Intelligent transmission line route selection method based on adaptive node optimization ant colony algorithm

Abstract

The application relates to a power transmission line intelligent route selection method based on an adaptive node optimization ant colony algorithm, which comprises the following steps: based on geographic information system, environment factors are determined, a hierarchical analytic hierarchy process model is established, and various environment factors are quantified; based on a GIS grid map, a geographic space is divided into a regular matrix composed of multiple square units, power transmission line construction costs of each basic area are obtained by combining the hierarchical analytic hierarchy process model, and total route costs are obtained; according to the total route costs and the GIS grid map, intelligent route selection is carried out through the adaptive node optimization ant colony algorithm, and an optimal path is obtained. Compared with a traditional ant colony algorithm used for intelligent route selection, the ant colony algorithm used in the application increases adaptive adjustment, local variation and node optimization mechanisms, and the convergence precision and stability of the algorithm are improved; the application realizes a new overhead power line automatic optimization wiring method, and is helpful for automatically selecting the most economical route to establish a new line.

Description

Technical Field

[0001] This invention relates to the field of intelligent transmission line selection technology, and in particular to an intelligent transmission line selection method based on an adaptive node optimization ant colony algorithm. Background Technology

[0002] With the deepening of my country's power industry system reform, the systems and methods of power engineering project management are constantly being explored and improved. However, due to the relatively short time since the power system reform, the construction management models and methods of power engineering projects are still lagging behind in many aspects. In particular, the application of scientific project management concepts in power transmission and transformation projects is still in the exploratory stage. Therefore, introducing the concept of optimized management into the scientific management of power transmission and transformation projects will become a focus of research for power system research institutes.

[0003] Currently, there is relatively little research on combining GIS with intelligent algorithms for power transmission engineering. In addition, the accurate quantification of environmental factors when establishing geographic information models and the applicability of the intelligent algorithms used to geographic information models are not high enough.

[0004] Traditional transmission line route planning typically involves four steps: indoor route selection, data collection, preliminary survey, and final route selection. Indoor route selection focuses on marking the start and end points on a topographic map of the planning area, comprehensively considering the actual terrain, avoiding protected areas and insurmountable obstacles, and adhering to line design principles to create multiple feasible routes as preliminary survey options, prioritizing the shortest distance between the start and end points. The preliminary survey primarily involves a comprehensive comparison of technical difficulties and economic factors, including line length and cross-regional obstacles. Final route selection requires revising the route plan based on actual conditions, considering the economic viability of corner towers, to determine the final route. However, the complexity of the geographical environment makes it impossible to guarantee real-time updates of the topographic map, leading to environmental information biases and ultimately, a final plan that fails to meet the needs of the power grid project. Summary of the Invention

[0005] The purpose of this invention is to provide a method for intelligent transmission line selection based on an adaptive node optimization ant colony algorithm, which enables automatic optimization of new overhead power line routing and helps to automatically select the most economical route for establishing new lines.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a method for intelligent transmission line selection based on an adaptive node-optimized ant colony algorithm, the method comprising the following sequential steps:

[0007] (1) Based on the geographic information system, identify environmental factors, establish a hierarchical analysis model, and quantify each environmental factor;

[0008] (2) Based on the GIS raster map, the geographic space is divided into a regular matrix composed of multiple square units. The corresponding two-dimensional coordinate system is defined. Each square unit represents a basic area and location. The construction cost of the transmission line in each basic area is obtained by combining the hierarchical analysis model, and the total cost of the route is obtained.

[0009] (3) Based on the total cost of the route and the GIS raster map, the optimal path best_path is obtained by intelligent route selection through the adaptive node optimization ant colony algorithm.

[0010] Step (1) specifically includes the following steps:

[0011] (1a) Geographic information in the topographic map obtained from the GIS raster map, including geological type, crossings, building facilities and guide costs;

[0012] (1b) The four types of geographic information in step (1a) are used as four criteria. The hierarchical analysis model of environmental factors for power transmission line selection is established by using the analytic hierarchy process (AHP). That is, the four criteria are used as criteria layers to determine the construction cost. At the same time, each criterion is affected by multiple environmental factors, including forests, construction land, water areas, nature reserves, houses, public facilities and roads.

