CN113240163B - Multi-objective optimization method for railway lines in complicated mountain areas - Google Patents
Multi-objective optimization method for railway lines in complicated mountain areas Download PDFInfo
- Publication number
- CN113240163B CN113240163B CN202110471218.5A CN202110471218A CN113240163B CN 113240163 B CN113240163 B CN 113240163B CN 202110471218 A CN202110471218 A CN 202110471218A CN 113240163 B CN113240163 B CN 113240163B
- Authority
- CN
- China
- Prior art keywords
- scheme
- line
- inferior solution
- particle
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 68
- 238000000034 method Methods 0.000 title claims abstract description 44
- 239000002245 particle Substances 0.000 claims abstract description 70
- 230000007613 environmental effect Effects 0.000 claims abstract description 18
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 10
- 238000010606 normalization Methods 0.000 claims abstract description 6
- 238000012545 processing Methods 0.000 claims abstract description 6
- 238000004364 calculation method Methods 0.000 claims abstract description 3
- 239000002689 soil Substances 0.000 claims description 16
- 230000006378 damage Effects 0.000 claims description 13
- 230000008569 process Effects 0.000 claims description 13
- 238000011156 evaluation Methods 0.000 claims description 12
- 239000002699 waste material Substances 0.000 claims description 11
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 11
- 238000012163 sequencing technique Methods 0.000 claims description 10
- 238000011234 economic evaluation Methods 0.000 claims description 8
- 238000011158 quantitative evaluation Methods 0.000 claims description 7
- 238000009412 basement excavation Methods 0.000 claims description 6
- 239000002893 slag Substances 0.000 claims description 6
- 238000012821 model calculation Methods 0.000 claims description 4
- 238000013461 design Methods 0.000 abstract description 11
- 230000006870 function Effects 0.000 description 21
- 238000004162 soil erosion Methods 0.000 description 4
- 238000010276 construction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000011435 rock Substances 0.000 description 1
- 238000004800 variational method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
- G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- Strategic Management (AREA)
- Databases & Information Systems (AREA)
- Marketing (AREA)
- Artificial Intelligence (AREA)
- General Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Remote Sensing (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Tourism & Hospitality (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a multi-objective optimization method for a railway line in a complex mountain area, which comprises the steps of establishing a comprehensive geographic information model, then establishing a multi-objective optimization model for the railway line which cooperatively considers economic and environmental factors, then carrying out calculation on the multi-objective optimization model for the railway line in a continuous space based on a particle swarm algorithm, and finally obtaining a result through normalization processing. The intelligent optimizing search of the railway line can be carried out in a continuous space, economy and environmental protection are comprehensively considered, a high-quality line alternative scheme is recommended for a designer, line design is assisted, and the quality and efficiency of the line design are improved; the method has the advantages of high automation degree, strong practicability, high operation efficiency and high popularization and application value.
Description
Technical Field
The invention relates to the technical field of railway line design, in particular to a multi-objective optimization method for a railway line in a complex mountain area.
Background
The railway route selection design is the core work of the overall railway construction overview, is a prerequisite for determining investment, operation, safety and environmental protection of projects, is also a fundamental problem for controlling the factors, and has the basic task of determining the technical standard and the spatial position of a line according to the functional requirements of the design projects and combining the natural environment, the social environment and the economic environment of a region where the design projects are located, and uniformly distributing various structures in a multi-target coordination manner. Therefore, the ideal line design not only needs the geometric design of space positions (planes and vertical sections), but also needs to realize the optimal configuration of various structures on the line according to the environments along the line (terrains, ground objects, surface vegetation coverage, existing road networks, forbidden zones and the like) so as to mutually match the structures to achieve the overall cooperative optimization.
Since the 60 s of the last century, students have conducted long and intensive studies on the line optimization problem at home and abroad, and a series of methods including variational methods (Howard, 1968), gradient projection (Zhan Zhenyan et al, 1982), grid optimization (tritsch et al, 1987), dynamic planning (Fwa et al, 1989; li et al, 2013), genetic algorithms (Jong, 1998; kim et al, 2005; kang et al, 2009; lai,2012; davey et al, 2017), distance transformation algorithms (de Smith,2006; li et al, 2016; pu et al, 2019) and the like have been proposed, and most of these methods are single-objective optimization methods mainly based on economic evaluation. However, in complex mountain areas, railway route selection is often faced with a fragile ecological environment, and the environmental friendliness of the route is also a main factor affecting scheme decisions. The existing single-objective optimization method mainly based on economic evaluation is difficult to balance the economical efficiency and the environmental protection of the line in the process of optimizing the line, so that a comprehensive optimal line scheme is designed.
