CN110046213B

CN110046213B - Power line selection method considering path distortion correction and cross crossing correction

Info

Publication number: CN110046213B
Application number: CN201910306291.XA
Authority: CN
Inventors: 周军义; 潘良军; 李宝昕; 王芝麟; 王喆; 郭瑾程; 陈本阳; 邹彬; 姚金雄; 张涵; 张超; 姜山; 王炜; 严欢; 闫娜; 万明忠; 李凤亮; 严研; 乔新辉; 陈思远
Original assignee: Beijing North Star Digital Remote Sensing Technology Co ltd; State Grid Shaanxi Electric Power Co Ltd; Economic and Technological Research Institute of State Grid Shaanxi Electric Power Co Ltd
Current assignee: Beijing North Star Digital Remote Sensing Technology Co ltd; State Grid Shaanxi Electric Power Co Ltd; Economic and Technological Research Institute of State Grid Shaanxi Electric Power Co Ltd
Priority date: 2018-11-20
Filing date: 2019-04-16
Publication date: 2023-04-18
Anticipated expiration: 2039-04-16
Also published as: CN110046213A

Abstract

The invention relates to a power line selection method considering path distortion correction and cross-over correction. The method comprises the following steps: firstly, storing data of each influence factor in a line selection range in a database; secondly, quantifying and weighting each influence factor; constructing the cost surface model again, and planning a path based on the model; then, carrying out path distortion correction on the generated path; and finally, carrying out cross-over correction on the path after the distortion correction to obtain a final path. Compared with the existing electric power line selection technology, the line selection method fully considers various line selection factors and is suitable for solving the electric power line selection problem; the distortion correction method fully considers the spatial position relation between the path and surrounding ground objects, carries out optimization processing such as corner point rejection, corner degree adjustment and the like on the path, effectively solves the problem of path distortion deformation of the machine-selected line, and improves the reasonability of the line; the cross-over correction method optimizes the cross-over position of the line, optimizes the corner point position near the cross-over point, and adjusts the position of the local path according to the spatial relationship with the ground object; the invention optimizes the data structure adopted by the shortest path algorithm, greatly improves the operation efficiency of the algorithm, reduces the calculated amount of path search by adopting a mode of limiting the search range, and further improves the line selection efficiency.

Description

Power line selection method considering path distortion correction and cross crossing correction

Technical Field

The invention belongs to the technical field of power transmission line design of a power system, and particularly relates to a power line selection method considering path distortion correction and cross crossing correction.

Background

The path selection of the transmission line is a precondition for the construction of the power network, and whether the path selection is reasonable in design is related to the investment operation cost and the operation reliability of the line, which is a crucial step in the line design. At present, the wire selection work of most design houses is mainly completed by manpower, and the wire selection work comprises two parts of planning wire selection and engineering wire selection. In the planning route selection stage, route selection personnel firstly draw out a plurality of preliminary route schemes on a 1/50000 or 1/100000 topographic map, and then carry out technical and economic comparison on each scheme according to collected related data (urban and rural planning, ecological or natural protection areas, industrial and mining and water conservancy facilities, military facilities, existing power lines, traffic lines, communication lines and the like) to abandon obvious and unreasonable schemes. The engineering route selection is to carry out on-site primary survey and measurement on a recommended scheme of planning route selection, carry out detailed investigation on the conditions of geology, hydrology, obstacles and the like of a route passing through a zone, draw a plane section diagram at some key positions, evaluate the feasibility of the scheme according to the survey result and adjust the primary selection scheme.

With the development of RS and GIS technologies, a small number of route selection works begin to utilize high-resolution satellite images or three-dimensional GIS environments or stereoscopic image mapping technologies to perform planning route selection, and LiDAR technologies are utilized to acquire terrain and earth surface features to perform engineering route selection. By means of high-resolution images or three-dimensional environments, more visual terrain expression, terrain analysis and the like can be carried out, but the problems existing in manual planning and line selection cannot be overcome: (1) The span of the general transmission line is as long as hundreds of kilometers, the range of the available line selection is tens of thousands of square kilometers, and the blind selection of one transmission line in such a large range by manpower is time-consuming and labor-consuming; (2) Factors such as line length, terrain, landform, geology, ice area, traffic, construction difficulty, operation and maintenance convenience degree, local planning and the like are comprehensively considered during path selection, line selection personnel need to be familiar with various factors in a working area and have enough line selection experience and responsibility, but along with replacement of the working personnel, many new line selection personnel often do not have the line selection experience; (3) With the development of social economy, the area for power wiring becomes more and more scarce, the wiring selection difficulty becomes more and more large, and a novel wiring selection method is urgently needed.

