the target area.

[0040] The second aspect of the present invention provides a centralized new energy power line path generation system, including:

[0041] A data acquisition module for acquiring a plurality of historical line path cases and their case parameters within a preset area of a centralized new energy power station;

[0042] A weight quantization module for quantifying the land type weight and the terrain information weight of the target area corresponding to the centralized new energy power station by using the traversal method according to each historical line path case and its case parameters;

[0043] A path generation module for generating and optimizing an initial power line path based on the land type weight and the terrain information weight, with the goal of the shortest power line path in the target area, through the Astar algorithm and the reinforcement learning algorithm to obtain an optimized power line path;

[0044] The tower arrangement module is used to arrange towers for the optimized path of the power line according to the environmental data of the target area and the current voltage level, and generate the final power line path arrangement strategy for the target area.

[0045] The third aspect of the present invention provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the centralized new energy power line path generation method as described above is implemented.

[0046] The fourth aspect of the present invention provides a computer-readable storage medium, which includes a stored computer program. When the device where the computer-readable storage medium is located executes the computer program, the centralized new energy power line path generation method as described above is implemented.

[0047] Compared with the prior art, the beneficial effects of the embodiments of the present invention are at least one of the following:

[0048] (1) When planning the path, the influence of terrain and land type on line construction is fully considered, avoiding the influence of poor geological conditions and land type on line stability, being able to generate the optimal line path, reducing the line length and complexity, and thus reducing the construction cost;

[0049] (2) By adopting the Astar algorithm and the reinforcement learning algorithm to quickly generate and optimize the power line path, the planning efficiency is greatly improved, which helps new energy power stations to quickly access the power grid and meet the growing energy demand;

[0050] (3) Through historical data-driven, optimization algorithms and intelligent arrangement strategies, the full process automation of the transmission line design of new energy power stations is realized, significantly improving the generation efficiency and accuracy of transmission lines, and enhancing the engineering reliability while reducing costs. Description of the Drawings

[0051] In order to more clearly illustrate the technical solutions of the present invention, the accompanying drawings required for implementation will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0052] Figure 1 It is a flowchart of a centralized new energy power line path generation method provided by an embodiment of the present invention;

[0053] Figure 2It is a flowchart of a method for generating a path of a centralized new energy power line provided by another embodiment of the present invention;

[0054] Figure 3 It is a structural diagram of a system for generating a path of a centralized new energy power line provided by an embodiment of the present invention;

[0055] Figure 4 It is a structural diagram of an electronic device provided by an embodiment of the present invention. Detailed implementation manners

[0056] Next, in combination with the accompanying drawings and embodiments, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

[0057] In the description of this application, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, features defined with "first", "second", "third", etc. may explicitly or implicitly include one or more of such features. In the description of this application, unless otherwise stated, the meaning of "a plurality" is two or more.

[0058] In the description of this application, it should be noted that unless otherwise clearly defined and limited, the terms "installed", "connected", "connected" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected, or indirectly connected through an intermediate medium, and it may be the communication inside two components. The terms "vertical", "horizontal", "left", "right", "up", "down" and similar expressions used herein are only for the purpose of illustration, rather than indicating or implying that the system or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be understood as a limitation of the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0059] In the description of the present application, it should be noted that unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0060] Most of the computer algorithms currently used in the design of auxiliary transmission lines are based on the current design requirements and land information, without performing a correlation analysis between the current design and the existing designs, nor exploring the data potential of a large number of existing transmission line schemes.

[0061] Based on this, in one embodiment, as Figure 1 shown, the first aspect of the present invention provides a method for generating a path of a centralized new energy power line, including:

[0062] S1. Obtain a plurality of historical line path cases and their case parameters within a preset area of a centralized new energy power station;

[0063] S2. According to each of the historical line path cases and their case parameters, use the traversal method to quantify the land type weight and terrain information weight of the target area corresponding to the centralized new energy power station;

[0064] S3. Based on the land type weight and the terrain information weight, with the goal of the shortest power line path in the target area, generate and optimize an initial power line path through the A star algorithm and the reinforcement learning algorithm to obtain an optimized power line path;

[0065] S4. Perform tower arrangement on the optimized power line path according to the environmental data and the current voltage level of the target area to generate a final power line path arrangement strategy for the target area.

[0066] The present invention is based on the intelligent fusion analysis of GIS data such as transmission line construction data, land type, and terrain information of historical line path cases in the area near a centralized new energy power station, and performs parameter clustering on similar historical cases to obtain the land type weight and terrain information weight of the target area to which the centralized new energy power station belongs. Then, based on this weight information, the A star optimization algorithm and the reinforcement learning algorithm are used to generate a power line path, and the tower arrangement and index calculation are automatically completed according to the design requirements to obtain a final power line path arrangement strategy, so as to effectively improve the generation efficiency and accuracy of the transmission line in the target area, specifically including:

[0067] First, obtain all historical line path cases that have been built or planned in the past in the area where the centralized new energy power station is located and its nearby preset area (such as a 50-kilometer radius around the new energy power station), or in a similar area (such as an area with the same or similar terrain and landform, etc.), including successful and partially failed cases (i.e., paths diverted due to geological problems). Extract the corresponding case parameters from each historical case, such as transmission type, land type, terrain information, transmission line type, line length, line orientation, tower type and tower coordinates on the line, voltage level, construction cost, operation and maintenance cost, etc. Then, through UTM projection, unify the coordinates of these case parameters and perform preprocessing operations such as removing obvious error data to construct a structured database, associating each historical case with case parameter tags for subsequent analysis. The present invention uses historical cases to generate transmission lines, providing a statistical basis for weight calculation and also avoiding repeating historical mistakes.