[0013] (1c) Quantify the environmental factors under a certain criterion according to the 1-9 labeling method in AHP to obtain evaluation data. Then, construct a comprehensive judgment matrix B based on the existing principal component analysis method PCA. Solve the weight of the environmental factors under a certain criterion through the comprehensive judgment matrix B, and then solve the comprehensive weight vector of each element to the total target construction cost.

[0014] Step (2) specifically includes the following steps:

[0015] (2a) Based on the GIS raster map, the geographic space is divided into a regular matrix of square units, each unit representing a basic area and location; for a given GIS raster map, a corresponding two-dimensional coordinate system is defined to locate each raster area, i.e. a discrete two-dimensional space.

[0016] (2b) In a GIS raster map, assuming a route passes through k+1 basic cells, it is represented by the set R:

[0017] R = {r0, ..., r} k ,...,r K}

[0018] ={(i0, j0), ..., (i k j k ), ..., (i K jK )}

[0019] In the formula, (i, j) represents the basic unit of the vertex in the selected path, i.e., the geographical location information. The line segment l is defined. k It is vertex r k-1 and r k If lines are connected between sets R, then the paths formed by these lines are represented by the set L:

[0020] L = {l1, ..., l} k , ..., l K};

[0021] (2c) Calculate the total cost of a route using the following formula:

[0022]

[0023] Wherein, g(r) k The basic region r is obtained from the hierarchical analysis model. k The construction cost of the power transmission line, h(l k ) is a line segment l k The cost of the route length is pt, and the cost of the route corner is pt.

[0024] Step (3) specifically includes the following steps:

[0025] (3a) Each value in the pheromone matrix η corresponds to the pheromone value of each basic area of ​​the GIS raster map. Initialize the pheromone matrix η = η0, the number of iterations K, the number of searches per group M, and initialize the neighbor raster cost influence factor α, the pheromone influence factor β, the pheromone volatility factor ρ, the pheromone intensity constant Q, the number of mutations n, and the best path best_path.

[0026] (3b) Calculate P based on the probability of the next node. i By using a roulette wheel to choose the next move, one can find the next position.

[0027] (3c) Determine the path points and path length, update the pheromone variable according to the adaptive mechanism, perform adaptive mutation on the path marked with the initial solution, determine the mutation length l according to the path length L′ of the previous iteration, randomly generate mutation points {b1, b2}, and then perform n local path searches with b1 as the starting point and b2 as the ending point, and record the mutated search path path. b And compare the mutation search path with historical path information;

[0028] (3d) Update path cost values ​​and taboo tables, record and update historical local best and worst paths, and update pheromones;

[0029] (3e) Record the search path and path cost of each ant in each iteration and update the current optimal search path. Determine whether the loop has been completed. If not, return to step (3b); otherwise, proceed to step (3f).

[0030] (3f) After the iteration is completed, output the current best path, best_path;

[0031] (3g) Optimize the nodes of the best path best_path and output the result.

[0032] The step (1c) specifically includes the following steps:

[0033] (1c1) The weights are solved using the following formula, that is, by solving for the largest eigenvalue and its corresponding eigenvector of the comprehensive judgment matrix B:

[0034]

[0035] Where, a′ ij a ij These are the corresponding elements of the two judgment matrices, b ij To comprehensively determine the corresponding elements of matrix B;

[0036] (1c2) The weights are solved using the following formula, that is, by solving for the largest eigenvalue and its corresponding eigenvector of the comprehensive judgment matrix B:

[0037] Bζ=λζ

[0038] Where ζ is the eigenvector and λ is the eigenvalue corresponding to the eigenvector; thus, the weight values ​​of each environmental factor under a certain criterion can be obtained;

[0039] (1c3) The combined weight vector of each element with respect to the overall objective is calculated using the following formula:

[0040]

[0041] in, This represents the weight value of each factor in the i-th layer to the overall objective. Let be the weight values ​​of all elements in the i-th layer with respect to each element in the j-th layer, and Let represent the weight value of the factor in layer i to layer i-1 with the j-th factor as the criterion. Calculate the weight value of each element relative to the target layer.

[0042] The method for calculating the probability of the next node in step (3b) specifically includes the following steps:

[0043] (3b1) The cost value of the path selected by the ant colony algorithm is calculated by combining the average cost value of the grid cells crossed by the line segment with the cost value of the unit transmission line as the cost value per unit distance of the path; the following formula is used to calculate the cost value of the ant from the grid position p. k-1 Reaching p k The cost per unit length of the path τ′(p) k-1 p k ):

[0044]

[0045] Among them, w(p) k ) for spanning grid p k Additional cost value, l s γ represents the path length traversed, and γ is the penalty coefficient.