The power line selection belongs to the field of path planning, and therefore calculation can be performed by means of a shortest path planning algorithm. The a-algorithm is a commonly used path search algorithm. The method is mainly characterized in that a heuristic search function is adopted, so heuristic information related to the problem can be added in the search to guide the search to be carried out towards the most promising direction, and the solving process of the problem is accelerated and the optimal solution is found. The heuristic search function, also known as a valuation function, calculates the estimated cost of moving from the current node to the target node.

To understand the workflow of the a-algorithm, several concepts need to be understood first. G value: the actual cost of moving from the starting point to the current node. H value: estimated cost from the current node to the end point. F value: sum of G value and H value. Open list (open): and searching a node list to be retrieved in the path searching process. Close list (close): a list of nodes that have already been retrieved (nodes within the list will not be retrieved again).

The process of searching the shortest path by the algorithm A is as follows:

(1) The starting point is added to the open list.

(2) The following work was repeated:

(2.1) find the node with the lowest F value in the open list, take it as the current node, and then move it to the closed list.

(2.2) performing the following operations on each adjacent node of the current node:

if it is not available or is already in the closed list, it is skipped, otherwise the following decision is made.

If it is not in the open list, it is added. The current node is taken as its parent. The F, G, and H values for the node are recorded.

If it is already in the open list, check if the new path is better with the value of G as a reference. A lower G value means a better path. If the G value of the path to the node via the current node is lower, the parent node of the node is changed to the current node, and its G and F values are recalculated. And then re-order the open list by F.

And (2.3) stopping. When the target node is added to the closed list, a path is found; or the target node is not found and the open list is empty, indicating that the path does not exist.

(3) The path is saved. Starting at the target node, the requested path is obtained by moving along the parent node of each node until returning to the starting point.

Disclosure of Invention

In order to solve the problems, the invention provides a novel power transmission line path planning method which manages various spatial data in the line selection process based on a GIS environment, utilizes a computer to automatically select a power transmission line, realizes the automatic line selection of the power based on a cost surface model, realizes the distortion correction and cross crossing correction of a path, effectively solves the problems of path distortion deformation and cross crossing angle overrun of the line, and improves the feasibility, rationality and reliability of the mechanically selected line.

The line selection method provided by the invention comprises the following steps:

(1) And (6) data is put in a warehouse. The spatial data in the line selection range is collected and is guided into a GIS spatial database for storage, processing and management.

(2) And (5) quantifying the influence factors. Firstly, excluding forbidden passing areas which do not need quantization; secondly, defining a uniform quantization interval; grading the influence factors again; and finally, allocating quantization values for the factors.

(3) A weight is assigned. The weighting steps are that firstly, a hierarchical structure model is established; secondly, constructing a judgment matrix in each layer by adopting a pair-wise comparison mode; carrying out hierarchical single sequencing and consistency check again; and finally, carrying out total hierarchical ordering and consistency check.

(4) A cost surface model is constructed. The basic steps for constructing the cost surface data model are: firstly, establishing a regular grid; secondly, defining a neighborhood mode, wherein the neighborhood mode adopts an 8-neighborhood mode; assigning quantization values to the cells by utilizing the spatial superposition analysis technology again; then calculating a cell cost value, wherein the cost value is the sum of products of the standardized values of the various influence factors and corresponding weights of the various influence factors; finally, the connection cost is calculated, which is the cost of moving from one cell to a cell in its neighborhood.

(5) And (6) planning a path. The basic flow of path planning is as follows: and obtaining the power transmission line path by adopting an A-star algorithm based on the cost surface model according to the specified starting point and the specified destination point. Wherein, this patent has carried out optimization in two aspects to the A star algorithm: the data structures of the open table and the close table are optimized firstly, and an advanced binary heap structure is adopted to store data, so that the time consumed by list operation is greatly reduced, and the operation efficiency of the algorithm is improved. Where a binary heap is a binary tree that satisfies the heap ordering attribute. Then, the patent also adopts a mode of limiting the search range to optimize the A-star algorithm, thereby reducing unnecessary calculation and further improving the algorithm efficiency.

(6) And correcting path distortion. The path distortion correction is to remove unnecessary saw-tooth points under the condition of ensuring that the smoothed path does not pass through the limitation of the specified avoidance area on the premise of keeping the trend of the selected path, thereby achieving the purpose of path smoothing. The function is to keep the basic shape of the line under the condition of considering the spatial position relation between the line and surrounding ground objects, eliminate redundant corner points and reduce the angles of the reserved corner points as much as possible, so that the optimized line does not pass through a passing-forbidden area or a high-cost area.

(7) And (4) cross crossing correction. And (3) cross crossing correction, namely, considering the cross angle between the path and a linear ground object (such as a traffic line, a power line, a communication line and the like), optimizing the cross crossing position of the line, optimizing the position of a corner point near a cross crossing point, and adjusting the position of a local path according to the spatial relationship with the ground object.