[0068] Next, according to each historical line path case and its case parameters, the traversal method is used to quantify the land type weight and terrain information weight of the target area corresponding to the centralized new energy power station, that is, the proportion of each land type and terrain information in the power line construction target area. Since the land type weight is a key parameter for generating the power line outgoing path curve, and in existing cases, only the generated outgoing path curves are available, lacking weight data for various land types. Based on this, the present invention performs interpolation analysis based on the empirical weights of land types in the target area and its vicinity in the past five years. If there are no similar land type weights in the vicinity, the land type weights are inversely deduced according to the transmission line parameters on similar land types in the same province or the whole country through interpolation analysis and the traversal method, and combined with curve similarity analysis to obtain the land type weights of historical cases to estimate the land type weights of the target area, and complete the fine-tuning of the land type weights of the target area by combining historical cases and current design requirements. Similarly, for the terrain information weight of the target area, according to historical cases and their related data, the traversal method is used to inversely deduce the land type weights of each historical case to calculate the curve similarity between the lines generated under each weight of the historical cases and the actual lines, and obtain the land type weights of each case to estimate the terrain information weight value of the currently designed target area.

[0069] Then, according to the calculated land type weights and terrain information weights of the target area, a heuristic function for power line path search is constructed with the goal of the shortest power line path in the target area. The A star algorithm is applied to search for the shortest path from the starting node (centralized new energy power station) to the ending node (power access point within the target area); then, the initial path generated by the A star algorithm is used as the initial state of the reinforcement learning algorithm, and the state space, action space, reward function, etc. of the reinforcement learning algorithm are defined. Through continuous iterative learning, the path is adjusted to maximize the reward function (such as the shortest path length, the lowest construction cost, etc.) to obtain the optimized power line path; it combines heuristic search and machine learning algorithms to generate the power line path, improving the efficiency and accuracy of path planning.

[0070] Finally, based on the environmental data of the target area and the current voltage level, obstacle point marking and span rule mapping are performed. Based on these rules, a tower intelligent layout algorithm (such as greedy strategy, genetic algorithm, etc.) is used to layout the towers for the optimized power line path, generating the final power line path layout strategy for the target area.

[0071] The present invention combines historical data-driven, multiple intelligent algorithms and intelligent layout strategies, realizing the full-process automation of the transmission line design of new energy power stations, significantly improving the efficiency and accuracy of strategy generation, and further reducing costs.

[0072] In one embodiment, step S2 includes:

[0073] Performing clustering analysis on each of the case parameters through a deep learning algorithm to obtain a clustering model, and inputting the design requirement information of the target area into the clustering model for processing to obtain a historical line path target case that matches the design requirement information;

[0074] Based on each of the historical line path target cases, the land type weights and terrain information weights of the target area are quantified using the traversal method.

[0075] Specifically, the present invention performs clustering analysis on each case parameter through a deep neural network to form a clustering model, and this process is carried out by the following formula:

[0076] X = {x (1) + x (2) + x (3) + x (4) + x (5) +... + x (n)}

[0077] C = f(X)

[0078] In the formula, X is the set of case parameters of the historical line path cases; x(1) is transmission type information; x (2) is terrain information; x (3) is voltage level; x (4) is transmission line; x (5) are the coordinates of the poles on the transmission line; n is the total number of case parameters; C is the clustering model, f() is the deep neural network model, which can be a multi-layer perceptron, a convolutional neural network model, a long short-term memory network model, etc., and the specific type is not limited here.

[0079] Next, the current design requirement information (construction cost, land type, terrain information, construction restrictions, etc.) of the target area is input into the clustering model for processing, and the case clustering result similar to the current design requirement can be obtained, that is, the historical line path target case. Based on the data of these historical line path target cases, the weights of the land type and terrain information of each historical target case are inversely deduced by the traversal method to calculate the curve similarity between the line generated by each historical target case under each weight and the actual line. Furthermore, according to the similarity calculation result, the weights of the land type and terrain information of each historical target are obtained to estimate the weight values of the land type and terrain information of the target area of the current design.

[0080] The present invention performs special clustering on the parameters of historical cases through a deep learning algorithm to realize the modeling of complex non-linear relationships, maps the design requirements (such as land type distribution, slope range) of the target area to the clustering space, screens the most similar historical cases through the similarity distance, reduces the dependence on artificial experience, and improves the matching accuracy; by inversely deducing the land type weight and terrain weight of historical cases, the data potential of historical cases is utilized, the deviation of traditional subjective weighting is avoided, and the data sparsity problem is solved at the same time.