[0046] (3b2) The probability of the improved ant colony algorithm being selected is obtained using the following formula.

[0047]

[0048] Among them, [η′(p i p j )] is from p i Transfer to p j Based on the pheromone concentration, the search is performed using the probability formula described above from the starting point. In the k-th iteration, the selectable neighboring grids for the next iteration are determined, and their positions at the next moment are determined by the roulette wheel method.

[0049] The step (3c) of updating the pheromone variable according to the adaptive mechanism specifically includes the following steps:

[0050] (3c1) The traditional ant colony algorithm updates the pheromone according to the local update rule, that is, after all ants have completed a transfer, it uses the following formula to update the pheromone:

[0051] η′ ij (k+1)=(1-ρ)η ij (k)+Δη ij (k)

[0052] Where ρ∈(0,1) is the pheromone volatile factor, η ij (k) is the accumulated pheromone concentration in the kth cycle;

[0053] (3c2) An improved global update rule is used, namely an update rule based on an adaptive mechanism, which incorporates each ant's perception of its surrounding environment. The pheromone concentration on each path is updated according to the existing wolf pack allocation principle. The update method is as follows, and the pheromone concentration of the optimal path and the local worst path is calculated using the following formula:

[0054]

[0055]

[0056] Where Q represents the pheromone intensity constant, and The costs (k) are calculated based on the locally optimal path through ij and the locally worst path through ij, respectively. m Let cost(k) represent the path cost of the m-th ant in the k-th iteration that passes through ant_ij. b and cost(k) w Let $i$ represent the local optimal path cost and the worst path cost in the kth iteration that passes through path $ij$.

[0057] (3c3) The pheromone is updated using the following formula:

[0058]

[0059] Where, Δη ij m(k) represents the pheromone concentration of the m-th ant on path ij in the k-th cycle.

[0060] The step (3g) specifically includes the following steps:

[0061] (3g1) Input grid and path information: the sequence of path points and the route from the starting point S to the target point E, denoted as queue = {s1, s2, ..., s}. n Let P be the route and area be the area that cannot be crossed on the map. Initialize i = 1, j = 1.

[0062] (3g2) Determine s j Is it the target point E? If so, then s j Add to the optimal path list to end the optimization; otherwise, add s i Add to the best path list best_list, proceed to step (3g3);

[0063] (3g3) Connection path s i s j+2 Calculations and node optimizations are performed on its paths;

[0064] (3g4) If the line does not pass through any insurmountable obstacle, then j = j + 1, and return to step (3g2); otherwise, return to step (3g5).

[0065] (3g5) path s i s j+2 After passing through insurmountable obstacles, compare paths (s)i ,s j+1 )+(s j+1 ,s j+2 ) and path (s i ,s j+2 If the cost value is smaller than the former, then i = j + 1 and return to step (3g2); otherwise, i = j + 2 and return to step (3g2). Steps (3c) and (3g) constitute the adaptive node optimization ant colony algorithm.

[0066] As can be seen from the above technical solution, the beneficial effects of the present invention are as follows: First, compared with the traditional ant colony algorithm for intelligent route selection, the ant colony algorithm used in the present invention adds adaptive adjustment, local mutation, and node optimization mechanisms, thereby improving the convergence accuracy and stability of the algorithm; Second, the intelligent route selection algorithm proposed in the present invention is based on the problem of optimizing line cost, and utilizes the computing power and visualization representation capabilities of geographic information systems to realize a new automatic optimization method for overhead power line routing, which helps to automatically select the most economical route to establish new lines; Third, the present invention can significantly reduce the workload of on-site surveys and manual labor, effectively improve the efficiency and effectiveness of power grid construction, and reduce the investment in route selection costs. Attached Figure Description

[0067] Figure 1 This is a flowchart of the method of the present invention;

[0068] Figure 2 This is a modeling diagram of the optimal route selection path provided in an embodiment of the present invention. Detailed Implementation

[0069] like Figure 1 As shown, a method for intelligent transmission line selection based on an adaptive node optimization ant colony algorithm is presented. This method includes the following steps in sequence:

[0070] (1) Based on the geographic information system, identify environmental factors, establish a hierarchical analysis model, and quantify each environmental factor;

[0071] (2) Based on the GIS raster map, the geographic space is divided into a regular matrix composed of multiple square units. The corresponding two-dimensional coordinate system is defined. Each square unit represents a basic area and location. The construction cost of the transmission line in each basic area is obtained by combining the hierarchical analysis model, and the total cost of the route is obtained.