A power line selection method considering path distortion correction and cross crossing correction comprises the following specific steps:

step 1: collecting spatial data in a line selection range, and importing the spatial data into a GIS spatial database for storage, processing and management; wherein the spatial data comprises: remote sensing data, topographic data, geological data, land utilization data, hydrometeorological data, ice region dirty region data and thunder damage risk region data;

step 2: quantifying impact factors, wherein the impact factors comprise: residential areas, planning areas, administrative districts, industrial areas, historical cultural trails, scenic areas, airports, military areas, natural or wild animal protection areas, water bodies, flood beaches, forest lands, cultivated lands, grasslands, deserts and bare earth surfaces, traffic access degrees, traffic lines, power lines, communication lines, slopes, geology, ice areas, dirty areas, distances and difficult line construction areas; firstly, defining a uniform quantization interval; grading the influence factors again; finally, distributing quantization values for all factors;

and 3, distributing weight to the influence factors, wherein the method for distributing the weight comprises the following steps: establishing a hierarchical structure model; adopting a pair-wise comparison mode to construct a judgment matrix in each layer, and comparing scales; sorting the hierarchical lists and checking consistency; checking the total hierarchical ordering and consistency;

step 4, constructing a cost surface model; the basic steps for constructing the cost surface data model are as follows: a. creating a regular grid; b. defining a neighborhood mode; c. assigning quantization values to the cells; d. calculating a cell cost value; e. calculating the connection cost;

step 5, planning a path; the optimized A-x algorithm is used as a path searching algorithm in a path planning stage, and the specific path planning step is as follows: a. setting a limited search range: l is less than or equal to phi l _min Wherein l = l ₁ +l ₂ ，l ₁ For the actual distance from the starting point to the current node, l ₂ For the estimated distance from the current node to the end point, [ phi ] is l and l _min The coefficient of proportionality of (1) generally ranges from (1,1.5)](ii) a b. Using an optimized A-star algorithm and taking a search range limiting formula as a limiting condition; c. outputting a result;

and 6: correcting path distortion; the path distortion correction process comprises the following steps: (1) inputting data: a starting point is a Start Line, an End point is an End Line, and passing Areas and some designated high-cost Areas are forbidden;

(2) Acquiring a coordinate set Points of Points capable of keeping the Line basic shape;

(3) Taking S = Start;

(4) Putting the S into a final path list FinALList, and finding out a first corner point C behind the S;

(5) If C = End, putting C into FinALList, and finishing optimization, otherwise finding out a first corner point D behind C;

(6) If Points contain C, S = C, and the step (4) is returned; if the Points do not contain C, connecting SD, and executing the step (7);

(7) Performing spatial calculation on the SD;

(7.1) if SD does not cross Areas, C = D, returning to step (5);

(7.2) if SD crosses Areas, making a perpendicular line of the SD from C, and setting the plumb foot as H; finding a point T on AH, wherein the point T needs to satisfy the nearest drop foot H, and ST and DT cannot cross Areas; adding T into a Candidate point set Candidate; calculating the intersection points of the ST and DT and the external rectangles of the Areas, and adding the intersection points into the Candidate; taking a point F from Candidate, wherein the point F needs to satisfy the condition that SF and DF do not cross Areas and the rotation angle between the SF and the DF is minimum; then taking S = F, and going to the step (4);

and 7: cross crossing correction; the cross-over correction process comprises the following steps: (1) acquiring any section of line on a path;

(2) Acquiring a linear ground object intersected with the section of the line;

(3) If more than one linear ground object is intersected with the section of the line, the midpoint of two adjacent intersections on the line is taken, and the line is divided into two or more sections; and repeating (1) to (3) until all the lines are traversed. To this end, the line segment between every two nodes of the path spans at most one linear feature.

(4) For each line SE, the intersection point of the line SE and the linear ground object is M, the intersection angle delta of the line SE and the linear ground object is calculated, and if the intersection angle delta > = the limit value theta, the next line is calculated;

(5) If delta is less than theta degrees, finding two end points S and E of the linear ground object, and solving two closest points on the linear ground object crossed with the two end points A and C respectively and an extension line SA 'of SA and an extension line EC' of EC;

(6) For A ', the crossing angle alpha between SA' and the linear ground object is obtained, if alpha < theta, the potential crossing point is not found in the MA section of the linear ground object (the MA section is simply abandoned), and the process goes to (8);

(7) If alpha > = theta, searching a point T with an included angle of theta degrees on the MA, and solving a corner beta at the position of T';

(7.1) if beta > the limit value gamma, abandoning the MA section and moving to (8);