[0081] In addition, for the calculation of the land type weight and terrain information weight of the target area, the combination of deep variational clustering and genetic algorithm can also be adopted: In the clustering stage: The DeepDPM model (deep non-parametric clustering) is used to automatically infer the number of clusters, avoiding the limitation of presetting the number of categories; In the weight optimization stage: The genetic algorithm is used to replace the traversal method, and the optimal weight combination is globally searched through crossover and mutation operations to solve the computational complexity problem of the high-dimensional weight space. Specifically, it includes: preprocessing the case parameters of each historical case to ensure the accuracy and consistency of the data, and using the DeepDPM model to cluster the preprocessed data. The DeepDPM model can automatically infer the number of clusters, avoiding the limitation of presetting the number of categories. During the training process, the DeepDPM model changes the number of clusters through splitting / fusion operations and maintains a sub-cluster pair for each cluster, thereby realizing the deep non-parametric clustering of the data. Analyze the clustering results of the DeepDPM model to obtain the cluster centers of different land types and terrain information and the corresponding sample points to calculate the similarity with the design requirement information of the target area, and select a preset number of historical cases with high similarity as historical target cases; Then, according to the historical target cases, generate a set of candidate solutions containing different land type and terrain information weights as the initial population of the genetic algorithm. Take the shortest power line path planning as the fitness function to evaluate the quality of each candidate solution. According to the fitness value, select a part of the candidate solutions as parents for subsequent crossover and mutation operations. The selection strategy can adopt roulette wheel selection, tournament selection, etc. Perform crossover operations on the selected parents to generate new offspring candidate solutions. The crossover strategy can adopt single-point crossover, two-point crossover, etc.; Perform mutation operations on the newly generated offspring candidate solutions to increase the diversity of the population. The mutation strategy can adopt single-point mutation, bit-by-bit mutation, etc. After forming a new population, iteratively execute steps such as fitness calculation, crossover and mutation until the maximum number of iterations is reached or the fitness value converges, etc., and select the candidate solution with the highest fitness value from the population as the land type weight and terrain information weight of the target area.

[0082] The present invention uses the DeepDPM model for deep non-parametric clustering, which can automatically infer the number of clusters, avoiding the limitation of presetting the number of categories. It can not only improve the clustering accuracy, but also reduce manual intervention and improve the automation degree of clustering; The genetic algorithm is used to replace the traversal method, and the optimal weight combination is globally searched through crossover and mutation operations. The genetic algorithm has the characteristics of global search ability and parallel computing, and can quickly find the optimal solution or approximate optimal solution in the high-dimensional weight space, thereby solving the computational complexity problem of the high-dimensional weight space; Combining the DeepDPM model and the genetic algorithm can realize the accurate calculation and optimization of the land type and terrain information weights of the target area, which can not only improve the efficiency of power line path planning, but also ensure the accuracy of the planning, reduce the construction cost and environmental impact.

[0083] In one embodiment, the method for quantifying the land type weight and terrain information weight of the target area by using the traversal method based on each of the historical line path target cases includes:

[0084] Obtain the actual curve of the transmission path and the actual land type data in each of the historical line path target cases;

[0085] Set a first weight search space, and traverse all land type weights for each of the historical line path target cases to obtain a first candidate weight combination;

[0086] Based on each of the first candidate weight combinations, use a path planning algorithm to generate a first virtual curve of the transmission path for each of the historical line path target cases under each of the first candidate weights, so as to quantify the similarity distance between each of the virtual curves and the actual curve of the transmission path, and obtain a first similarity calculation result;

[0087] According to the first similarity calculation result, take the first candidate weight with the minimum similarity distance as the land type weight of the corresponding historical line path target case;

[0088] Obtain the current land type data of the target area in the design requirement information, so as to quantify the similarity distance between the current land type data and the actual land type data of each of the historical line path target cases, and obtain a second similarity calculation result;

[0089] According to the second similarity calculation result, adjust the land type weight of the historical line path target case with the minimum similarity distance to be used as the land type weight of the target area.

[0090] Specifically, the present invention takes the actual curve C of the transmission path in each of the historical line path target cases i and its actual land type data L i as inputs (i is the number of historical line path target cases), and inversely deduces the implicit land type weights of each historical target case (such as the proportion of land types such as cultivated land, forest land, water area, and wasteland in their respective historical target cases). First, set a first weight search space corresponding to the land type weight: assume that the value range of each land type weight W j is [0, 1], and the step size is set to ΔW (such as 0.01). For each historical target case, traverse all possible land type weights W j (j = 1, 2,..., m) to obtain a first candidate weight combination, and based on this combination, use a path planning algorithm (such as the A* algorithm, the best-first search BFS algorithm, etc.) to generate a first virtual curve of the transmission path for each historical target case under each of the first candidate weights, so as to calculate the similarity distance between each of the virtual curves and the corresponding actual curve of the transmission path, which is represented by the following formula:

[0091]

[0092] Wherein, d ij is the first similarity calculation result; C ijk is the k-th point coordinate of the first outgoing path virtual curve of the i-th historical line path target case; K is the total number of point coordinates; C ik is the k-th point coordinate of the actual curve of the outgoing path of the i-th historical line path target case.

