A railway route selection method, medium and device based on the environmental suitability gravitational field

Abstract

The present invention relates to the technical field of railway route selection, and discloses a railway route selection method, medium and device based on the environmental suitability gravitational field. The method includes: abstracting the route selection environment into a series of three-dimensional voxels, and clustering to obtain areas suitable and unsuitable for line passage; abstracting the areas suitable for line passage into a gravitational field, and abstracting the areas unsuitable for line passage into a repulsive field; obtaining the vector sum of the gravitational and repulsive forces received by each voxel in the route selection environment, and integrating the environmental suitability resultant forces of all voxels to obtain the environmental suitability gravitational field of the route selection environment; using the environmental suitability gravitational field of the route selection environment to correct the three-dimensional random tree algorithm, so that it can converge efficiently and obtain an optimized route plan. The present invention introduces the artificial potential field theory, models through the environmental suitability gravitational field of the railway route selection environment, and then uses this environmental suitability gravitational field to guide the route search process of the three-dimensional random tree algorithm, so that it can converge efficiently and generate an optimized route plan.

Description

Technical Field

[0001] The present invention relates to the technical field of railway route selection, and particularly to a railway route selection method, medium and device based on the gravitational field of environmental suitability. Background Art

[0002] As the pioneer and overall core work of railway construction, railway route selection has a profound impact on the safety, economy, etc. in the subsequent railway construction and operation stages. With the increasingly complex railway construction environment, the railway network is gradually expanding towards complex and dangerous mountainous areas, and its construction faces steep terrain and widespread and dense obstacles. Artificial designers often need to spend a lot of time and energy to design a feasible line plan, and may even ignore valuable line directions. Existing intelligent route selection methods provide an effective solution to this problem. They mainly adopt a route determination strategy of "searching for the environment with a line", that is, first generating alternative line plans, then evaluating the feasibility of the plans, and then calculating the objective function of the feasible plans and establishing a progressive optimization mechanism to make it decline, and iteratively optimizing the line plan until convergence. This route selection strategy can be classified as a "trial and error method", and it can still generate optimized plans in route selection areas with relatively flat terrain and sparse obstacle distribution. However, in complex and dangerous mountainous areas, since the areas unsuitable for line passage occupy the main body of the search space, it is easy to cause a large amount of computing time and resources to be consumed in the blind exploration of areas unsuitable for line passage, thus greatly reducing the search efficiency and optimization quality of existing intelligent route selection methods.

[0003] During the manual route selection process, designers often adopt the design idea of "evaluating first and then determining the line", that is, first evaluating the environment of the route selection area based on experience and prior knowledge, screening out the areas suitable for line passage, and then focusing the route selection work on these high-quality corridors to improve the design efficiency. Therefore, there is an urgent need for a modeling theory of railway route selection environmental suitability and its intelligent navigation technology applied to line search to improve the search efficiency and optimization quality and transform "blind search" into "directed search". Summary of the Invention

[0004] The purpose of the present invention is to provide an efficient railway route selection method based on environmental suitability evaluation, and its specific technical solution is as follows:

[0005] A railway route selection method based on the gravitational field of environmental suitability includes the following steps:

[0006] S1: Abstract the route selection environment into a series of three-dimensional voxels, and cluster to obtain areas suitable and unsuitable for line passage;

[0007] S2: Abstract the areas suitable for line passage into a gravitational field, and the areas unsuitable for line passage into a repulsive field;

[0008] S3: Calculate the vector sum of the gravitational and repulsive forces exerted on each voxel in the route selection environment, and integrate the resultant environmental suitability forces of all voxels to obtain the gravitational field of the route selection environmental suitability;

[0009] S4: Use the gravitational field of the route selection environmental suitability to correct the three-dimensional random tree algorithm so that it can converge and obtain an optimized route plan.

[0010] Preferably, the S1 is specifically:

[0011] Abstract the route selection area into a series of three-dimensional voxels, and divide the voxels into different structure layers according to the difference between the central elevation of the voxel and the elevation of the corresponding ground line, and the hierarchical threshold of the railway line structure;

[0012] Solve the values of multiple route passage suitability evaluation indicators including cost, structure stability, geological disaster risk, and carbon emission contained in the voxel according to the structure layer where the voxel is located. The quantization calculation formulas of the route passage suitability evaluation indicators in different structure layers are different. Use the linear weighting method to fuse the route passage suitability evaluation indicators to obtain the route passage environmental suitability value ES of all voxels in the route selection area. The larger the ES, the more suitable the route is to pass through this voxel;

[0013] According to the ES value of the voxel, use the k-mean algorithm to divide the voxels in the entire route selection area into the area RF suitable for route passage and the area RU not suitable for route passage.