[0072] (3) Based on the total cost of the route and the GIS raster map, the optimal path best_path is obtained by intelligent route selection through the adaptive node optimization ant colony algorithm.

[0073] Step (1) specifically includes the following steps:

[0074] (1a) Geographic information in the topographic map obtained from the GIS raster map, including geological type, crossings, building facilities and guide costs;

[0075] (1b) The four types of geographic information in step (1a) are used as four criteria. The hierarchical analysis model of environmental factors for power transmission line selection is established by using the analytic hierarchy process (AHP). That is, the four criteria are used as criteria layers to determine the construction cost. At the same time, each criterion is affected by multiple environmental factors, including forests, construction land, water areas, nature reserves, houses, public facilities and roads.

[0076] (1c) Quantify the environmental factors under a certain criterion according to the 1-9 labeling method in AHP to obtain evaluation data. Then, construct a comprehensive judgment matrix B based on the existing principal component analysis method PCA. Solve the weight of the environmental factors under a certain criterion through the comprehensive judgment matrix B, and then solve the comprehensive weight vector of each element to the total target construction cost.

[0077] Many environmental factors need to be considered in the selection of power transmission lines. Based on the geographical information in the topographic map obtained by GIS, this invention mainly considers aspects such as geological type, crossings, building facilities and conductor costs. In the process of power transmission line planning, important areas or areas that cannot be crossed should be avoided as much as possible. For example, it is forbidden to cross important units such as airports and large factories. In addition, the demolition of residential buildings and deforestation should also be considered when selecting the route to reduce the economic compensation of the line project.

[0078] like Figure 2 As shown, step (2) specifically includes the following steps:

[0079] (2a) Based on the GIS raster map, the geographic space is divided into a regular matrix of square units, each unit representing a basic area and location; for a given GIS raster map, a corresponding two-dimensional coordinate system is defined to locate each raster area, i.e. a discrete two-dimensional space.

[0080] (2b) In a GIS raster map, assuming a route passes through k+1 basic cells, it is represented by the set R:

[0081] R = {r0,..,r} k ,...,r K}

[0082] ={(i0,j0),...,(i k ,j k ),...,(i K ,j K )}

[0083] In the formula, (i,j) represents the basic unit of the vertex in the selected path, i.e., the geographical location information. The line segment l is defined. k It is vertex r k-1 and r k If lines are connected between sets R, then the paths formed by these lines are represented by the set L:

[0084] L={l1,..,l k ,...,l K};

[0085] (2c) Calculate the total cost of a route using the following formula:

[0086]

[0087] Wherein, g(r) k The basic region r is obtained from the hierarchical analysis model. k The construction cost of the power transmission line, h(l k ) is a line segment l k Line length cost, p t It is the cost of turning points on the route.

[0088] Step (3) specifically includes the following steps:

[0089] (3a) Each value in the pheromone matrix η corresponds to the pheromone value of each basic area of ​​the GIS raster map. Initialize the pheromone matrix η = η0, the number of iterations K, the number of searches per group M, and initialize the neighbor raster cost influence factor α, the pheromone influence factor β, the pheromone volatility factor ρ, the pheromone intensity constant Q, the number of mutations n, and the best path best_path.

[0090] (3b) Calculate P based on the probability of the next node. i By using a roulette wheel to choose the next move, one can find the next position.

[0091] (3c) Determine the path points and path length, update the pheromone variable according to the adaptive mechanism, perform adaptive mutation on the path marked with the initial solution, determine the mutation length l according to the path length L′ of the previous iteration, randomly generate mutation points {b1, b2}, and then perform n local path searches with b1 as the starting point and b2 as the ending point, and record the mutated search path path. b And compare the mutation search path with historical path information;

[0092] (3d) Update path cost values ​​and taboo tables, record and update historical local best and worst paths, and update pheromones;

[0093] (3e) Record the search path and path cost of each ant in each iteration and update the current optimal search path. Determine whether the loop has been completed. If not, return to step (3b); otherwise, proceed to step (3f).