Disclosure of Invention
The invention aims to provide a multi-objective optimization method for a railway line in a complicated mountain area, which is capable of carrying out intelligent optimizing search on the railway line in a continuous space by cooperatively considering economic and environmental factors and improving the quality and efficiency of line design.
The technical scheme adopted by the invention is as follows:
the multi-objective optimization method for the railway lines in the complicated mountain areas is characterized by comprising the following steps of:
the method comprises the following steps:
S 1 : establishing a comprehensive geographic information model:
S 2 : constructing a railway line multi-objective optimization model which cooperatively considers economic and environmental factors:
S 3 : and carrying out multi-objective optimization model calculation on the railway line under continuous space based on a particle swarm algorithm:
S 4 : will pass through step S 3 Obtained byAnd carrying out normalization processing on each optimization target of the line schemes contained in the non-inferior solution scheme set.
Said step S 1 Comprising the following steps:
S 1-1 : establishing a railway route selection information storage structure based on a grid form;
S 1-2 : and collecting various information required by line searching, dispersing the information into a grid, and establishing a railway line selection comprehensive geographic information model.
Said step S 2 Comprising the following steps:
S 2-1 : analyzing key parameters affecting railway space line positions, and determining independent variables of a multi-objective optimization model of the railway line;
S 2-2 : establishing a functional relation between independent variables of a railway multi-objective optimization model and economic evaluation targets, and constructing a line economic evaluation target function;
S 2-3 : establishing a functional relation between independent variables of a railway multi-objective optimization model and environmental impact evaluation targets, and constructing a line environmental impact evaluation target function;
S 2-4 : and establishing constraint conditions of the railway line optimization model.
Said step S 3 Comprising the following steps:
S 3-1 : generating an initial particle population, wherein the initial position of each particle represents each initial line scheme;
S 3-2 : constructing a storage file of the non-inferior solution scheme of each particle, storing the non-inferior solution scheme generated in the iterative evolution process of each particle into the respective storage file, generating a non-inferior solution scheme set of each particle, and respectively selecting an individual optimal scheme of each particle from the non-inferior solution scheme set;
S 3-3 : constructing a non-inferior solution scheme storage file of the whole population, storing the non-inferior solution scheme generated in the iterative evolution process of the whole population into the file, generating a non-inferior solution scheme set of the whole population, and selecting a global optimal scheme from the non-inferior solution scheme set;
S 3-4 : guiding population evolution based on the individual optimal scheme and the global optimal scheme;
S 3-5 : laminationAnd (5) evolving the population until reaching the iteration termination condition.