[0093] Then, for each historical line path target case, the first candidate weight value corresponding to the minimum similarity distance d ij is used as the land type weight of the historical line path target case, and the current land type data of the target area is extracted from the design requirement information to calculate the similarity with the land types of each historical line path target case, that is:

[0094]

[0095] Wherein, d i is the second similarity calculation result; L newj is the proportion of the j-th current land type data in the target area; L ij is the proportion of the j-th land type in the i-th historical line path target case; m is the number of land type categories.

[0096] For each historical line path target case, the land type weight of the historical line path target case with the minimum similarity distance d i is fine-tuned by combining experience as the land type weight of the target area; alternatively, the land type weight of the historical line path target case with the minimum similarity distance d i is weighted and averaged according to the inverse similarity distance and then adjusted by introducing domain knowledge as the land type weight of the target area. The present invention makes full use of the data potential of historical cases and efficiently and accurately estimates the land type weight of the current design through the traversal method, providing reliable data support for subsequent line generation and optimization.

[0097] In one embodiment, based on each of the historical line path target cases, using the traversal method to quantify the land type weight and terrain information weight of the target area further includes:

[0098] Obtaining the actual terrain information data in each of the historical line path target cases;

[0099] Setting a second weight search space to traverse all terrain information weights for each of the historical line path target cases to obtain a second candidate weight combination;

[0100] Based on each of the second candidate weight combinations, a path planning algorithm is used to generate virtual second delivery path curves of each of the historical route path target cases under each second candidate weight, so as to quantify the similarity distance from the actual curve of each delivery path, and a third similarity calculation result is obtained;

[0101] According to the third similarity calculation result, the second candidate weight with the minimum similarity distance is used as the terrain information weight corresponding to the historical route path target case;

[0102] Obtain the current terrain information data of the target area in the design requirement information to quantify the similarity distance from the actual terrain information data of each historical route path target case, and obtain a fourth similarity calculation result;

[0103] According to the fourth similarity calculation result, the terrain information weight of the historical route path target case with the minimum similarity distance is used as the terrain information weight of the target area.

[0104] Specifically, similar to the quantification process of the land type weight, the present invention uses the actual delivery path curve and its actual terrain information data in each historical route path target case as inputs to reverse-infer the terrain information weights (such as elevation, slope ratio, etc.) of each historical target case. First, set the second weight search space corresponding to the terrain information weight, and for each historical target case, traverse all possible terrain information weights to obtain second candidate weight combinations; based on this combination, use the path planning algorithm to generate virtual second delivery path curves of each historical target case under each second candidate weight to calculate the similarity distance between each of them and the corresponding actual delivery path curve, and obtain a third similarity result; for each historical route path target case, use the second candidate weight value corresponding to the minimum distance in the third similarity result as the terrain information weight of this historical route path target case, and extract the current terrain information data of the target area from the design requirement information to perform similarity calculation with the terrain information of each historical route path target case to obtain a fourth similarity result; for each historical route path target case, use the terrain information weight of the historical route path target case with the minimum distance in the fourth similarity calculation result as the terrain information weight of the target area, that is, the elevation and slope parameters of the current design; among them, the similarity distance calculation formula for the terrain information can refer to the similarity calculation formula of the aforementioned land type, and will not be elaborated here. The present invention makes full use of the data potential of historical cases and efficiently and accurately estimates the land type weight of the current design through the traversal method, providing reliable data support for subsequent route generation and optimization.

[0105] In one embodiment, step S3 includes:

[0106] The target area is divided into grids, and each divided grid is labeled with the corresponding land type weight and terrain information weight based on the land type weight and terrain information weight of the target area;

[0107] A cost function is constructed according to the land type weight and terrain information weight of the target area, and a heuristic function is constructed based on the distance between the current node and the target node in the target area;

[0108] Set the initial node and the termination node of the target area. With the goal of the shortest power line path in the target area, and based on the cost function and the heuristic function, calculate the cost and heuristic value of adjacent grids starting from the initial node, and expand the node with the minimum total cost until reaching the termination node;

[0109] Trace back the parent node from the termination node in reverse to generate the initial power line path.

[0110] Specifically, the present invention divides the area into grids of equal size (such as 100m×100m), and labels each grid with the land type weight, elevation and slope; then constructs a cost function Cost(i,j) according to the land type weight and terrain information weight of the target area, which is shown by the following formula:

[0111] Cost(i,j)=w land *L ij +w elev *ΔH ij +*w slope +tan(θ ij )

[0112] In the formula, L ij is the land type weight of the grid (i,j); ΔH ij is the elevation difference between adjacent grids; θ ij is the grid slope value, and tan(θ) is used to quantify the slope influence; w land , w elev , w slope are weight coefficients (which need to be determined through historical data) respectively.

[0113] And a heuristic function h(n) is constructed based on the distance between the current node and the target node in the target area, which is shown by the following formula:

[0114]

[0115] In the formula, (x n ,y n ) are the coordinates of the current node; (x goal ,y goal ) are the coordinates of the target node.