[0094] (3f) After the iteration is completed, output the current best path, best_path;

[0095] (3g) Optimize the nodes of the best path best_path and output the result.

[0096] The step (1c) specifically includes the following steps:

[0097] (1c1) The weights are solved using the following formula, that is, by solving for the largest eigenvalue and its corresponding eigenvector of the comprehensive judgment matrix B:

[0098]

[0099] Where, a′ ij a ij These are the corresponding elements of the two judgment matrices, b ij To comprehensively determine the corresponding elements of matrix B;

[0100] (1c2) The weights are solved using the following formula, that is, by solving for the largest eigenvalue and its corresponding eigenvector of the comprehensive judgment matrix B:

[0101] Bζ=λζ

[0102] Where ζ is the eigenvector and λ is the eigenvalue corresponding to the eigenvector; thus, the weight values ​​of each environmental factor under a certain criterion can be obtained;

[0103] (1c3) The combined weight vector of each element with respect to the overall objective is calculated using the following formula:

[0104]

[0105] in, This represents the weight value of each factor in the i-th layer to the overall objective. Let be the weight values ​​of all elements in the i-th layer with respect to each element in the j-th layer, and Let represent the weight value of the factor in layer i to layer i-1 with the j-th factor as the criterion. Calculate the weight value of each element relative to the target layer.

[0106] The method for calculating the probability of the next node in step (3b) specifically includes the following steps:

[0107] (3b1) The cost value of the path selected by the ant colony algorithm is calculated by combining the average cost value of the grid cells crossed by the line segment with the cost value of the unit transmission line as the cost value per unit distance of the path; the following formula is used to calculate the cost value of the ant from the grid position p. k-1 Reaching p k The cost per unit length of the path τ′(p) k-1 ,p k ):

[0108]

[0109] Among them, w(p) k ) for spanning grid p k Additional cost value, l s γ represents the path length traversed, and γ is the penalty coefficient.

[0110] (3b2) The probability of the improved ant colony algorithm being selected is obtained using the following formula.

[0111]

[0112] Among them, [η′(p i ,p j )] is from p i Transfer to p j Based on the pheromone concentration, the search is performed using the probability formula described above from the starting point. In the k-th iteration, the selectable neighboring grids for the next iteration are determined, and their positions at the next moment are determined by the roulette wheel method.

[0113] The step (3c) of updating the pheromone variable according to the adaptive mechanism specifically includes the following steps:

[0114] (3c1) The traditional ant colony algorithm updates the pheromone according to the local update rule, that is, after all ants have completed a transfer, it uses the following formula to update the pheromone:

[0115] η′ ij (k+1)=(1-ρ)η ij (k)+Δη ij (k)

[0116] Where ρ∈(0,1) is the pheromone volatile factor, η ij (k) is the accumulated pheromone concentration in the kth cycle;

[0117] (3c2) An improved global update rule is used, namely an update rule based on an adaptive mechanism, which incorporates each ant's perception of its surrounding environment. The pheromone concentration on each path is updated according to the existing wolf pack allocation principle. The update method is as follows, and the pheromone concentration of the optimal path and the local worst path is calculated using the following formula:

[0118]

[0119]

[0120] Where Q represents the pheromone intensity constant, and The costs (k) are calculated based on the locally optimal path through ij and the locally worst path through ij, respectively. m Let cost(k) represent the path cost of the m-th ant in the k-th iteration that passes through ant_ij. b and cost(k) w Let $i$ represent the local optimal path cost and the worst path cost in the kth iteration that passes through path $ij$.

[0121] (3c3) The pheromone is updated using the following formula:

[0122]

[0123] Where, Δη ij m(k) represents the pheromone concentration of the m-th ant on path ij in the k-th cycle.

[0124] The step (3g) specifically includes the following steps:

[0125] (3g1) Input grid and path information: the sequence of path points and the route from the starting point S to the target point E, denoted as queue = {s1, s2, ..., s}. n Let P be the route and area be the area that cannot be crossed on the map. Initialize i = 1, j = 1.

