An intelligent transmission line path planning method integrating pixel-level image segmentation and ant colony algorithm

Abstract

The present invention proposes an intelligent transmission line path planning method that combines pixel-level image segmentation and the ant colony algorithm, including the following steps: Step S1: Obtain the data parameters of the candidate line area through a drone equipped with a three-dimensional lidar, combine them into a corresponding transmission line path data set, preprocess the transmission line path data set, and label the tags of different regions to obtain a corresponding preprocessed data set; Step S2: Use an image segmentation task-based deep neural network model to perform road network node calculation processing on the preprocessed data set to obtain suitable node targets; Step S3: Use the ant colony algorithm to obtain the shortest path between nodes to achieve intelligent transmission line path planning. Step S4: Visualize the algorithm results and provide an external use interface.

Description

Technical Field

[0001] The present invention belongs to the technical fields of transmission line engineering design and computer vision, and particularly relates to an intelligent transmission line path planning method integrating pixel-level image segmentation and ant colony algorithm. Background Art

[0002] The development of society and cities is inseparable from the supply and replenishment of electric power resources. In order to better meet the sustainable social and economic development of the region, detailed and comprehensive power planning work needs to be carried out in specific regions to ensure the power supply demand in the later stage of regional development and improve the power supply efficiency of the regional transmission lines. For this reason, it is necessary to carry out the spatial calculation of the service nodes of the distribution and transmission lines, analyze the power supply capacity of the planning area from the geographical space field, and meet the power supply demand in line with the regional economic development.

[0003] In the process of calculating the service nodes of the transmission lines, the specific actual situation of the power supply area is often compared and analyzed through the calculated installed capacity and load value table, ignoring the spatial geographical information of the planning area, making the node analysis and calculation have certain irrationality and affecting the distribution of the power supply capacity demand in the planning area. Summary of the Invention

[0004] In view of this, the present invention provides an intelligent transmission line path planning method integrating pixel-level image segmentation and ant colony algorithm, including the following steps: Step S1: Obtain the data parameters of the line selection area through a drone equipped with a three-dimensional lidar, combine them into a corresponding transmission line path data set, preprocess the transmission line path data set, and label the tags of different regions to obtain a corresponding preprocessed data set; Step S2: Adopt an image segmentation task-based deep neural network model to perform road network node calculation processing on the preprocessed data set to obtain suitable node targets; Step S3: Adopt the ant colony algorithm to obtain the shortest path between nodes to achieve intelligent transmission line path planning. Step S4: Visualize the algorithm results and provide an external use interface to be able to more finely plan the optimal path for the transmission line route selection.

[0005] The present invention specifically adopts the following technical solutions:

[0006] An intelligent transmission line path planning method integrating pixel-level image segmentation and ant colony algorithm, characterized by including the following steps:

[0007] Step S1: Obtain the data parameters of the line selection area through a drone equipped with a three-dimensional lidar, combine them into a corresponding transmission line path data set, preprocess the transmission line path data set, and label the tags of different regions to obtain a corresponding preprocessed data set;

[0008] Step S2: Use an image segmentation task-based deep neural network model to perform road network node calculation processing on the preprocessed data set obtained in Step S1, and obtain appropriate node targets;

[0009] Step S3: Use the ant colony algorithm to obtain the shortest path between nodes and achieve intelligent transmission line path planning;

[0010] Step S4: Visualize the algorithm results and provide an external usage interface.

[0011] Furthermore, Step S1 specifically includes the following steps:

[0012] Step S11: Use a drone equipped with a 3D lidar to obtain data parameters of the candidate line area, including real-time data such as multimedia photo data, distance image data, and aerial digital image data, to form a transmission line path space data set;

[0013] Step S12: Process the data set and divide the transmission line path space pictures into obstacle areas and topographic and geomorphic areas;

[0014] Step S13: Make training labels label for each image image in the transmission line path space data set image , the labels are divided into two categories, represented by the number 1 and the number 0; for the area in the image where line selection is appropriate, that is, the topographic and geomorphic area image pixel (appropriate), mark it as the number 1, and for the area where line selection is not appropriate, that is, the obstacle area image pixel (inappropriate) mark it as the number 0; the formula is as follows:

[0015]

[0016]

[0017] Step S14: Organize the labels to obtain the data set labels required for the training model.