[0116] Set the centralized new energy power station as the starting node and the power access point in the target area as the ending node. With the goal of the shortest power line path in the target area, it is shown by the following formula:

[0117]

[0118] In the formula, L min is the length of the shortest power line path; d i is the length of each section of the line; n is the total number of line sections.

[0119] Then starting from the starting point, calculate the cost Cost(i,j) (cumulative movement cost) and heuristic value h(n) (straight-line distance to the end point) of the adjacent grids, select the node with the smallest total cost f(n)=Cost(i,j)+h(n) for expansion until reaching the end point, and trace back the parent node in reverse from the end point to generate the initial power line path.

[0120] By dividing the target area into grids, the present invention can simplify the complex geographical environment into a series of regular units. Each unit (grid) contains important weights of land type and terrain information, enabling path planning to more accurately consider factors such as terrain undulation and land use type, thereby improving the accuracy of planning; the cost function is constructed based on land type weights and terrain information weights, and can comprehensively reflect the cost when the path crosses different grids. The heuristic function estimates the potential length of the path according to the distance between the current node and the target node, which helps to guide the search process towards the target node. The combined use of the two can significantly improve the efficiency of path search, reduce unnecessary search space; enhance the adaptability and flexibility of path planning; the solution adopts a search algorithm based on the cost function and heuristic function, which can automatically find the optimal path without manual intervention, not only improving the planning efficiency, but also reducing the requirements for the professional skills of planners.

[0121] In an embodiment, step S3 further includes:

[0122] Taking the grid coordinates of the current node, the land type weights and terrain information weights of the adjacent grids as the state space, and taking the movement direction and the preset range of path adjustment as the action space;

[0123] Construct a reward function based on path length change, path turning times penalty, path tortuosity coefficient, and high-cost area crossing penalty;

[0124] Train the DQN model according to the state space, the action space, and the reward function, and use the initial power line path as the initial policy to optimize with the trained DQN model to obtain the optimized power line path.

[0125] Specifically, for the optimization of the initial power line path, the present invention uses the grid coordinates of the current node and the land type, elevation, and slope information of the surrounding grids (such as a 3×3 neighborhood) as the state space, and the moving directions: 8 adjacent grids (east, south, west, north, northeast, etc.) and the path adjustment preset range: allowing a small adjustment of the path point position (such as ±1 grid) as the action space; then based on the path length change (shortening is a positive reward), the number of path turning penalties (fewer times is a positive reward), the path tortuosity coefficient (decreasing is a positive reward), and the high-cost area crossing penalty (crossing a high-cost area is a negative reward) to construct a reward function, and uses the initial power line path as the initial strategy, executes actions in the simulation environment, collects state-action-reward sequences for interactive sampling, and maximizes the cumulative reward through gradient ascent, adjusts the parameters of the DQN model for training the DQN model until the preset convergence condition is reached or the maximum number of iterations is reached; finally, optimizes the initial power line path through the trained DQN model, that is, selects the maximum Q-value action for each state, generates a power line optimized path, and outputs it after B-spline smoothing and mechanical verification. In addition, the PPO (Proximal Policy Optimization) policy gradient algorithm can also be used to generate a power line optimized path by balancing exploration and exploitation.

[0126] The present invention realizes the intelligent optimization of the power line path through deep reinforcement learning, significantly exceeding traditional methods in dimensions such as path length, economy, and compliance, and is particularly suitable for complex terrain and multi-constraint scenarios.

[0127] In addition, for the path generation and optimization in step S3, the present invention can also be realized by a path optimization method based on graph theory and simulated annealing algorithm, including:

[0128] First, abstract the power line path planning problem into a graph theory problem. In this process, divide the target area into several grids, regard each grid as a node of the graph, and the connection relationship between the grids constitutes the edges of the graph. The weights of the edges are comprehensively determined according to the land type weights, terrain information weights, and the distances between the grids;

[0129] Then, after the graph theory model is constructed, generate an initial solution, that is, a path from the starting node to the ending node, by randomly selecting nodes or using other heuristic methods;

[0130] Then, the simulated annealing algorithm is used to search within the neighborhood of the initial solution, accepting inferior solutions with a certain probability, and finally converging to the global optimal solution or an approximate optimal solution: Generate a new solution within the neighborhood of the current solution, that is, try to change the selection of some nodes or edges on the path; According to the objective function value of the new solution (such as path length, number of turns, etc.) and the current temperature, decide whether to accept the new solution as the current solution based on the Metropolis criterion, that is, accept a worse solution with a certain probability to jump out of the local optimal solution; By gradually reducing the temperature, control the probability of accepting inferior solutions during the search process. The higher the temperature, the greater the probability of accepting inferior solutions; The lower the temperature, the smaller the probability of accepting inferior solutions.

[0131] Finally, repeat the neighborhood search, acceptance criterion judgment, and cooling strategy application until the termination condition is met (such as reaching the maximum number of iterations, the change in the solution is less than the preset threshold, etc.). During the iteration process, the path will gradually converge to the global optimal solution or an approximate optimal solution. When the termination condition is met, output the finally optimized power line path.