[0126] (3g2) Determine s j Is it the target point E? If so, then s j Add to the optimal path list to end the optimization; otherwise, add s i Add to the best path list best_list, proceed to step (3g3);

[0127] (3g3) Connection path s i s j+2 Calculations and node optimizations are performed on its paths;

[0128] (3g4) If the line does not pass through any insurmountable obstacle, then j = j + 1, and return to step (3g2); otherwise, return to step (3g5).

[0129] (3g5) path s i s j+2 After passing through insurmountable obstacles, compare paths (s)i ,s j+1 )+(s j+1 ,s j+2 ) and path (s i ,s j+2 If the former is smaller than the former, i = j + 1 and return to step (3g2); otherwise, i = j + 2 and return to step (3g2).

[0130] In summary, compared with traditional ant colony algorithms for intelligent route selection, the ant colony algorithm used in this invention adds adaptive adjustment, local mutation, and node optimization mechanisms, thereby improving the convergence accuracy and stability of the algorithm. The intelligent route selection algorithm proposed in this invention is based on the problem of optimizing route cost. It utilizes the computing power and visualization capabilities of geographic information systems to realize a new method for automatically optimizing the wiring of overhead power lines, which helps to automatically select the most economical route to establish new lines.

Claims

1. A power transmission line intelligent route selection method based on an adaptive node optimization ant colony algorithm, characterized in that: The method includes the following steps in sequence: (1) Based on the geographic information system, identify environmental factors, establish a hierarchical analysis model, and quantify each environmental factor; (2) Based on the GIS raster map, the geographic space is divided into a regular matrix composed of multiple square units, and a corresponding two-dimensional coordinate system is defined. Each square unit represents a basic area and location. The construction cost of the transmission line in each basic area is obtained by combining the hierarchical analysis model, and the total cost of the route is obtained. (3) According to the total cost of the route and the GIS grid map, intelligent route selection is carried out through an adaptive node optimization ant colony algorithm to obtain the optimal path ; Step (3) specifically includes the following steps: (3a) Pheromone matrix Each value in the matrix corresponds to the pheromone value of each basic area in the GIS raster map, and the pheromone matrix is ​​initialized. Number of iterations Number of searches per group Initialize the neighbor grid cost impact factor Pheromones influencing factors pheromone volatile factors pheromone intensity constant Number of mutations and optimal path ; (3b) Calculated based on the probability of the next node By using a roulette wheel to choose the next move, one can find the next position. (3c) Determine the waypoints and path lengths, update the pheromone variables according to the adaptive mechanism, and mark the paths of the initial solutions. Perform adaptive mutation based on the path length of the previous iteration. Determine the length of variation Randomly generate mutation points Then follow the instructions. Starting from, For the end point Secondary local path search, recording the search path after mutation. And compare the mutation search path with historical path information; (3d) Update path cost values ​​and taboo tables, record and update historical local best and worst paths, and update pheromones; (3e) Record the search path and path cost of each ant in each iteration and update the current optimal search path. Determine whether the loop has been completed. If not, return to step (3b); otherwise, proceed to step (3f). (3f) After the iteration is completed, output the current optimal path. ; (3g) for the optimal path Perform node optimization and output; The step (3c) of updating the pheromone variable according to the adaptive mechanism specifically includes the following steps: (3c1) The traditional ant colony algorithm updates the pheromone according to the local update rule, that is, after all ants have completed a transfer, it uses the following formula to update the pheromone: ； in, As a pheromone volatile factor, It is the first The pheromone concentration accumulated in the previous cycle; (3c2) An improved global update rule is used, namely an update rule based on an adaptive mechanism, which incorporates each ant's perception of its surrounding environment. The pheromone concentration on each path is updated according to the existing wolf pack allocation principle. The update method is as follows, and the pheromone concentration of the optimal path and the local worst path is calculated using the following formula: ； in, Represents the pheromone intensity constant. and They are based on the process Local optimal path and route The results were obtained by calculating the local worst-case path. Indicates the first In the second loop Only ants and passed by The path cost value, and They represent the first In the second loop and traversed by the path The cost values ​​of the locally optimal path and the cost values ​​of the worst path; (3c3) The pheromone is updated using the following formula: ； in, Indicates the first In the second loop Only ants on the path pheromone concentration; The method for calculating the probability of the next node in step (3b) specifically includes the following steps: (3b1) The cost value of the path selected by the ant colony algorithm is calculated by taking the average cost value of the grid crossings of the line segment and the cost value of the unit transmission line as the cost value per unit distance of the path; the following formula is used to calculate the cost value of the ant from the grid position. arrive Cost per unit length of path : ； in, To cross grid Additional cost value, The length of the path traversed. This is the penalty coefficient; (3b2) The probability of the improved ant colony algorithm being selected is obtained using the following formula. : ； in, From Transferred to The pheromone concentration is searched using the probability formula described above, starting from the initial point, and at the pheromone concentration... In the next iteration, the neighboring grid cells that can be selected for the next iteration are determined, and their positions at the next moment are determined by the roulette wheel method.