[0018] Furthermore, Step S2 specifically includes the following steps:

[0019] Step S21: Input the image image in the preprocessed data set into the image segmentation neural network model. The model extracts image features and outputs the classification results of each pixel image pixel in the image. If the output is 1, the pixel is an appropriate line selection area, and if the output is 0, the pixel is an inappropriate line selection area;

[0020] Step S22: Use the cross-entropy loss function CrossEntropy_Loss to calculate the model loss value Loss between the model output result and the image label. The loss function formula is as follows:

[0021]

[0022] In the formula, number is the total number of image pixels, index is the pixel subscript, representing the index-th pixel, prob() represents the event probability, represents the category to which the index-th pixel pixel in the image image in the label belongs, represents the category to which the index-th pixel pixel in the image image output by the model belongs; then represents the probability that the pixel belongs to the label category;

[0023] Step S23: After calculating the loss value by the cross-entropy loss function, use the stochastic gradient descent optimization method to use the loss value to train the neural network model until the network model converges to achieve the highest accuracy;

[0024] Step S24: Input the image of the area where the transmission line route needs to be planned into the trained image segmentation neural network model to obtain the regional nodes suitable for route selection in this area.

[0025] Furthermore, in step S5: Step S3 specifically includes the following steps:

[0026] Step S31: Use the ant colony algorithm to select the regional nodes that need to be passed between two points; the optimal path in the ant colony algorithm is determined by the final pheromone concentration Potency(final_time), and final_time represents the termination time; using the ant colony algorithm first requires initializing the pheromone concentration Potency(0):

[0027] potency(0) = ant num / distance_rand

[0028] In the formula, ant num represents the number of ants, and distance_rand represents the path length between any two regional nodes;

[0029] Step S32: Select the next visited node nex_point for each ant. The probability formula for selecting the node is as follows:

[0030] prob(nex point ) = potency ij (time) / potency sum (time)

[0031] where potency ij (time) represents the pheromone concentration of the path between nodes i and j at time, and potency sum (time) represents the pheromone concentration of all paths at time;

[0032] Step S33: To avoid the ant colony algorithm falling into a local optimal solution, it is necessary to adjust the pheromone concentration left by a single ant at different times:

[0033]

[0034] where represents the pheromone left by the kth ant on the path between nodes i and j, distance_sum represents the total path length obtained by the kth ant after walking the entire path, the pheromone concentration adjustment parameter α belongs to hyperparameters and is set to 0.5; the concentration of pheromone left by a single ant decreases gradually over time;

[0035] Step S34: In the original algorithm, after each round of walking by the ants, the pheromone on all paths will evaporate; to avoid the ant colony algorithm falling into a local optimal solution, it is necessary to set the pheromone retention time potency_exist_time:

[0036]

[0037] Since the pheromone retention time decreases gradually with the number of iterations, to effectively avoid the ant colony algorithm falling into a local optimal solution; all ants release pheromone on the edges they passed through this round according to the path lengths they constructed, and the formula is as follows:

[0038]

[0039] where time represents time, lamda represents the evaporation rate of pheromone, which belongs to hyperparameters and is set to 0.5, kth represents the kth ant, represents the pheromone left by the kth ant on the path between nodes i and j, distance_sum represents the total path length obtained by the kth ant after walking the entire path;

[0040] Step S35: Iterate steps S32 and S33 multiple times until the algorithm model converges, and obtain the path with the highest pheromone concentration, which is the best path suitable for route selection between the nodes in this area.

[0041] Furthermore, in step S6: Step S4 specifically includes the following steps:

[0042] Step S41: Implement visual graphical display of the optimal transmission line path based on Python;

[0043] Step S42: Provide an external interface for processing data for easy use.

[0044] Compared with the related technology, the present invention and its preferred solutions have the following beneficial effects:

[0045] 1. The present invention applies the cutting-edge image segmentation technology to the transmission line path planning. According to the geographical and spatial environmental characteristics, the image segmentation neural network model is used to select the regional nodes suitable for route selection, which greatly saves manpower and improves the planning efficiency;

[0046] 2. The present invention applies the ant colony algorithm to the path planning. The optimal route is calculated based on the distances between the nodes, and a pheromone-related formula is constructed to avoid the ant colony algorithm falling into the local optimal solution, making the calculation results more accurate. It greatly saves manpower and improves the planning efficiency.

[0047] 3. The present invention graphically displays the planning results and provides an external interface for easy use. Description of the Drawings

[0048] The following further details the present invention in conjunction with the drawings and specific embodiments:

[0049] The drawings constituting a part of this application are used to provide a further understanding of the present invention. The illustrative embodiments and descriptions of the present invention are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0050] Figure 1 It is the overall scheme flow chart of the embodiment of the present invention.

[0051] The following specific embodiments will further illustrate the present invention in conjunction with the above drawings. Specific Embodiments

[0052] In order to make the purpose, technical solutions and advantages of this application clearer, the following describes and explains this application in conjunction with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application. Based on the embodiments provided in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope protected by this application.