(7.2) if beta < = gamma, judging whether ST ', T' E presses the ground object, if yes, dividing AT into N equal parts, traversing the N points from T to A direction in sequence, searching points with beta < = gamma degrees and without pressing the ground object until the end of beta = gamma, if yes, determining the feasible points, if not, abandoning the MA section, and turning to (8); if the ground object is not pressed, T' is a feasible point;

(8) Calculating the MC section, and repeating the steps (6) to (7);

(9) If the feasible points exist, adding the feasible points into the path;

(10) If no feasible point exists, comparing the sizes of SM and ME, finding a point O in a longer part from a cross point M, making MO equal to the length of the smaller part or making the length of MO greater than a certain distance, and repeating (4) to (8);

(11) If the feasible point exists, adding the feasible point into the path, and then, moving the O to the E direction for a certain distance to reduce the corner at the feasible point and the corner at the O point; if there is still no feasible point, no adjustment is made.

The optimized A-star algorithm flow comprises the following steps: initializing a starting node and a target node; (2) adding the initial node into an open table; (3) As long as the close table has no target nodes and the open table is not empty, the following steps are repeated: (4.1) taking the node with the lowest F value in the open table as the current node, and removing the current node from the open table and putting the current node into a close table; (4.2) judging whether each node in the neighborhood of the current node meets the limiting condition, and if not, marking the node as not-passable; (4.3) for each node in the neighborhood of the current node which meets the constraint and is not in the close table, if the node is not in the open table, moving the node into the table, marking the current node as a parent node of the node, recording a F, G value of the node, if the new G value is lower than the existing G value in the open table, marking the current node as the parent node of the node, and recording a F, G value of the node; the open table is used for storing all nodes which are known but not tested, the close table is used for storing the nodes which are tested, the value F is an estimated value from an initial node to a target node through the known nodes, the value G is an actual value from the initial node to the known nodes, and the value H is an optimal path estimated value from the known nodes to the target node.

Compared with the prior power transmission line path planning technology, the invention has the beneficial effects that: (1) The path planning of the power transmission line is essentially a path analysis problem of a continuous space, and the cost surface model can simulate the continuous space; (2) Because the moving direction of the cell is limited in the range set by the neighborhood, the calculated line has saw-toothed shape and serious distortion, and the reasonability of the path planning result is difficult to ensure. The technical scheme of the invention designs a method for correcting path distortion. The method fully considers the spatial position relation between the path and surrounding ground objects, carries out optimization processing such as corner point elimination and corner degree adjustment on the path, effectively solves the problem of path distortion and deformation of the line, and improves the reasonability of the line. This approach is not available with other path planning methods. (3) The calculation force limited by a computer is limited, and the crossing angle of a path and a linear ground object (such as a traffic line, a power line, a communication line and the like) cannot be considered in the path planning process, so that the problem that the included angle at the crossing part is over-limited is caused. The technical scheme of the invention designs a cross-over correction method. The method mainly optimizes the crossing and crossing positions of the lines, optimizes the positions of corner points near the crossing and crossing points, and adjusts the positions of local paths according to the spatial relationship with the ground objects. This approach is not available with other path planning methods. (4) This patent has carried out two aspects's optimization to the shortest path algorithm A who adopts: the method comprises the steps of firstly optimizing data structures of an open table and a close table, and storing data by adopting an advanced binary heap structure, so that the time consumed by list operation is greatly reduced, and the operation efficiency of an algorithm is improved. Wherein the binary heap is a binary tree that satisfies the heap ordering attribute. Then, the patent also adopts a mode of limiting the search range to optimize the A-star algorithm, thereby reducing unnecessary calculation and further improving the algorithm efficiency.

Drawings

FIG. 1 is a flow diagram of a power line selection method that accounts for path distortion correction and cross-over correction;

FIG. 2 is a diagram of a quantized value interval;

FIG. 3 is a schematic view of a hierarchical model;

FIG. 4 is a schematic diagram of a decision matrix;

FIG. 5 is a schematic of a surface view;

FIG. 6 is a schematic diagram of a neighborhood pattern;

FIG. 7 is a schematic diagram of cell assignments;

figure 8 is a schematic diagram of calculating cell cost values;

FIG. 9 is a schematic diagram of computing connection costs;

FIG. 10 is a schematic diagram of a cost surface based power line selection method;

FIG. 11 is a memory structure of a binary heap;

FIG. 12 is a schematic diagram of a limited search range;

FIG. 13a is a schematic diagram of an optimal path in path distortion;

FIG. 13b is a schematic diagram of a sawtooth path in the path distortion;

FIG. 13c is a schematic diagram of a sawtooth path in the path distortion;

FIG. 14 is a flow chart of a path distortion correction algorithm;

fig. 15 is a schematic diagram of cross-over adjustment front (left) and back (right).