2. The intelligent transmission line selection method based on adaptive node optimization ant colony algorithm according to claim 1, characterized in that: Step (1) specifically includes the following steps: (1a) Geographic information in the topographic map obtained from the GIS raster map, including geological type, crossings, building facilities and traverse costs; (1b) The four types of geographic information in step (1a) are used as four criteria. The hierarchical analysis model of environmental factors for power transmission line selection is established by using the analytic hierarchy process (AHP). That is, the four criteria are used as criteria layers to determine the construction cost. At the same time, each criterion is affected by multiple environmental factors, including forests, construction land, water areas, nature reserves, houses, public facilities and roads. (1c) Quantify the environmental factors under a certain criterion according to the 1-9 labeling method in AHP to obtain evaluation data. Then, construct a comprehensive judgment matrix B based on the existing principal component analysis method PCA. Solve the weight of the environmental factors under a certain criterion through the comprehensive judgment matrix B, and then solve the comprehensive weight vector of each element to the total target construction cost.

3. The intelligent transmission line selection method based on adaptive node optimization ant colony algorithm according to claim 1, characterized in that: Step (2) specifically includes the following steps: (2a) Based on the GIS raster map, the geographic space is divided into a regular matrix of square units, each unit representing a basic area and location; for a given GIS raster map, a corresponding two-dimensional coordinate system is defined to locate each raster area, i.e. a discrete two-dimensional space. (2b) In a GIS raster map, assuming a route passes through k+1 basic cells, then use a set To indicate: ； In the formula, The basic unit representing the vertices in the selected path is the geographical location information, and the line segment is defined. It is the vertex and Connecting the two sets, then for each set... The paths formed are represented by a set. express: ； (2c) Calculate the total cost of a route using the following formula: ； in, Each basic region is obtained from a hierarchical analysis model. The construction cost of power transmission lines, It is a line segment Line length cost, It is the cost of turning points on the route.

4. The intelligent transmission line selection method based on adaptive node optimization ant colony algorithm according to claim 2, characterized in that: The step (1c) specifically includes the following steps: (1c1) The weights are solved using the following formula, that is, by solving for the largest eigenvalue and its corresponding eigenvector of the comprehensive judgment matrix B: ； in, , These are the corresponding elements of the two judgment matrices. To comprehensively judge the corresponding elements of matrix B; (1c2) The weights are solved using the following formula, that is, by solving for the largest eigenvalue and its corresponding eigenvector of the comprehensive judgment matrix B: in, For feature vectors, These are the eigenvalues ​​corresponding to the eigenvectors; from these, the weights of each environmental factor under a certain criterion can be calculated. (1c3) The comprehensive weight vector of each element with respect to the overall objective is calculated using the following formula: ； in, Indicates the first The weight values ​​of each factor in the overall objective. For the first All elements of the layer respectively for the first The weight values ​​of each element in the layer, and Indicates the first Layer factors Layer and first Using the criterion of weight values, calculate the weight value of each element relative to the target layer.

5. The intelligent transmission line selection method based on adaptive node optimization ant colony algorithm according to claim 1, characterized in that: The step (3g) specifically includes the following steps: (3g1) Input grid and path information: starting point Target point The path point sequence and route, the sequence is denoted as The line is denoted as And the uncrossable areas in the map, initialized. , ; (3g2) Judgment Is it the target point? If so, Add to the optimal path list and end the optimization; otherwise, Add to the best path list best_list, proceed to step (3g3); (3g3) connection path , Calculations and node optimizations are performed on its paths; (3g4) If the line does not pass through any insurmountable obstacle, then If yes, return to step (3g2); otherwise, return to step (3g5). (3g5) path , After passing through insurmountable obstacles, compare paths. With path Between the cost values, if the former is smaller, Return to step (3g2); otherwise Return to step (3g2).

Images