[0053] Obviously, the accompanying drawings in the following description are only some examples or embodiments of the present application. For those of ordinary skill in the art, without creative efforts, the present application can also be applied to other similar scenarios based on these drawings. In addition, it can also be understood that although the efforts made in such a development process may be complex and lengthy, for those of ordinary skill in the art related to the content disclosed in the present application, some design, manufacturing, or production changes based on the technical content disclosed in the present application are only conventional technical means and should not be understood as the content disclosed in the present application being insufficient.

[0054] In the present application, the mention of "embodiment" means that the specific features, structures, or characteristics described in connection with the embodiment can be included in at least one embodiment of the present application. The appearance of this phrase at various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those of ordinary skill in the art explicitly and implicitly understand that the embodiments described in the present application can be combined with other embodiments without conflict.

[0055] Unless otherwise defined, the technical terms or scientific terms involved in the present application should have the ordinary meaning understood by those of ordinary skill in the technical field to which the present application belongs. The words such as "a", "an", "one kind", "the" and the like involved in the present application do not indicate a quantity limitation and can represent a singular or plural number. The terms "including", "comprising", "having" and any variations thereof involved in the present application are intended to cover non-exclusive inclusion; for example, a process, method, system, product or device including a series of steps or modules (units) is not limited to the listed steps or units, but may further include unlisted steps or units, or may further include other steps or units inherent to these processes, methods, products or devices. The terms "connected", "coupled" and the like involved in the present application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "multiple" involved in the present application refers to two or more. "And / or" describes the association relationship of associated objects and indicates that three relationships may exist. For example, "A and / or B" may represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after. The terms "first", "second", "third", etc. involved in the present application are only used to distinguish similar objects and do not represent a specific order for the objects.

[0056] Please refer to Figure 1 , the present invention provides an intelligent transmission line path planning method integrating pixel-level image segmentation and ant colony algorithm, including the following steps:

[0057] Step S1: Use a drone equipped with a 3D lidar to obtain the data parameters of the candidate line area, combine them into a corresponding transmission line path dataset, preprocess the transmission line path dataset, and label the tags of different areas to obtain a corresponding preprocessed dataset;

[0058] Step S2: Adopt an image segmentation task-based deep neural network model to perform road network node calculation processing on the preprocessed dataset to obtain appropriate node targets;

[0059] Step S3: Use the ant colony algorithm to obtain the shortest path between nodes and achieve intelligent transmission line path planning.

[0060] Step S4: Visualize the algorithm results and provide an external usage interface.

[0061] In this embodiment, step S1 specifically includes the following steps:

[0062] Step S11: Use a drone equipped with a 3D lidar to obtain the data parameters of the candidate line area, including multimedia data (photo data, jpg format), distance image data (real-time data), and aerial digital image data (aerial photos, jpg format), and organize them into a transmission line path space dataset.

[0063] Step S12: Process the dataset and divide the transmission line path space pictures into obstacle areas and topographic and geomorphic areas. Obstacle areas include but are not limited to residential areas, crowded areas, construction sites, etc. Topographic and geomorphic areas include but are not limited to plains, hills, mountains, lakes, etc.

[0064] Step S13: Make training labels label for each image image in the transmission line path space dataset image , and the labels are divided into two categories, represented by the number 1 and the number 0. For the areas in the image where line selection is appropriate, that is, the topographic and geomorphic areas image pixel (appropriate), mark it as the number 1, and for the areas where line selection is not appropriate, that is, the obstacle areas image pixel (inappropriate), mark it as the number 0. The formula is as follows:

[0065]

[0066]

[0067] Step S14: Organize the labels to obtain the dataset labels required for the training model.

[0068] In this embodiment, step S2 specifically includes the following steps:

[0069] Step S21: Input the image "image" in the preprocessed dataset into the image segmentation neural network model. The model extracts image features and outputs the classification results of each pixel "image" in the image "image". If the output is 1, the pixel is a suitable line selection area; if the output is 0, the pixel is an unsuitable line selection area. pixel The classification result of the pixel. If the output is 1, the pixel is a suitable line selection area; if the output is 0, the pixel is an unsuitable line selection area.

[0070] Step S22: Use the cross-entropy loss function CrossEntropy_Loss to calculate the model loss value Loss between the model output result and the image label. The formula of the loss function is as follows:

[0071]

[0072] In the formula, number is the total number of image pixels, index is the pixel subscript, representing the index-th pixel, prob() represents the event probability, represents the category to which the index-th pixel "pixel" in the image "image" belongs in the label, represents the category to which the index-th pixel "pixel" in the image "image" output by the model belongs. Then represents the probability that the pixel belongs to the label category.