[0014] Preferably, in the S2, calculate the gravitational force exerted on the route by the area suitable for route passage and abstract it into a gravitational field. The calculation formula is as follows:

[0015]

[0016] Among them, F att (i) represents the gravitational force of the i-th voxel exerted by the RF; F is the number of voxels in the RF; ES a represents the ES value of the a-th voxel located in the RF; ||p a -p i || represents the three-dimensional distance from voxel i to voxel a; U ia is the unit vector pointing from voxel i to voxel a; d ie is the distance from voxel i to the end point of the route; d ae is the distance from voxel a to the end point of the route;

[0017] The voxels in the RF closer to voxel i have a greater attraction to it, and the voxels in the RF farther from the end point than voxel i do not exert an attraction on it.

[0018] Preferably, in the S2, calculate the repulsive force exerted on the route by the area not suitable for route passage and abstract it into a repulsive field. The calculation formula is as follows:

[0019]

[0020] Among them, F rep (i) represents the repulsive force of the RU received by the i-th voxel; U is the number of voxels in the RU; ES b represents the ES value of the b-th voxel located in the RU; ||p i -p b || represents the three-dimensional distance from voxel i to voxel b; U bi is the unit vector pointing from voxel b to voxel i; d be is the distance from voxel b to the end point of the line;

[0021] The voxels in the RU closer to voxel i have a greater repulsive force on it, and the voxels in the RU farther from the end point than voxel i do not generate a repulsive force on it.

[0022] Preferably, the step S4 specifically includes:

[0023] Step 4.1: Add the starting point of the line to the nodes of the three-dimensional random tree and set the expansion step length L of the tree E ;

[0024] Step 4.2: Randomly generate a plane point PH in the line selection environment, and find the node PN closest to PH on the plane projection point of the three-dimensional random tree node; on the plane, extend a length of L from PN in the direction pointing to PH E to obtain the plane projection point PGH of the newly generated node PG; in the longitudinal plane, collect the elevation of PG at a specified spacing DS in the z-axis direction of PGH, and select the connection direction with the minimum cost between the three-dimensional points PN and PG as the original expansion direction D of the random tree ori ;

[0025] The cost F C (N, G) between two three-dimensional points is calculated by the following formula:

[0026]

[0027] Among them, ES p is the environmental suitability value of the voxel p through which the line connecting the three-dimensional points PN and PG passes; P is the number of voxels through which the line connecting PN and PG passes;

[0028] Step 4.3: Determine the final expansion direction of the random tree through the original expansion direction of point PN and the gravitational direction D fes of the gravitational field of the line selection environmental suitability;

[0029] Determine the final expansion step length according to the original expansion step length and the magnitude of the gravitational force of the environmental suitability;

[0030] Then the coordinates (X M , Y M , Z M ) of the newly generated node PM after correction are calculated as follows:

[0031]

[0032] Where X N , Y N , Z N are the three-dimensional coordinates of point PN; D orix , D oriy , D oriz are the projection vectors of the original direction unit vector on the x, y, and z axes respectively; D fex , D fey , D fez are the projection vectors of the normalized vector of the resultant force direction of environmental suitability on the x, y, and z axes respectively;

[0033] Step 4.4: Repeat Steps 4.1 to 3 until the random tree node closest to the end point of the line meets the requirement of the specified threshold D min and the cost fluctuation of the optimized line scheme is less than the set threshold ΔF C , and an optimized line scheme is obtained.