Detailed Description

The following description of the embodiments of the present invention is provided for illustrative purposes, and other advantages and effects of the present invention will become apparent to those skilled in the art from the present disclosure.

Please refer to fig. 1 to 15. It should be understood that the structures, ratios, sizes, and the like shown in the drawings and described in the specification are only used for matching with the disclosure of the specification, so as to be understood and read by those skilled in the art, and are not used to limit the conditions under which the present invention can be implemented, so that the present invention has no technical significance, and any structural modification, ratio relationship change, or size adjustment should still fall within the scope of the present invention without affecting the efficacy and the achievable purpose of the present invention. In addition, the terms "upper", "lower", "left", "right", "middle" and "one" used in the present specification are for clarity of description, and are not intended to limit the scope of the present invention, and the relative relationship between the terms and the terms is not to be construed as a scope of the present invention.

In order to clearly illustrate the technical features of the present invention, the following detailed description of the overall technical solution of the present invention is provided with the accompanying drawings. As shown in fig. 1, the present invention comprises the steps of:

step 1: and (6) warehousing the data. The GIS has strong functions of spatial data storage, management, operation and the like, and a spatial database of the GIS is adopted to store spatial data in a line selection range. Firstly, collecting remote sensing data, topographic data, geological data, land utilization data, hydrological meteorological data, ice region dirty region data, thunder damage risk region data and the like in a line selection range. Then, storing the data into a spatial database of the GIS through GIS software, wherein the specific storage format is as follows: the remote sensing image and the DEM are stored in a raster layer; the topographic map data comprises terrain distribution and terrain relief conditions, and the terrain and each type of terrain are stored in different vector layers; the geological data is mainly a bad geological zone distribution map and is stored in a vector map layer; dividing the land utilization data into different vector layers according to land types for storage; hydrological and meteorological data are stored in different vector layers; the ice area and dirt area data refer to an ice area distribution diagram and a dirt area distribution diagram and are stored in different vector layers; and the lightning risk area is stored in a vector layer.

Step 2: and (5) quantifying the influence factors. Influence factors to be considered in the electric power line selection include residential areas, planning areas (core planning areas and non-core planning areas), administrative areas (refined to towns), industrial areas (factories, mining areas, wind power plants and the like), historic sites, scenic areas, airports, military areas, natural or wild animal protection areas, water bodies (including rivers, lakes, wetlands, reservoirs and the like), river flood beaches, woodlands, cultivated lands, grasslands, deserts and bare earth surfaces, traffic accessibility, traffic lines, power lines, communication lines, slopes, geology, ice areas, dirty areas, difficult areas for distance and line construction, ecological red lines and the like proposed in recent years.

These impact factors are both quantitative and qualitative, with each factor having a different unit or metric, which results in the inability to perform subsequent cost value calculations and optimal path calculations. Therefore, this step quantifies the influencing factors so that they have a uniform metric scale.

The forbidden passing areas are firstly picked from the influence factors, and the forbidden paths pass through the forbidden passing areas and are not quantized any more. Secondly, the quantization value interval of other factors is defined as 1 to 9,1 to represent the most suitable frame line, and 9 represents the least suitable frame line, as shown in fig. 2. The factors of the residential area, the ice area, the dirty area, the gradient, the traffic convenience, the cross spanning and the like are graded again, and other factors are not graded, as shown in table 1. Finally, quantized values are assigned to the factors, where different levels correspond to different quantized values, as shown in Table 2.

TABLE 1 impact factor ranking

/>

TABLE 2 quantized values of the impact factors

And step 3: different factors have different effects on the transmission line path and need to be assigned different weights. Whether the weight is reasonable or not directly influences the accuracy of the line selection result. This patent sets the weights of the factors to values between [0,1 ]. The weight determination method adopts a classical analytic hierarchy process. The method comprises the steps of (1) establishing a hierarchical structure model shown in figure 3; (2) A judgment matrix in each layer is constructed by adopting a pair-wise comparison mode, as shown in fig. 4, and the comparison scale is shown in table 3; (3) carrying out hierarchical single sequencing and consistency inspection; and (4) carrying out hierarchical total sorting and consistency check.

TABLE 3 meanings on scale

/>

And 4, step 4: a cost surface model is constructed. The basic steps for constructing the cost surface data model are as follows:

(1) And creating a regular grid. The whole spatial area under study is covered by a regular grid, and then the continuous working area is divided according to a certain rule to form regular polygons, and each polygon is called a grid unit. The regions may be partitioned into various mesh levels, and cost surfaces of various cell sizes, 200m × 20m, 150m × 150m, etc., may be created. The size of the grid cells has a great influence on the result of line selection, the calculation speed of larger cells is high, a rough line channel can be selected, the calculation time is long and the line is more specific. Thus, cost surfaces of desired precision may be created according to particular needs, with cost surfaces of different precision as shown in FIG. 5.