[0073] Step S23: After calculating the loss value by the cross-entropy loss function, use the stochastic gradient descent optimization method to use the loss value to train the neural network model until the network model converges to achieve the highest accuracy.

[0074] Step S24: Input the image of the area where the transmission line route needs to be planned into the trained image segmentation neural network model to obtain the area nodes suitable for line selection in this area.

[0075] In this embodiment, step S3 specifically includes the following steps:

[0076] Step S31: Use the ant colony algorithm to select the area nodes that need to be passed between two points. The optimal path in the ant colony algorithm is determined by the final pheromone concentration Potency(final_time), and final_time represents the termination time. First, it is necessary to initialize the pheromone concentration Potency(0) when using the ant colony algorithm:

[0077] potency(0) = ant num / distance_rand

[0078] In the formula, ant num represents the number of ants, and distance_rand represents the path length between any two area nodes.

[0079] Step S32: Select the next visited node nex_point for each ant. The probability formula for node selection is as follows:

[0080] prob(nex point )=potency ij (time) / potency sum (time)

[0081] In the formula, potency ij (time) represents the pheromone concentration of the path between nodes i and j at time time, and potency sum (time) represents the pheromone concentration of all paths at time time.

[0082] Step S33: To prevent the ant colony algorithm from falling into a local optimal solution, it is necessary to adjust the pheromone concentration left by a single ant at different times:

[0083]

[0084] In the formula represents the pheromone left by the kth ant on the path between nodes i and j, distance_sum represents the total path length obtained after the kth ant walks through the entire path, and the pheromone concentration adjustment parameter α belongs to the hyperparameter and is set to 0.5. It can be seen from the formula that the concentration of pheromone left by a single ant gradually decreases over time.

[0085] Step S34: In the original algorithm, after each round that the ants walk through, the pheromone on all paths will evaporate. To prevent the ant colony algorithm from falling into a local optimal solution, it is necessary to set the pheromone retention time potency_exist_time:

[0086]

[0087] It can be seen from the above formula that the pheromone retention time gradually decreases with the number of iterations, effectively preventing the ant colony algorithm from falling into a local optimal solution. All ants release pheromone on the edges they passed through this round according to the path lengths they constructed. The formula is as follows:

[0088]

[0089] In the formula, time represents time, lamda represents the pheromone evaporation rate, which belongs to the hyperparameter and is set to 0.5, kth represents the kth ant, represents the pheromone left by the kth ant on the path between nodes i and j, and distance_sum represents the total path length obtained after the kth ant walks through the entire path.

[0090] Step S35: Iterate steps S32 and S33 multiple times until the algorithm model converges, and obtain the path with the highest pheromone concentration, which is the optimal path for route selection between nodes in this area.

[0091] In this embodiment, step S4 specifically includes the following steps:

[0092] Step S41: Implement visual graphical display of the optimal transmission line path based on Python.

[0093] Step S42: Provide an external interface for processing data for easy use.

[0094] Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0095] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0096] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0097] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide means for implementing the functions specified in Figure 1One process or multiple processes and / or boxes Figure 1 Steps of functions specified in one box or multiple boxes.

[0098] The technical features of the above-described embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above 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 described in this specification.

[0099] The above-described embodiments only represent several implementation manners of the present application. 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 noted that for those of ordinary skill in the art, without departing from the concept of the present application, several deformations and improvements can be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims. This patent is not limited to the above best implementation manner. Anyone can obtain other various forms of an intelligent transmission line path planning method that combines pixel-level image segmentation and ant colony algorithm under the inspiration of this patent. All equal changes and modifications made according to the scope of the patent application of the present invention shall fall within the scope covered by this patent.