[0132] Through the construction of the graph theory model, the present invention abstracts the problem into a more easily handled graph theory problem; The application of the simulated annealing algorithm can search for the optimal solution globally, avoid the problem of falling into the local optimal solution, efficiently handle the path planning problem in a large-scale complex network environment, improve the path planning efficiency and flexibility, optimize the path quality, and enhance the robustness of the algorithm.

[0133] In an embodiment, step S4 includes:

[0134] Determine the positions of obstacles in each grid according to the environmental data to label each grid, and set the span standard based on the current voltage level;

[0135] Based on the span standard, use the greedy strategy to initially arrange the poles and towers for the optimized power line path, and optimize the initial arrangement result through the improved genetic algorithm to obtain the final power line path arrangement strategy for the target area.

[0136] Specifically, based on the obstacle positions (residential areas, nature reserves, high-voltage line intersections) included in the environmental data, each grid is marked with obstacles, and the span standard is set based on the current voltage level. For example, for a 110 kV voltage, the standard span is 300 - 400 m, the maximum allowable span is 450 m, and the priority for tower type selection is mainly straight towers, with tension towers spaced 3 - 5 spans apart; for a 220 kV voltage, the standard span is 400 - 500 m, the maximum allowable span is 550 m, and the priority for tower type selection is a ratio of straight towers to tension towers of 4:1; for a 500 kV voltage, the standard span is 500 - 700 m, the maximum allowable span is 800 m, and the priority for tower type selection is an increase in the proportion of tension towers to 30%. Then, based on the span standard, the first tower (tension tower) is set at the starting point of the path, with the marked coordinates being T0(x0, y0). Then, starting from T i extend in the path direction and calculate the maximum allowable span D max :

[0137] D max = min(standard span, terrain-limited span, obstacle safety distance)

[0138] And within the range of D max select the optimal position T i+1 , and the objective function is:

[0139] f(T i+1 ) = a1 * land cost + a2 * slope penalty + a3 * construction accessibility

[0140] In the formula, a1, a2, and a3 are weight coefficients, which can be calculated from historical data.

[0141] Then, chromosome coding is performed on the tower position sequence, and the total cost required to set up the towers and the safety margin of the tower setting scheme are used as the fitness function. The cross operation is to exchange the tower position subsequences of the two paths, and the mutation operation is to randomly fine-tune the tower positions (such as ±2 grids) or replace the tower type. The iterative process of the conventional genetic algorithm is carried out until the change in fitness between two adjacent times < 1% or the preset maximum number of iterations is reached. The final power line path layout strategy for the target area is output, and the proportion of various land types, line length, estimated cost, and tortuosity coefficient of this scheme are automatically calculated.

[0142] Through the three-step closed-loop of environmental data-driven, voltage level rule constraint, and intelligent algorithm optimization, the present invention realizes the leap from rough to fine tower arrangement, making the final path not only meet the mechanical safety and construction requirements, but also significantly reduce the project cost, and is applicable to the power grid construction in complex terrain areas.

[0143] In the embodiments of the present application, in view of the problem of how to improve the generation efficiency and accuracy of the power transmission lines of centralized new energy power stations, a method for generating the path of centralized new energy power lines is designed. By obtaining multiple historical line path cases and their case parameters in the area where the centralized new energy power station is located and its nearby areas, valuable data support is provided for subsequent weight evaluation and path planning. Based on the obtained historical data, the land type weight and terrain information weight of the target area are quantified by using the traversal method to ensure that the influence of terrain and land type on line construction can be fully considered during path planning. According to the quantified weight information, with the goal of the shortest power line path in the target area, a heuristic search and a machine learning algorithm are combined to generate the power line path, improving the efficiency and accuracy of path planning. According to the environmental data of the target area and the current voltage level, the pole towers are arranged for the optimized path of the power line to generate the final power line path layout strategy for the target area, thereby ensuring the feasibility and economy of the line in actual construction. Another method for generating the path of centralized new energy power lines is as Figure 2 shown. Through the combination of historical data-driven, multiple intelligent algorithms and intelligent layout strategies, the present invention realizes the full-process automation of the design of the transmission lines of new energy power stations, significantly improves the strategy generation efficiency and accuracy, and further reduces the cost.

[0144] It should be noted that although the steps in the above flowcharts are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders.

[0145] In another embodiment, as Figure 3 shown, the second aspect of the present invention provides a system for generating the path of centralized new energy power lines, including:

[0146] A data acquisition module 10, configured to acquire multiple historical line path cases and their case parameters in a preset area of a centralized new energy power station;

[0147] A weight quantification module 20, configured to quantify the land type weight and terrain information weight of the target area corresponding to the centralized new energy power station by using the traversal method according to each of the historical line path cases and their case parameters;

[0148] A path generation module 30, configured to generate and optimize an initial power line path based on the land type weight and the terrain information weight, with the goal of the shortest power line path in the target area, and obtain an optimized power line path through the A star algorithm and the reinforcement learning algorithm;

[0149] The tower arrangement module 40 is configured to arrange towers for the optimized path of the power line according to the environmental data of the target area and the current voltage level, and generate a final power line path arrangement strategy for the target area.