[0034] Applying the technical solution of the present invention has the following beneficial effects:

[0035] The railway route selection method based on the environmental suitability gravitational field provided by the present invention includes: S1: Abstracting the route selection environment into a series of three-dimensional voxels, and clustering to obtain areas suitable and unsuitable for line passage; S2: Abstracting the areas suitable for line passage into a gravitational field, and abstracting the areas unsuitable for line passage into a repulsive field; S3: Calculating the vector sum of the gravitational and repulsive forces received by each voxel in the route selection environment, and integrating the resultant environmental suitability force of all voxels to obtain the environmental suitability gravitational field of the route selection environment; S4: Using the environmental suitability gravitational field of the route selection environment to correct the three-dimensional random tree algorithm, so that it can converge efficiently and obtain an optimized line scheme. Aiming at the problem that the existing intelligent line search method is difficult to efficiently generate an optimized scheme in complex and dangerous mountainous areas, the present invention proposes a search strategy of "environment attracting line", introduces the artificial potential field theory, models the environmental suitability gravitational field of railway route selection, and then uses this environmental suitability gravitational field to guide the line search process of the three-dimensional random tree algorithm, so that it can converge efficiently and generate an optimized line scheme.

[0036] The present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the aforementioned railway route selection method based on the environmental suitability gravitational field is implemented.

[0037] The present invention also provides an electronic device, including: a processor; and a memory for storing executable instructions of the processor; wherein, the processor is configured to execute the foregoing railway route selection method based on the environmental suitability gravitational field by executing the executable instructions.

[0038] In addition to the purposes, features, and advantages described above, the present invention has other purposes, features, and advantages. The following will refer to the drawings for a further detailed description of the present invention. Description of the Drawings

[0039] The drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0040] Figure 1 is a flowchart of the method provided according to an exemplary embodiment of the present invention;

[0041] Figure 2 is a schematic diagram of areas suitable and unsuitable for line passage;

[0042] Figure 3 is a schematic diagram of the environmental suitability gravitational field. Detailed Embodiments

[0043] The following will describe the embodiments of the present invention in detail with reference to the drawings. However, the present invention can be implemented in many different ways defined and covered by the claims.

[0044] The following will further illustrate a railway route selection method based on the environmental suitability gravitational field of the present invention by taking a complex mountain railway section with dense obstacles as an example.

[0045] Reference Figure 1 , a railway route selection method based on the environmental suitability gravitational field, includes the following steps:

[0046] First, perform gravitational field modeling on the suitability distribution of the railway route selection environment, specifically including S1 to S3.

[0047] S1: Abstract the route selection environment into a series of three-dimensional voxels, and cluster to obtain areas suitable and unsuitable for line passage;

[0048] Specifically, abstract the route selection area into a cube voxel set with a side length of 90 m; according to the difference between the central elevation of the voxel and the elevation of the corresponding ground line, and the railway line structure layer threshold (preferably, the bridge-fill threshold is set to 20 m, and the cut-tunnel threshold is set to 30 m), divide the voxels into different structure layers;

[0049] Solve the values of multiple line passage suitability evaluation indicators including cost, structure stability, geological disaster risk, and carbon emissions for the voxels according to the structure layer where the voxels are located. Among them, the quantization calculation formulas of the line passage suitability evaluation indicators for different structure layers are different. Use the linear weighting method to fuse the line passage suitability evaluation indicators to obtain the line passage environment suitability value ES of all voxels in the route selection area. The larger the ES, the more suitable the line is to pass through this voxel.

[0050] According to the ES value of the voxels, use the k-mean algorithm to divide the voxels in the entire route selection area into the area RF suitable for line passage and the area RU not suitable for line passage.

[0051] S2: Abstract the area suitable for line passage as a gravitational field and the area not suitable for line passage as a repulsive field. Specifically:

[0052] Calculate the gravitational force generated by the area suitable for line passage on the line and abstract it as a gravitational field. The calculation formula is as follows:

[0053]

[0054] Among them, F att (i) represents the gravitational force of the i-th voxel from RF; F is the number of voxels in RF; ES a represents the ES value of the a-th voxel located in RF; ||p a -p i || represents the three-dimensional distance from voxel i to voxel a; U ia is the unit vector from voxel i to voxel a; d ie is the distance from voxel i to the end of the line; d ae is the distance from voxel a to the end of the line;

[0055] The voxels in RF closer to voxel i have a greater attraction to it, and the voxels in RF farther from the end than voxel i do not generate attraction to it.

[0056] Calculate the repulsive force generated by the area not suitable for line passage on the line and abstract it as a repulsive field. The calculation formula is as follows:

[0057]

[0058] Among them, F rep (i) represents the repulsive force of the i-th voxel from RU; U is the number of voxels in RU; ES b represents the ES value of the b-th voxel located in RU; ||p i -p b || represents the three-dimensional distance from voxel i to voxel b; U bi is the unit vector from voxel b to voxel i; dbe is the distance from voxel b to the end of the line;

[0059] The closer the voxel in the RU is to voxel i, the greater the repulsive force it exerts on voxel i. Voxels in the RU that are farther from the end point than voxel i do not exert a repulsive force on it.