Said step S 2-3 The constructed environment evaluation objective function comprises a vegetation destruction degree objective function and a soil erosion degree objective function:
vegetation damage level objective function: introducing a normalized vegetation index NDVI into a railway line optimization model, dispersing NDVI data into each grid of a comprehensive geographic information model, adding positive NDVI values of a road passing region in a roadbed form, and taking the accumulated NDVI values as quantitative evaluation indexes of vegetation damage degrees of the road scheme, wherein the quantitative evaluation indexes are shown in a formula (1):
wherein SV is NDVI An accumulated value of NDVI of the line passing area, C (k) The kth grid for the passage of the line in the form of a roadbed, U S For a grid set where the line passes in the form of a road bed,NDVI values for the kth grid through which the line passes in a roadbed form, max { } represents taking the maximum value;
water and soil loss degree objective function: taking the waste soil and waste slag volume as a representative value of the water and soil loss degree, and the waste soil and waste slag volume V S By the volume V of the roadbed excavation SC Volume V of tunnel excavation TC Volume V of roadbed filling SF Calculation, as shown in formula (2):
V S ＝max{V SC +V TC -V SF ,0} (2)。
said step S 3 The obtaining process of the individual optimal scheme specifically comprises the following steps:
firstly, establishing a storage file for each particle, storing all non-inferior solution schemes obtained by searching the current generation by each particle, and establishing a non-inferior solution scheme set of each particle; then, each optimization target of each non-inferior solution scheme is compared in sequence, the winning value of each non-inferior solution scheme is calculated by establishing a tournament equation (3), the sorting value of each scheme is calculated by establishing a sorting function formula (5), the non-inferior solution schemes are sorted, and the larger the sorting value of the scheme is, the higher the ranking is; finally, selecting the first scheme as an individual optimal scheme of each particle;
wherein phi is i For the non-bad solution set of the ith particle, T (j) (a,Φ i ) Optimizing the winning value of the objective, |Φ, for the j-th term of the a-scheme in the non-bad solution set of the i-th particle i I is the number of non-bad solutions of the ith particle, t (j) (a, b) optimizing an objective function for a constructed line for a winning value of a j-th optimization objective of a scheme and b scheme in a non-inferior solution scheme set of the i-th particle, the smaller the target value, the better the scheme, thus t (j) (a, b) is calculated from formula (4):
wherein f (j) (a) And f (j) (b) Optimizing target values for the j-th items of the a scheme and the b scheme respectively;
wherein R (a) is the sorting value of the scheme a, m is the optimized target quantity, w j The target weights are optimized for each item.
Said step S 3 The global optimal scheme is obtained by the following steps:
firstly, storing a non-inferior solution scheme obtained by searching the current generation by all particles, and establishing a global non-inferior solution scheme set; then, calculating the distance between each scheme in the global non-inferior solution set and other schemes in a multidimensional space through a formula (6), constructing a crowding degree function (7), calculating the crowding degree of each scheme, and sequencing each scheme in the global non-inferior solution set according to the crowding degree, wherein the lower the crowding degree is, the more front the sequencing is, so as to obtain a sequence R1; sequencing all schemes in the global non-inferior solution set through a tournament algorithm to obtain a sequence R2; finally, based on the roulette idea, constructing a probability decision equation (8) comprehensively considering R1 and R2, and selecting a scheme with the maximum probability value as a global optimal solution scheme;
wherein D (a, b) is the distance between the a scheme and the b scheme in the multi-dimensional solution space in the global non-inferior solution set, m is the number of optimization targets, f (j) (a) Optimizing target value for j-th item of a scheme in global non-inferior solution scheme set, f (j) (b) Optimizing the target value for the j-th term of the b-scheme in the global non-bad solution set,maximum value of the objective of optimization for item j in the global non-bad solution,/>Optimizing the minimum value of the target for the j-th item in the global non-inferior solution scheme;
wherein ρ is a For the crowding value of the a scheme in the global non-inferior solution scheme set, k is a constant to represent the first k schemes closest to the a scheme in the global non-inferior solution scheme set,concentrating the (r) scheme closest to the (a) scheme for the global non-inferior solution scheme;
wherein P is a The probability decision value of the scheme a is in the global non-inferior solution scheme set, and the phi is the number of the global non-inferior solution schemes.
The invention has the following advantages:
1) The intelligent optimizing and searching method can conduct intelligent optimizing and searching of the railway line in a continuous space, comprehensively considers economy and environmental protection, recommends a high-quality line alternative scheme for a designer, assists line design, and improves quality and efficiency of the line design.
2) The method has the advantages of high automation degree, strong practicability, high operation efficiency and high popularization and application value.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a schematic diagram of a set of non-bad solutions for each particle;
fig. 3 is a plan layout of the intelligent optimization scheme of the county-Taiping railway.
Fig. 4 is a longitudinal circuit diagram of the intelligent optimization scheme of the county-Taiping railway.
Detailed Description
The present invention will be described in detail with reference to the following embodiments.
The invention relates to a multi-objective optimization method for a railway line in a complex mountain area, which comprises the following steps of:
S 1 : establishing a comprehensive geographic information model:
S 1-1 : establishing a railway route selection information storage structure based on a grid form;
S 1-2 : and collecting various information required by line searching, dispersing the information into a grid, and establishing a railway line selection comprehensive geographic information model.