(2) A neighborhood mode is defined. The neighborhood pattern defines the adjacency between cells on the cost surface. During path planning, each cell can only move to cells in its neighborhood. Path planning cannot be done on the cost surface without specifying a neighborhood pattern because the connectivity graph between cells cannot be constructed without a neighborhood pattern. The cost surface model has many neighborhood patterns, such as 4, 8, 16, etc., as shown in FIG. 6.

(3) The cells are assigned quantization values. The electric power line selection needs to digitize factors influencing line selection, and the step mainly extracts various qualitative and quantitative influence factors into cells of the cost surface model by utilizing a space superposition analysis technology according to the quantization values set in the step 3 through data processing and conversion. Firstly, calculating the gradient by using DEM data, and extracting the average gradient value in each grid by using the functions of superposition analysis and statistical analysis; thirdly, generating different types of vector image layers according to different types of influence factors, such as various data of towns, villages, rivers, roads, scenic spots, mining areas, land utilization and the like, and extracting the influence levels of the various influence factors in the grids by data fusion and various spatial analyses of various influence factor data including forbidden areas; finally, a corresponding quantization value is set for each attribute in each grid cell. The process of assigning quantization values to cells is shown in fig. 7.

(4) A cell cost value is calculated. Each cell in the cost surface model holds a cost value (i.e., grid reachability) through the cell, the cost value being the sum of products of normalized values of the various impact factors and their respective weights. The cell cost values of the cost surface model are calculated in a weighted linear combination, as shown in figure 8.

(5) The connection cost is calculated. Moving from one cell to a cell in its neighborhood requires a certain cost, which is called the neighborhood moving cost or the connection cost, such as the line length, the material cost, the construction difficulty, the removal cost, etc., and these costs can be estimated by calculating the distance and the direction. As shown in fig. 9.

And 5, planning the path. The power line selection process based on the cost surface is shown in fig. 10, and a shortest path algorithm is adopted to plan a power transmission line path based on the cost surface model according to the specified start point and stop point and the necessary passing point. Because the shortest path problem between a single pair of vertexes is solved by the power transmission line path planning, and the heuristic search algorithm has higher efficiency in solving the shortest path problem between the single pair of vertexes, the A-star algorithm is adopted as the path searching algorithm in the path planning stage.

In the algorithm, the open table and the close table are optimized, and an advanced binary heap structure is adopted to store data, so that the time consumed by list operation is greatly reduced, the operation efficiency of the algorithm is improved, and the efficiency of the algorithm is improved by nearly one hundred times after test optimization, which is also a characteristic of the patent. The binary heap is a binary tree that satisfies the heap ordering attribute, and its time complexity is lower than that of the conventional list, and its storage structure is shown in fig. 11. Besides, the patent also adopts a mode of limiting the search range to optimize the A-star algorithm. The specific steps of planning the path by the optimized A-algorithm are as follows:

(1) A limited search range is set. Considering the requirements of the transmission line design rule on the line length: generally not greater than the linear distance l between the start and stop points _min 1.1 times, the scheme is based on the idea of a branch pruning method, and the estimated total length l of the line and the linear distance l between the start point and the stop point are calculated _min Is taken into account in the a algorithm as a constraint, which is shown in equation 1. This limits the search area to an ellipse, as shown in FIG. 12, in which the sum of the distances from each point on the ellipse to the connecting line between the start and stop points is not greater than l _min Phi times. Where l consists of two parts, i.e. the actual distance l from the starting point to the current node ₁ And an estimated distance l from the current node to the end point ₂ Phi is l and l _min The coefficient of proportionality of (1) generally ranges from (1,1.5)]。

l≤φl _min Wherein l = l ₁ +l ₂ Formula 1

(2) And (3) adopting an A-algorithm, and taking the formula 1 as a limiting condition. The basic flow of the algorithm is as follows:

(2.1) initializing a starting node and a target node;

(2.2) adding the initial node into an open table;

(2.3) repeating the following steps as long as the close table has no target node and the open table is not empty:

(2.4.1) taking the node with the lowest F value in the open table as the current node, removing the current node from the open table and placing the current node into a close table;

(2.4.2) judging whether each node in the neighborhood of the current node meets the limiting condition, and if not, marking the node as not-passing (except for the target node);

(2.4.3) for each node in the neighborhood of the current node that meets the constraint and is not in the close table, if it is not in the open table, moving it into the table, marking the current node as the parent node of the node, recording the F, G value of the node, if in the open table, marking the current node as the parent node of the node if the new G value is lower than the existing value, and recording the F, G value of the node.

(3) And outputting the result.