[0150] It should be noted that each module in the above centralized new energy power line path generation system can be implemented in whole or in part by software, hardware, and their combinations. The above modules can be embedded in the processor of the computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above respective modules. For the specific limitations of a centralized new energy power line path generation system, refer to the limitations on a centralized new energy power line path generation method in the above text. The two have the same functions and effects and will not be elaborated here.

[0151] The third aspect of the present invention provides an electronic device, which includes:

[0152] A processor, a memory, and a bus;

[0153] The bus is used to connect the processor and the memory;

[0154] The memory is used to store operation instructions;

[0155] The processor is configured to execute, by calling the operation instructions, instructions to cause the processor to perform operations corresponding to a centralized new energy power line path generation method as shown in the first aspect of the present application.

[0156] In an alternative embodiment, an electronic device is provided, as Figure 4 shown, Figure 4 The electronic device 5000 shown includes a processor 5001 and a memory 5003. Among them, the processor 5001 and the memory 5003 are connected, such as through a bus 5002. Optionally, the electronic device 5000 may further include a transceiver 5004. It should be noted that in actual applications, the transceiver 5004 is not limited to one, and the structure of the electronic device 5000 does not constitute a limitation to the embodiments of the present application.

[0157] The processor 5001 can be a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logical blocks, modules, and circuits described in combination with the disclosed content of the present application. The processor 5001 can also be a combination that realizes computing functions, such as a combination including one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0158] The bus 5002 may include a path for transmitting information between the above components. The bus 5002 can be a PCI bus, an EISA bus, or the like. The bus 5002 can be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity of representation, Figure 4 it is only represented by a thick line in Figure 4 , but it does not mean that there is only one bus or one type of bus.

[0159] The memory 5003 can be a ROM or other types of static storage devices that can store static information and instructions, a RAM or other types of dynamic storage devices that can store information and instructions, or it can also be an EEPROM, a CD-ROM, or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but not limited thereto.

[0160] The memory 5003 is used to store the application program code for implementing the solution of this application and is controlled by the processor 5001 to execute. The processor 5001 is used to execute the application program code stored in the memory 5003 to implement the content shown in any of the foregoing method embodiments.

[0161] Among them, the electronic device includes but is not limited to: mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc.

[0162] The fourth aspect of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements a centralized new energy power line path generation method shown in the first aspect of this application.

[0163] Another embodiment of this application provides a computer-readable storage medium, on which a computer program is stored, and when it runs on a computer, it enables the computer to execute the corresponding content in the foregoing method embodiments.

[0164] In addition, an embodiment of the present invention also proposes a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the steps of the above method.

[0165] In summary, the present invention relates to the technical field of electric power energy, and discloses a method, system, device and medium for generating a path of a centralized new energy power line. The method includes obtaining a plurality of historical line path cases and their case parameters within a preset area of a centralized new energy power station; according to each of the historical line path cases and their case parameters, using the traversal method to quantify the land type weight and the terrain information weight of the target area corresponding to the centralized new energy power station; based on the land type weight and the terrain information weight, with the goal of the shortest power line path in the target area, generating and optimizing an initial power line path through the A star algorithm and the reinforcement learning algorithm to obtain an optimized power line path; arranging poles and towers for the optimized power line path according to the environmental data and the current voltage level of the target area, and generating a final power line path layout strategy for the target area, effectively improving the generation efficiency and accuracy of the power transmission line of the centralized new energy power station.

[0166] Each embodiment in this specification is described in a progressive manner. For parts that are the same or similar in each embodiment, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiment. It should be noted that the above technical features of the embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the above technical features in the embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0167] The above embodiments only represent several preferred implementation manners of the present application, and the description is relatively specific and detailed, but it cannot be understood as a limitation on the scope of the invention patent. It should be pointed out that for those of ordinary skill in the art in this technical field, without departing from the technical principle of the present invention, several improvements and substitutions can still be made, and these improvements and substitutions should also be regarded as the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for generating a path of a centralized new energy power line, characterized in that, Including: Obtain multiple historical line path cases and their case parameters within a preset area of a centralized new energy power station; According to each of the historical line path cases and their case parameters, use the traversal method to quantify the land type weight and terrain information weight of the target area corresponding to the centralized new energy power station; Based on the land type weight and the terrain information weight, with the goal of the shortest power line path in the target area, generate and optimize the initial power line path through the Astar algorithm and the reinforcement learning algorithm to obtain the optimized power line path; Arrange the power line towers according to the environmental data and the current voltage level of the target area for the optimized power line path to generate the final power line path arrangement strategy for the target area.

2. The centralized new energy power line path generation method according to claim 1, wherein The step of using the traversal method to quantify the land type weight and terrain information weight of the target area corresponding to the centralized new energy power station according to each of the historical line path cases and their case parameters includes: Perform clustering analysis on each of the case parameters through a deep learning algorithm to obtain a clustering model, and input the design requirement information of the target area into the clustering model for processing to obtain a historical line path target case that matches the design requirement information; Based on each of the historical line path target cases, use the traversal method to quantify the land type weight and terrain information weight of the target area.