S 2 : constructing a railway line multi-objective optimization model which cooperatively considers economic and environmental factors:
S 2-1 : analyzing key parameters affecting railway space line positions, and determining independent variables of a multi-objective optimization model of the railway line;
S 2-2 : establishing a functional relation between independent variables of a railway multi-objective optimization model and economic evaluation targets, and constructing a line economic evaluationA valence target function;
S 2-3 : establishing a functional relation between independent variables of a railway multi-objective optimization model and environmental impact evaluation targets, and constructing a line environmental impact evaluation target function;
S 2-4 : and establishing constraint conditions of the railway line optimization model.
S 3 : and carrying out multi-objective optimization model calculation on the railway line under continuous space based on a particle swarm algorithm:
S 3-1 : generating an initial particle population, wherein the initial position of each particle represents each initial line scheme;
S 3-2 : constructing a storage file of the non-inferior solution scheme of each particle, storing the non-inferior solution scheme generated in the iterative evolution process of each particle into the respective storage file, generating a non-inferior solution scheme set of each particle, and respectively selecting an individual optimal scheme of each particle from the non-inferior solution scheme set;
S 3-3 : constructing a non-inferior solution scheme storage file of the whole population, storing the non-inferior solution scheme generated in the iterative evolution process of the whole population into the file, generating a non-inferior solution scheme set of the whole population, and selecting a global optimal scheme from the non-inferior solution scheme set;
S 3-4 : guiding population evolution based on the individual optimal scheme and the global optimal scheme;
S 3-5 : and iterating the evolutionary population until reaching an iteration termination condition.
S 4 : will pass through step S 3 And carrying out normalization processing on each optimization target of the line schemes contained in the obtained non-inferior solution set.
Said step S 2-3 The constructed environment evaluation objective function comprises a vegetation destruction degree objective function and a soil erosion degree objective function:
(1) Vegetation damage level objective function: the invention introduces a normalized vegetation index (NDVI) in the remote sensing field into a railway line optimization model for the first time, and provides a railway line scheme vegetation damage degree quantitative evaluation method based on the normalized vegetation index.
NDVI is the most widely used vegetation detection index in the remote sensing field, and can be used to detect vegetation growth status and vegetation coverage, where NDVI values are between-1 and 1, negative values indicate that the ground coverage is cloudy, water, snow, etc., high reflection to visible light, 0 indicates that there is rock or bare soil, etc., positive values indicate that there is vegetation coverage, and increase as coverage increases. The latest NDVI data can be downloaded at NASA official website or geographical national condition monitoring cloud platform. The invention disperses the downloaded NDVI data into each grid of the comprehensive geographic information model, adds positive NDVI values of the areas where the lines pass through in the form of roadbed, and takes the accumulated NDVI values as quantitative evaluation indexes of vegetation damage degree of the line scheme, as shown in a formula (1).
Wherein SV is NDVI An accumulated value of NDVI of the line passing area, C (k) The kth grid for the passage of the line in the form of a roadbed, U S For a grid set where the line passes in the form of a road bed,NDVI values, max { }, for the kth grid through which the line passes in a roadbed form represent maximum values.
(2) Water and soil loss degree objective function: the invention provides a railway line scheme soil erosion degree quantitative evaluation method based on spoil and slag removal quantity. The invention uses the volume of the waste soil as the representative value of the water and soil loss degree to evaluate the water and soil loss degree of different line schemes because the waste soil and waste slag is the main reason of water and soil loss caused by railway engineering construction. Spoil volume (V S ) Through the volume of the roadbed (V) SC ) Volume of tunnel excavation (V TC ) Roadbed filling volume (V SF ) And (3) calculating as shown in the formula (2).