[0060] S3: Calculate the vector sum of the gravitational and repulsive forces received by each voxel in the route selection environment, and integrate the resultant force of the environmental suitability of all voxels to obtain the gravitational field of the environmental suitability of the route selection environment;

[0061] Then, use the gravitational field of the environmental suitability of the route selection environment to guide the search process of the three-dimensional random tree as follows:

[0062] S4: Use the gravitational field of the environmental suitability of the route selection environment to correct the three-dimensional random tree algorithm so that it can converge efficiently and obtain an optimized route plan, specifically including:

[0063] Step 4.1: Add the starting point of the line to the nodes of the three-dimensional random tree and set the expansion step length L of the tree E , which is preferably set to 5000m in this embodiment;

[0064] Step 4.2: Randomly generate a plane point PH in the route selection environment, and find the node PN closest to PH on the plane projection point of the three-dimensional random tree node; on the plane, extend a length of L from PN in the direction pointing to PH E to obtain the plane projection point PGH of the newly generated node PG; in the longitudinal plane, collect the elevation of PG at a specified spacing DS in the z-axis direction of PGH, and select the connection direction with the lowest cost between the three-dimensional points PN and PG as the original expansion direction D of the random tree ori ; in this embodiment, the longitudinal sampling spacing of PGH is set to 1m.

[0065] The cost F C (N, G) between two three-dimensional points is calculated by the following formula:

[0066]

[0067] where ES p is the environmental suitability value of the voxel p passed through by the connection line between the three-dimensional points PN and PG; P is the number of voxels passed through by the connection line between PN and PG;

[0068] Step 4.3: Determine the final expansion direction of the random tree through the original expansion direction of point PN and the gravitational direction D of the gravitational field of the environmental suitability of the route selection environment fes ;

[0069] Determine the final expansion step length according to the original expansion step length and the magnitude of the environmental suitability gravity;

[0070] Then, the coordinates (X M, Y M , Z M ) The calculation is as follows:

[0071]

[0072] Where X N , Y N , Z N are the three-dimensional coordinates of point PN; D orix , D oriy , D oriz are the projection vectors of the original direction unit vector on the x, y, and z axes respectively; D fex , D fey , D fez are the projection vectors of the normalized vector of the resultant force of environmental suitability on the x, y, and z axes respectively;

[0073] At this time, one node expansion of the random number search is completed.

[0074] Step 4.4: Repeat steps 4.1 to 3 until the random tree node closest to the end point of the line satisfies the specified threshold D min and the cost fluctuation of the optimized line scheme is less than the set threshold ΔF C , and an optimized line scheme is obtained.

[0075] In this embodiment, it is preferably set that the requirement threshold D min for the end point of the line to be connectable is 500 m, and the cost fluctuation threshold ΔF C of the optimized line scheme is 0.01 times the cost of the first connectable line scheme.

[0076] This embodiment also discloses a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the railway route selection method based on the environmental suitability gravitational field described in this embodiment is implemented.

[0077] This embodiment also discloses an electronic device, including:

[0078] A processor;

[0079] And a memory for storing executable instructions of the processor;

[0080] Wherein, the processor is configured to execute the railway route selection method based on the environmental suitability gravitational field as described in this embodiment by executing the executable instructions.