3. The centralized new energy power line path generation method according to claim 2, characterized in that, The step of using the traversal method to quantify the land type weight and terrain information weight of the target area based on each of the historical line path target cases includes: Obtain the actual curve of the transmission path and the actual land type data in each of the historical line path target cases; Set a first weight search space to traverse all land type weights for each of the historical line path target cases to obtain a first candidate weight combination; Based on each of the first candidate weight combinations, use a path planning algorithm to generate a first virtual curve of the transmission path for each of the historical line path target cases under each of the first candidate weights to quantify the similarity distance with each of the actual curves of the transmission path, and obtain a first similarity calculation result; According to the first similarity calculation result, take the first candidate weight with the smallest similarity distance as the land type weight of the corresponding historical line path target case; Obtain the current land type data of the target area in the design requirement information to quantify the similarity distance with the actual land type data of each of the historical line path target cases, and obtain a second similarity calculation result; According to the second similarity calculation result, adjust the land type weight of the historical line path target case with the smallest similarity distance to be used as the land type weight of the target area.

4. A centralized new energy power line path generation method according to claim 3, characterized in that The step of using the traversal method to quantify the land type weight and terrain information weight of the target area based on each of the historical line path target cases further includes: Obtain the actual terrain information data in each of the historical line path target cases; Set a second weight search space to traverse all terrain information weights for each of the historical line path target cases to obtain a second candidate weight combination; Based on each of the second candidate weight combinations, use a path planning algorithm to generate second virtual curves of the delivery paths of each of the historical route path target cases under each second candidate weight, so as to quantify the similarity distance with the actual curves of each delivery path, and obtain a third similarity calculation result; According to the third similarity calculation result, use the second candidate weight with the smallest similarity distance as the terrain information weight corresponding to the historical route path target case; Obtain the current terrain information data of the target area in the design requirement information, so as to quantify the similarity distance with the actual terrain information data of each historical route path target case, and obtain a fourth similarity calculation result; According to the fourth similarity calculation result, use the terrain information weight of the historical route path target case with the smallest similarity distance as the terrain information weight of the target area.

5. A method for generating a path of a centralized new energy power line according to claim 1, characterized in that, Based on the land type weight and the terrain information weight, with the goal of the shortest power line path in the target area, generate and optimize an initial power line path through the Astar algorithm and the reinforcement learning algorithm to obtain an optimized power line path, including: Divide the target area into grids, and label each divided grid with the corresponding land type weight and terrain information weight based on the land type weight and terrain information weight of the target area; Construct a cost function according to the land type weight and terrain information weight of the target area, and construct a heuristic function based on the distance between the current node and the target node in the target area; Set the initial node and the termination node of the target area, with the goal of the shortest power line path in the target area, and based on the cost function and the heuristic function, calculate the cost and heuristic value of adjacent grids starting from the initial node, and expand the node with the smallest total cost until reaching the termination node; Trace back to the parent node from the termination node to generate an initial power line path.

6. A method for generating a path of a centralized new energy power line according to claim 5, characterized in that, Based on the land type weight and the terrain information weight, with the goal of the shortest power line path in the target area, generate and optimize an initial power line path through the Astar algorithm and the reinforcement learning algorithm to obtain an optimized power line path, further including: Use the grid coordinates of the current node and the land type weight and terrain information weight of adjacent grids as the state space, and use the moving direction and the preset range of path adjustment as the action space; Construct a reward function based on the path length change, the path turning times penalty, the path tortuosity coefficient, and the high-cost area crossing penalty; Train the DQN model according to the state space, the action space, and the reward function, and use the initial power line path as the initial strategy to optimize with the trained DQN model to obtain an optimized power line path.

7. A method for generating a path of a centralized new energy power line according to claim 5, characterized in that The arranging of power line towers for the optimized power line path according to the environmental data and the current voltage level of the target area to generate a final power line path arrangement strategy for the target area includes: Determine the positions of obstacles in each grid according to the environmental data to label each grid, and set the span standard based on the current voltage level; Based on the span standard, the initial arrangement of poles for the optimized path of the power line is carried out using a greedy strategy, and the initial arrangement result is optimized by an improved genetic algorithm to obtain the final power line path arrangement strategy for the target area.

8. A centralized new energy power line path generation system, characterized in that, It includes: A data acquisition module, configured to acquire multiple historical line path cases and their case parameters within a preset area of a centralized new energy power station; A weight quantification module, configured to quantify the land type weight and terrain information weight of the target area corresponding to the centralized new energy power station by using the traversal method according to each of the historical line path cases and their case parameters; A path generation module, configured to generate and optimize an initial power line path based on the land type weight and the terrain information weight, with the goal of the shortest power line path in the target area, to obtain an optimized power line path through the Astar algorithm and the reinforcement learning algorithm; A pole arrangement module, configured to arrange poles for the optimized power line path according to the environmental data of the target area and the current voltage level, and generate the final power line path arrangement strategy for the target area.

9. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the centralized new energy power line path generation method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program. When the device where the computer-readable storage medium is located executes the computer program, it implements the centralized new energy power line path generation method according to any one of claims 1 to 7.

Images