V S ＝max{V SC +V TC -V SF ,0} (2)
Said step S 3 In the method, aiming at a railway line multi-objective optimization model, the invention provides a multi-objective optimization based on a particle swarm algorithmThe model resolving method abstracts the railway line optimizing process into a process of continuously searching the optimal position of a group of particles in a multidimensional space. During the search, each location in the population where each particle passes through in the multidimensional space represents a routing scheme. And updating the respective positions of each particle based on the current individual optimal particle position and the global optimal particle position to obtain a new line scheme. The individual optimal particle positions are the optimal positions in the positions where single particles pass, each particle has its optimal position, i.e. the individual optimal scheme, and the global optimal particle positions are the optimal positions in the positions where all particles in the population pass, i.e. the global optimal scheme. However, for the multi-objective optimization problem of railway lines, a line scheme with all optimization objectives being optimal cannot be generally found, and a set of non-inferior solution schemes which cannot be governed by each other is obtained. In contrast, the invention provides a decision method of an individual optimal scheme and a global optimal scheme aiming at the railway line multi-objective optimization problem
(1) Individual optimal scheme decision method
Individual optimal solution decisions should take into account the goodness of the solution. The individual optimal solution should have the ability to direct the evolution of particles in a direction that improves the solution's goodness. In this regard, the invention provides an individual optimal scheme decision method based on the concept of tournament. Firstly, establishing a storage file for each particle, storing all non-inferior solution schemes obtained by searching the current generation by each particle, and establishing a non-inferior solution scheme set of each particle; then, each optimization target of each non-inferior solution scheme is compared in turn, the winning value of each non-inferior solution scheme is calculated by establishing a tournament equation (formula (3)), and the sorting value of each non-inferior solution scheme is calculated by establishing a sorting function (formula (5)), so that the sorting value of each non-inferior solution scheme is sorted, and the larger the sorting value of the scheme is, the more front the ranking is; and finally, selecting the first scheme as an individual optimal scheme of each particle.
Wherein phi is i Is the ith grainNon-bad solution scheme set of sub-T (j) (a,Φ i ) Optimizing the winning value of the objective, |Φ, for the j-th term of the a-scheme in the non-bad solution set of the i-th particle i I is the number of non-bad solutions of the ith particle, t (j) (a, b) optimizing an objective function for a line constructed by the present invention for the winning value of the j-th optimization objective of the a-scheme and the b-scheme in the non-inferior solution scheme set of the i-th particle, the smaller the objective value, the better the scheme, and thus t (j) (a, b) can be calculated from formula (4).
Wherein f (j) (a) And f (j) (b) The target value is optimized for the j-th item of the a-scheme and the b-scheme, respectively.
Wherein R (a) is the sorting value of the scheme a, m is the optimized target quantity, w j The target weights are optimized for each item.
(2) Global optimal scheme decision method
The global optimal scheme decision should consider not only the quality of the scheme, but also the uniformity of the population distribution throughout the solution space. The global optimal scheme should have the capability of improving the performance of each scheme in the population and the diversity of the population. In this regard, the present patent proposes a global optimal solution decision method combining the concept of tournaments and congestion. Firstly, storing a non-inferior solution scheme obtained by searching the current generation by all particles, and establishing a global non-inferior solution scheme set; then, calculating the distance between each scheme in the global non-inferior solution set and other schemes in a multidimensional space through a formula (6), constructing a crowding degree function (formula (7)), calculating the crowding degree of each scheme, and sequencing each scheme in the global non-inferior solution set according to the crowding degree, wherein the lower the crowding degree is, the more front the sequencing is, so as to obtain a sequence R1; sequencing all schemes in the global non-inferior solution set through a tournament algorithm to obtain a sequence R2; and finally, constructing a probability decision equation (formula (8)) comprehensively considering R1 and R2 based on the roulette idea, and selecting a scheme with the maximum probability value as a global optimal solution scheme.
Wherein D (a, b) is the distance between the a scheme and the b scheme in the multi-dimensional solution space in the global non-inferior solution set, m is the number of optimization targets, f (j) (a) Optimizing target value for j-th item of a scheme in global non-inferior solution scheme set, f (j) (b) Optimizing the target value for the j-th term of the b-scheme in the global non-bad solution set,maximum value of the objective of optimization for item j in the global non-bad solution,/>And optimizing the minimum value of the target for the j-th item in the global non-inferior solution scheme.
Wherein ρ is a For the crowding value of the a scheme in the global non-inferior solution scheme set, k is a constant to represent the first k schemes closest to the a scheme in the global non-inferior solution scheme set,the r-th scheme closest to the a-scheme is concentrated for the global non-bad solution scheme.
Wherein P is a The probability decision value of the scheme a is in the global non-inferior solution scheme set, and the phi is the number of the global non-inferior solution schemes.