In the process, an open table is used for storing all nodes which are known but not tested, a close table is used for storing the tested nodes, an F value is an estimated value from an initial node to a target node through the known nodes, a G value is an actual value from the initial node to the known nodes, and an H value is an optimal path estimated value from the known nodes to the target node.

And 6: and correcting path distortion. Limited by the neighborhood pattern of the cost surface model, the computer-selected path tends to have jaggies, as is the case for the optimal path in FIG. 13a, and for both jagged paths in FIGS. 13b and 13 c. The path distortion correction is to remove unnecessary jagged points and reduce the angle of the reserved corner point as much as possible under the condition that the path after the distortion correction is not limited by the specified avoidance area on the premise of reserving the trend of the selected path, thereby achieving the purpose of smoothing the path. This method is not available in other power line selection methods.

The specific flow of the algorithm is described as follows, as shown in fig. 14:

(1) Inputting data: line "starting at Start and ending at End, forbidden to pass through zones and some designated high cost zones Areas.

(2) And acquiring a coordinate set Points of the Points capable of keeping the Line basic shape.

(3) Take S = Start.

(4) And putting the S into a FinalList of final paths, and finding out a first corner point C behind the S.

(5) If C = End, then put C into FinALList, and the optimization ends. Otherwise, the first corner point D after C is found.

(6) If Points contain C, S = C, and the step (4) is returned; if the Points do not contain C, SD is connected and step (7) is executed.

(7) The SD is spatially calculated.

(7.1) if SD does not cross Areas, C = D, and the process returns to step (5).

(7.2) if SD crosses Areas, perpendicular to SD from C, and the foot is H. Find a point T on AH that should satisfy the closest drop H, and ST and DT cannot cross Areas. T is added to the Candidate set Candidate. The intersection points of ST and DT with the circumscribed rectangle of Areas are calculated and added to Candidate. Taking point F from Candidate, point F must satisfy that SF and DF do not cross Areas, and the turn angle between them is minimal. Then, take S = F and go to step (4).

And 7: and (4) cross crossing correction. The calculation power of the computer is limited, and the intersection angle between the path and linear ground objects (such as traffic lines, power lines, communication lines and the like) cannot be considered in the path planning process, so that the selected path has the problem of exceeding the intersection crossing angle, and needs to be optimized. The method mainly optimizes the crossing and crossing positions of the lines, optimizes the positions of corner points near the crossing and crossing points, and adjusts the positions of local paths according to the spatial relationship with the ground objects. This method is not available in other power line selection methods. The method has the following flow as shown in fig. 15:

(1) Acquiring any section of line;

(2) Acquiring a linear ground object intersected with the section of the line;

(7.1) if β > the limit value γ, abandoning the MA section, and going to (8);

(8) Calculating the MC section, and repeating the steps (6) to (7);

(9) If the feasible points exist, adding the feasible points into the path (if two feasible points exist, adding the feasible points into the path is needed, so that the feasible points can be manually selected);

(11) If there are feasible points, the feasible points are added into the path (if there are two feasible points, the feasible points are also added, so that the feasible points can be selected), and then, the O is moved to the E direction for a certain distance (the corner edge can not press the ground object) to reduce the corners at the feasible points and the corners at the O point. If there are still no feasible points, no adjustment is made.

The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Those skilled in the art can modify or change the above-described embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.

Claims

1. A method for selecting a power line in consideration of path distortion correction and cross-over correction, comprising the steps of:

step 2: quantifying an influence factor, wherein the influence factor comprises: residential areas, planning areas, administrative districts, industrial areas, historical cultural trails, scenic areas, airports, military areas, natural or wild animal protection areas, water bodies, flood beaches, forest lands, cultivated lands, grasslands, deserts and bare earth surfaces, traffic access degrees, traffic lines, power lines, communication lines, slopes, geology, ice areas, dirty areas, distances and difficult line construction areas; firstly, defining a uniform quantization interval 1-9; grading the influence factors again; finally, distributing quantization values for all factors, wherein different quantization values correspond to different grades;

and 3, distributing weight to the influence factors, wherein the method for distributing the weight comprises the following steps: establishing a hierarchical structure model; adopting a pair-wise comparison mode to construct a judgment matrix in each layer, and comparing scales; sequencing the hierarchical lists and checking the consistency; checking the total sequence and consistency of the layers;

step 4, constructing a cost surface model; the basic steps for constructing the cost surface data model are: a. creating a regular grid; b. defining a neighborhood mode; c. assigning quantization values to the cells; d. calculating a cell cost value; e. calculating the connection cost;

step 5, planning a path; the optimized A-algorithm is adopted as a path finding algorithm in a path planning stage, and the specific path planning step is as follows: a. setting a restricted searchThe formula of the search range is:

where l = l1+ l2, l1 is the actual distance from the starting point to the current node, l2 is the estimated distance from the current node to the end point, and->