Claims

1. An intelligent transmission line path planning method integrating pixel-level image segmentation and ant colony algorithm, characterized in that, it includes the following steps: Step S1: Use a drone equipped with a 3D lidar to obtain data parameters of the candidate line area, combine them into a corresponding transmission line path data set, preprocess the transmission line path data set, and label the tags of different areas to obtain a corresponding preprocessed data set; Step S2: Adopt an image segmentation task-based deep neural network model to perform road network node calculation processing on the preprocessed data set obtained in Step S1 to obtain suitable node targets; Step S3: Adopt the ant colony algorithm to obtain the shortest path between nodes to achieve intelligent transmission line path planning; Step S3 specifically includes the following steps: Step S31: Use the ant colony algorithm to select the area nodes that need to be passed between two points; the optimal path in the ant colony algorithm is determined by the final pheromone concentration Potency(final_time), and final_time represents the termination time; when using the ant colony algorithm, it is first necessary to initialize the pheromone concentration Potency(0): potency(0) = ant num / distance_rand where ant num represents the number of ants, and distance_rand represents the path length between any two regional nodes; Step S32: Select the next access node nex_point for each ant, and the probability formula for selecting the node is as follows: prob(nex point ) = potency ij (time) / potency sum (time) where potency ij (time) represents the pheromone concentration of the path between nodes i and j at time, potency sum (time) represents the pheromone concentration of all paths at time; Step S33: In order to avoid the ant colony algorithm falling into a local optimal solution, it is necessary to adjust the pheromone concentration left by a single ant at different times: where represents the pheromone left on the path between nodes i and j by the k-th ant, distance_sum represents the total path length obtained after the k-th ant walks through the entire path, the pheromone concentration adjustment parameter α belongs to the hyperparameter and is set to 0.5; the concentration of pheromone left by a single ant gradually decreases over time; Step S34: In the original algorithm, after each round that the ant walks, the pheromone on all paths will evaporate; in order to avoid the ant colony algorithm falling into a local optimal solution, it is necessary to set the pheromone retention time potency_exist_time: Since the pheromone retention time gradually decreases with the number of iterations, in order to effectively avoid the ant colony algorithm falling into a local optimal solution; all ants release pheromone on the edges they passed through this round according to the path lengths they constructed, and the formula is as follows: where time represents time, lada represents the evaporation rate of pheromone, which is a hyperparameter and is set to 0.5, kth represents the kth ant, represents the pheromone left by the kth ant on the path between nodes i and j, and distance_sum represents the total path length obtained after the kth ant walks through the entire path; Step S35: Iterate Steps S32 and S33 multiple times until the algorithm model converges, and obtain the path with the highest pheromone concentration, which is the best path for suitable line selection between regional nodes; Step S4: Visualize the algorithm results and provide an external usage interface.

2. The intelligent transmission line path planning method integrating pixel-level image segmentation and ant colony algorithm according to claim 1, characterized in that, Step S1 specifically includes the following steps: Step S11: Use a drone equipped with a 3D lidar to obtain data parameters of the candidate line area, including real-time data such as multimedia photo data and distance image data, and aerial digital image data, to form a transmission line path space data set; Step S12: Process the data set, and divide the transmission line path space picture into obstacle areas and terrain and landform areas; Step S13: Generate training labels for each image in the transmission line path spatial dataset image , the labels are divided into two categories, represented by the numbers 1 and 0; for the areas in the image that are suitable for route selection, i.e., the terrain and landform areas pixel (appropriate), mark as the number 1, and for the areas that are not suitable for route selection, i.e., the obstacle areas pixel (inappropriate), mark as the number 0; the formula is as follows: Step S14: Sort out the tags to obtain the data set tags required for the training model.

3. The intelligent transmission line path planning method integrating pixel-level image segmentation and ant colony algorithm according to claim 2, characterized in that, Step S2 specifically includes the following steps: Step S21: Input the image image in the preprocessed dataset into the image segmentation neural network model. The model extracts image features and outputs the classification results of each pixel image in the image image. If the output is 1, the pixel is a suitable line selection area; if the output is 0, the pixel is an unsuitable line selection area. pixel If the output is 1, the pixel is a suitable line selection area; if the output is 0, the pixel is an unsuitable line selection area. Step S22: Use the cross-entropy loss function CrossEntropy_Loss to calculate the model loss value Loss between the model output result and the image label. The loss function formula is as follows: where number is the total number of image pixels, index is the pixel subscript, representing the index-th pixel, prob() represents the event probability, representing the category to which the index-th pixel pixel in the image image in the label belongs, representing the category to which the index-th pixel pixel in the image image output by the model belongs; then represents the probability that the pixel belongs to the label category; Step S23: After the loss value is calculated by the cross-entropy loss function, use the stochastic gradient descent optimization method to use the loss value to train the neural network model until the network model converges to achieve the highest accuracy; Step S24: Input the image of the area where the transmission line route needs to be planned into the trained image segmentation neural network model to obtain the regional nodes suitable for route selection in this area.

4. The intelligent transmission line path planning method integrating pixel-level image segmentation and ant colony algorithm according to claim 1, characterized in that, Step S4 specifically includes the following steps: Step S41: Implement visual graphical display of the best path of the transmission line based on Python; Step S42: Provide an external interface for processing data for easy use.

Images