Step S 4 In (2) through Min-max normalizationThe method (formula (9)) performs normalization processing on each optimization target of the line scheme:
wherein x is i Optimizing target value for item i, min x i Optimizing the minimum value of the objective for the ith item in all schemes, max x i The maximum value of the objective is optimized for the ith term in all schemes.
Examples:
the technical scheme of the invention is further described by taking a county-Tai flat railway as an example. As shown in fig. 1, the multi-objective optimization method for the railway line in the complex mountain area comprises the following steps:
S 1 : establishing a comprehensive geographic information model:
S 1-1 : dividing a rectangular area with a line selection range of 56km multiplied by 27km into a regular grid containing 1867 multiplied by 900 square cells, wherein the width of each cell is 30m;
S 1-2 : and collecting various information required by line searching, including terrain, ground objects, vegetation coverage, engineering unit price and starting and ending point coordinates. Wherein the topographic information is digital elevation information; the ground object information comprises river information, existing road network information and forbidden zone information; the vegetation coverage information is normalized vegetation index (NDVI) information reflecting vegetation growth status and coverage; the engineering unit price information comprises land occupation cost information, track laying cost information, filling engineering cost information, excavation engineering cost information, bridge engineering cost information, tunnel portal cost information, bridge pier cost information and operation cost information; the coordinates of the starting point and the ending point of the line are respectively as follows: flourishing county (3506944.120,35390718.808,1508.91), too flat (3556583.808,35380569.235,2275.204). And dispersing the collected information into a grid, and establishing a railway route selection comprehensive geographic information model.
S 2 : constructing a railway line multi-objective optimization model which cooperatively considers economic and environmental factors:
S 2-1 : to be used forRailway plane line intersection point coordinates (X) i ,Y i ) Radius R of railway plane line circle curve i Railway longitudinal surface slope change point mileage K i Elevation H of slope changing point of railway longitudinal surface i For the multi-objective optimization model independent variable of the railway line, constructing a model independent variable vector: x, Y, R, K, H;
S 2-2 : based on the independent variables of the line multi-objective optimization model, construction and consideration of bridge engineering cost (C B ) Tunnel engineering cost (C) T ) Earthwork costs (C) E ) Track laying cost (C) L ) Cost of occupation (C) R ) The operating costs associated with the total line angle (C A ) An operation fee (C) related to the total length of the line M ) Line gradient dependent operating costs (C G ) As shown in the formula (13);
f C (X,Y,R,K,H)＝Δ(C B +C T +C E +C C +C R )+C A +C M +C G (13)
S 2-3 : constructing and considering vegetation damage degree (SV) based on independent variables of line multi-objective optimization model NDVI ) And the soil erosion degree (V) S ) The line environmental impact evaluation objective function of (2) is represented by the following formulas (14) and (15);
S 2-4 : and establishing railway line optimization model constraint conditions including gradient constraint, forbidden zone constraint, maximum bridge height constraint and maximum tunnel length constraint. The gradient constraint of the railway case (maoxian-tai) is 24 per mill, the maximum bridge height constraint is 100m, and the maximum tunnel length is 20000m.
S 3 : and carrying out multi-objective optimization model calculation on the railway line under continuous space based on a particle swarm algorithm:
S 3-1 : randomly generating 100 initial particles, wherein the initial position of each particle represents each initial line scheme;
S 3-2 : constructing a storage file (shown in figure 2) of the non-inferior solution schemes of each particle, storing the non-inferior solution schemes generated in the iterative evolution process of each particle into respective storage files, generating a non-inferior solution scheme set of each particle, sequentially comparing economic evaluation targets, vegetation damage degree evaluation targets and water and soil loss degree damage evaluation targets of each non-inferior solution scheme based on a tournament idea, calculating the ranking value of each non-inferior solution scheme, and selecting the scheme with the highest ranking value as an individual optimal scheme in each iterative evolution;
S 3-3 : constructing a non-inferior solution storage file of the whole population, storing the non-inferior solution generated in the iterative evolution process of the whole population into the file, generating a non-inferior solution set of the whole population, sequentially calculating the crowdedness of each non-inferior solution based on the crowdedness idea, sorting the non-inferior solution based on the crowdedness value, sorting the non-inferior solution based on the tournament idea, calculating the probability value of each non-inferior solution according to a probability decision equation combining two sorting values, and selecting the scheme with the maximum probability value as the global optimal scheme in each iterative evolution;
S 3-4 : updating the particle position according to formula (16), formula (17) based on the individual optimal solution and the global optimal solution;
Vel i (t+1)＝w·Vel i (t)+C 1 ·r 1 (t)·[gbest(t)-Pos i (t)]+C 2 ·r 2 (t)·[pbest(t)-Pos i (t)] (16)
Pos i (t+1)＝Pos i (t)+Vel i (t+1) (17)
wherein, vel i (t) and Vel i (t+1) is the velocity of the ith particle at the t-th and t+1-th generations, C 1 And C 2 As learning factors, C is usually taken 1 ＝C 2 ＝2，r 1 And r 2 Is uniformly distributed in [0,1 ]]Random number in interval, w is inertial weight, pos i (t) and Pos i (t+1) distribution as the firstThe i particles are at the t-th and t+1-th generation positions.