Is the proportionality coefficient of l and lmin, and the coefficient value range is (1,1.5)](ii) a b. Using an optimized A-star algorithm and taking a search range limiting formula as a limiting condition; c. outputting a result;

and 6: correcting path distortion; the path distortion correction process comprises the following steps: (61) input data: a starting point is Start, a terminal point is End, and passing Areas and some designated high-cost Areas are forbidden;

(62) Acquiring a coordinate set Points of Points capable of keeping the Line basic shape;

(63) Taking S = Start;

(64) Putting the S into a final path list FinALList, and finding out a first corner point C behind the S;

(65) If C = End, putting C into FinALList, and finishing optimization; otherwise, finding out a first corner point D behind the C;

(66) If Points contains C, S = C, go back to step (64); if the Points do not contain C, connecting SD, and executing step (67);

(67) Performing spatial calculation on the SD;

(67.1) if SD does not cross Areas, C = D, return to step (65);

(67.2) if SD crosses Areas, making a perpendicular line of SD from C, and the vertical foot is H; finding a point T on AH, wherein the point T needs to be closest to the plumb leg H, and ST and DT cannot cross Areas; adding T into a Candidate point set Candidate; calculating the intersection points of the ST and DT and the external rectangles of the Areas, and adding the intersection points into the Candidate; taking a point F from Candidate, wherein the point F needs to satisfy the condition that SF and DF do not cross Areas and the rotation angle between the SF and the DF is minimum; then, taking S = F, and going to step (64);

and 7: cross-over correction; the cross-over correction process comprises the following steps:

(71) Acquiring any section of line;

(72) Acquiring a linear ground object intersected with the section of the line;

(73) If more than one linear ground object is intersected with the section of the line, the midpoint of two adjacent intersections on the line is taken, and the line is divided into two or more sections; repeating (71) - (73) until all lines are traversed; so far, the line segment between every two nodes of the path spans at most one linear ground object;

(74) For each line SE, the intersection point of the line SE and the linear ground object is M, the intersection angle delta of the line SE and the linear ground object is calculated, and if the intersection angle delta > = the limit value theta, the next line is calculated;

(75) If delta is less than theta degrees, finding two end points S and E of the linear ground object, and solving two closest points on the linear ground object crossed with the two end points A and C respectively and an extension line SA 'of SA and an extension line EC' of EC;

(76) For A ', the crossing angle alpha of SA' and the linear ground object is obtained, if alpha < theta, the possible crossing point is not found in the MA section of the linear ground object (the MA section is simply abandoned), and the process goes to (78);

(77) If alpha > = theta, searching a point T with an included angle of theta degrees on the MA, and solving a corner beta at the position of T';

(77.1) if β > the limit γ, abandoning the MA segment, and proceeding to (78);

(77.2) if beta < = gamma, judging whether ST ', T' E presses the ground object, if yes, dividing AT into N equal parts, traversing the N points from T to A direction in sequence, searching points which are beta < = gamma degrees and do not press the ground object until the end of beta = gamma, if yes, determining feasible points, if not, abandoning MA sections, and turning to (78); if the ground object is not pressed, T' is a feasible point;

(78) Calculating the MC section, and repeating the steps (76) to (77);

(79) If the feasible points exist, adding the feasible points into the path;

(710) If there is no feasible point, comparing the sizes of SM and ME, finding a point O in the longer part from the cross point M, making MO equal to the smaller part or MO with a length greater than a certain distance, and repeating (74) - (78);

(711) If the feasible point exists, adding the feasible point into the path, and then, moving the O to the E direction for a certain distance to reduce the corner at the feasible point and the corner at the O point; if there is still no feasible point, no adjustment is made.

2. The electric power line selection method according to claim 1, characterized in that: the optimized A-star algorithm flow comprises the following steps: initializing an initial node and a target node; (2) adding the initial node into an open table; (3) As long as the close table has no target nodes and the open table is not empty, the following steps are repeated: (4.1) taking the node with the lowest F value in the open table as the current node, and removing the current node from the open table and putting the current node into a close table; (4.2) judging whether each node in the neighborhood of the current node meets the limiting condition, and if not, marking that the node cannot pass; (4.3) for each node in the neighborhood of the current node that meets the constraint and is not in the close table, if it is not in the open table, moving it into the table, marking the current node as the parent node of the node, recording the F, G value of the node, if in the open table, marking the current node as the parent node of the node if the new G value is lower than the existing G value, and recording the F, G value of the node; the open table is used for storing all nodes which are known but not tested, the close table is used for storing the nodes which are tested, the value F is an estimated value from an initial node to a target node through the known nodes, the value G is an actual value from the initial node to the known nodes, and the value H is an optimal path estimated value from the known nodes to the target node.