[0081] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A railway line selection method based on environmental suitability gravity field, characterized in that: The steps include: S1: The line selection environment is abstracted into a series of three-dimensional voxels, and the areas suitable and unsuitable for line passage are obtained by clustering; S2: The area suitable for the route to pass is abstracted as the gravitational field, and the area not suitable for the route to pass is abstracted as the repulsive field; S3: Calculate the vector sum of the gravitational force and repulsive force on each voxel in the line selection environment, and integrate the environmental suitability of all voxels to obtain the gravitational field of the line selection environment suitability; S4: Use the gravitational field of the environmental suitability of the line selection to correct the three-dimensional random tree algorithm so that it can converge and obtain the optimal line plan; The S1 is specifically: The line selection area is abstracted into a series of three-dimensional voxels, and the voxels are divided into different structural layers according to the difference between the center elevation of the voxel and the corresponding ground line elevation and the railway line structure layering threshold; According to the structural layer where the voxel is located, the values ​​of multiple route suitability evaluation indicators of the voxel, including construction cost, structural stability, geological disaster risk and carbon emission, are solved. The quantitative calculation formulas of the route suitability evaluation indicators of different structural layers are different. The route suitability evaluation indicators are integrated using the linear weighted method to obtain the route travel environment suitability value ES of all voxels in the line selection area. The larger the ES, the more suitable the route is for passing through this voxel. According to the ES value of the voxel, the k-mean algorithm is used to divide the voxels in the entire line selection area into the area RF suitable for line passage and the area RU unsuitable for line passage; The step S4 specifically includes: Step 4.1: Add the starting point of the line to the node of the three-dimensional random tree and set the tree expansion step size L E ; Step 4.2: Generate a plane point PH randomly in the line selection environment, find the node PN closest to PH on the plane projection point of the three-dimensional random tree node; on the plane, extend an L from PN in the direction pointing to PH E Get the plane projection point PGH of the newly generated node PG; on the vertical plane, collect the elevation of PG at a specified interval DS in the z-axis direction of PGH, and select the connection direction with the smallest cost between the three-dimensional point PN and PG as the original expansion direction D of the random tree ori ; Cost between two 3D points F C (N,G) is calculated by the following formula: Among them ES p is the environmental suitability value of the voxel p through which the line connecting the three-dimensional points PN and PG passes; P is the number of voxels through which the line connecting PN and PG passes; Step 4.3: The original extension direction of the point PN and the gravitational direction D of the gravitational field of the environmental suitability of the selected line fes Determine the final expansion direction of the random tree; The final expansion step length is determined according to the original expansion step length and the gravity of environmental suitability; Then the coordinates of the newly generated node PM after correction (X M ,Y M ,Z M ) is calculated as follows: Where X N ,Y N ,Z N is the three-dimensional coordinate of point PN; D orix ,D oriy ,D oriz are the projection vectors of the original direction unit vector on the x, y, and z axes respectively; D fex ,D fey ,D fez are the projection vectors of the normalized vector of the resultant force direction of environmental suitability on the x, y, and z axes respectively; Step 4.4: Repeat steps 4.1 to 4.3 until the random tree node closest to the end of the line meets the specified threshold D min The requirement is that the cost fluctuation of the optimized line solution is less than the set threshold ΔF C , and get the optimized line plan.

2. A railway line selection method based on environmental suitability gravitational field according to claim 1, characterized in that: Calculate the gravitational force generated by the area suitable for line passage on the line and abstract it into a gravitational field. The calculation formula is as follows: Among them, F att (i) represents the gravitational force of RF on the i-th voxel; F is the number of voxels in RF; ES a represents the ES value of the ath voxel in RF; ||p a -p i || represents the three-dimensional distance from voxel i to voxel a; U ia is the unit vector pointing from voxel i to voxel a; d ie is the distance from voxel i to the end point of the line; d ae is the distance from voxel a to the end point of the line; The voxels in the RF that are closer to the voxel i have a greater attraction to it, and the voxels in the RF that are farther away from the end point than the voxel i have no attraction to it.

3. The method for railway line selection based on environmental suitability gravitational field according to claim 2, characterized in that: The repulsive force generated by the area that is not suitable for the route to pass on the route is calculated and abstracted into a repulsive field. The calculation formula is as follows: Among them, F rep (i) represents the repulsive force of the RU on the i-th voxel; U is the number of voxels in the RU; ES b represents the ES value of the bth voxel in RU; ||p i -p b || represents the three-dimensional distance from voxel i to voxel b; U bi is the unit vector pointing from voxel b to voxel i; d be is the distance from voxel b to the end point of the line; The voxels in the RU that are closer to the voxel i have a greater repulsive force on it, and the voxels in the RU that are farther away from the end point than the voxel i do not have any repulsive force on it.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the railway route selection method based on the environmental suitability gravitational field as described in any one of claims 1 to 3 is implemented.

5. An electronic device, characterized in that: include: processor; and a memory for storing executable instructions for the processor; Wherein, the processor is configured to execute the railway route selection method based on the environmental suitability gravitational field as described in any one of claims 1-3 by executing the executable instructions.

Images