S 3-5 : the population is continuously evolved until 500 generations are reached, terminating the iteration.
S 4 : will pass through step S 3 And carrying out normalization processing on each optimization target of the line schemes contained in the obtained non-inferior solution scheme set to obtain a flat and longitudinal plane scheme of the county-Tai flat railway recommended scheme, wherein the flat and longitudinal plane scheme is shown in figure 3.
The content of the invention is not limited to the examples listed, and any equivalent transformation to the technical solution of the invention that a person skilled in the art can take on by reading the description of the invention is covered by the claims of the invention.
Claims (2)
1. The multi-objective optimization method for the railway lines in the complicated mountain areas is characterized by comprising the following steps of:
the method comprises the following steps:
S 1 : establishing a comprehensive geographic information model:
S 2 : constructing a railway line multi-objective optimization model which cooperatively considers economic and environmental factors:
S 3 : and carrying out multi-objective optimization model calculation on the railway line under continuous space based on a particle swarm algorithm:
S 4 : will pass through step S 3 Carrying out normalization processing on each optimization target of the line scheme contained in the obtained non-inferior solution scheme set;
said step S 3 Comprising the following steps:
S 3-1 : generating an initial particle population, wherein the initial position of each particle represents each initial line scheme;
S 3-2 : constructing a storage file of the non-inferior solution scheme of each particle, storing the non-inferior solution scheme generated in the iterative evolution process of each particle into the respective storage file, generating a non-inferior solution scheme set of each particle, and respectively selecting an individual optimal scheme of each particle from the non-inferior solution scheme set;
S 3-3 : constructing a storage file of a non-inferior solution scheme of the whole population, and storing the non-inferior solution scheme generated in the iterative evolution process of the whole population into the storage fileGenerating a non-inferior solution set of the whole population, and selecting a global optimal solution from the non-inferior solution set;
S 3-4 : guiding population evolution based on the individual optimal scheme and the global optimal scheme;
S 3-5 : iteratively evolving the population until reaching an iteration termination condition;
said step S 3 The global optimal scheme is obtained by the following steps:
firstly, storing a non-inferior solution scheme obtained by searching the current generation by all particles, and establishing a global non-inferior solution scheme set; then, calculating the distance between each scheme in the global non-inferior solution set and other schemes in a multidimensional space through a formula (6), constructing a crowding degree function (7), calculating the crowding degree of each scheme, and sequencing each scheme in the global non-inferior solution set according to the crowding degree, wherein the lower the crowding degree is, the more front the sequencing is, so as to obtain a sequence R1; sequencing all schemes in the global non-inferior solution set through a tournament algorithm to obtain a sequence R2; finally, based on the roulette idea, constructing a probability decision equation (8) comprehensively considering R1 and R2, and selecting a scheme with the maximum probability value as a global optimal solution scheme;
wherein D (a, b) is the distance between the a scheme and the b scheme in the multi-dimensional solution space in the global non-inferior solution set, m is the number of optimization targets, f (j) (a) Optimizing target value for j-th item of a scheme in global non-inferior solution scheme set, f (j) (b) Optimizing the target value for the j-th term of the b-scheme in the global non-bad solution set,the maximum value of the objective is optimized for the jth term in the global non-bad solution,optimizing the minimum value of the target for the j-th item in the global non-inferior solution scheme;
wherein ρ is a For the congestion level value of scheme a in the global non-bad solution set, the rank of scheme a, ordered according to the congestion level value, is noted as R1 (a) K is a constant representing the top k schemes closest to the a scheme in the set of global non-bad solutions,concentrating the (r) scheme closest to the (a) scheme for the global non-inferior solution scheme;
wherein P is a The probability decision value of the scheme a is concentrated for the global non-inferior solution scheme, and the I phi I is the number of the global non-inferior solution schemes; r1 (a) The sorting value of the scheme a after sorting according to the congestion degree value; r2 (a) Ranking the tournament values for scheme a;
said step S 3 The obtaining process of the individual optimal scheme specifically comprises the following steps:
firstly, establishing a storage file for each particle, storing all non-inferior solution schemes obtained by searching the current generation by each particle, and establishing a non-inferior solution scheme set of each particle; then, each optimization target of each non-inferior solution scheme is compared in sequence, the winning value of each non-inferior solution scheme is calculated by establishing a tournament equation (3), the sorting value of each scheme is calculated by establishing a sorting function formula (5), the non-inferior solution schemes are sorted, and the larger the sorting value of the scheme is, the higher the ranking is; finally, selecting the first scheme as an individual optimal scheme of each particle;
wherein phi is i For the non-bad solution set of the ith particle, T (j) (a,Φ i ) Optimizing the winning value of the objective, |Φ, for the j-th term of the a-scheme in the non-bad solution set of the i-th particle i I is the number of non-bad solutions of the ith particle, t (j) (a, b) optimizing an objective function for a constructed line for a winning value of a j-th optimization objective of a scheme and b scheme in a non-inferior solution scheme set of the i-th particle, the smaller the target value, the better the scheme, thus t (j) (a, b) is calculated from formula (4):
wherein f (j) (a) And f (j) (b) Optimizing target values for the j-th items of the a scheme and the b scheme respectively;
wherein R2 (a) A championship match sequencing value of a scheme, m is an optimization target quantity, w j Optimizing target weights for each item;
said step S 2 Comprising the following steps:
S 2-1 : analyzing key parameters affecting railway space line positions, and determining independent variables of a multi-objective optimization model of the railway line;
S 2-2 : establishing a functional relation between independent variables of a railway multi-objective optimization model and economic evaluation targets, and constructing a line economic evaluation target function;
S 2-3 : establishing a functional relation between independent variables of a railway multi-objective optimization model and environmental impact evaluation targets, and constructing a line environmental impact evaluation target function;
S 2-4 : establishing constraint conditions of a railway line optimization model;
said step S 2-3 In the method, the constructed environment evaluation objective function comprises a vegetation destruction degree objective function and water and soilLoss degree objective function:
vegetation damage level objective function: introducing a normalized vegetation index NDVI into a railway line optimization model, dispersing NDVI data into each grid of a comprehensive geographic information model, adding positive NDVI values of a road passing region in a roadbed form, and taking the accumulated NDVI values as quantitative evaluation indexes of vegetation damage degrees of the road scheme, wherein the quantitative evaluation indexes are shown in a formula (1):
wherein SV is NDVI An accumulated value of NDVI of the line passing area, C (k) The kth grid for the passage of the line in the form of a roadbed, U S For a grid set where the line passes in the form of a road bed,NDVI values for the kth grid through which the line passes in a roadbed form, max { } represents taking the maximum value;
water and soil loss degree objective function: taking the waste soil and waste slag volume as a representative value of the water and soil loss degree, and the waste soil and waste slag volume V S By the volume V of the roadbed excavation SC Volume V of tunnel excavation TC Volume V of roadbed filling SF Calculation, as shown in formula (2):
V S ＝max{V SC +V TC -V SF ,0} (2)。
2. the multi-objective optimization method for the railway line in the complex mountain area according to claim 1, wherein the method comprises the following steps:
said step S 1 Comprising the following steps:
S 1-1 : establishing a railway route selection information storage structure based on a grid form;
S 1-2 : and collecting various information required by line searching, dispersing the information into a grid, and establishing a railway line selection comprehensive geographic information model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110471218.5A CN113240163B (en) | 2021-04-29 | 2021-04-29 | Multi-objective optimization method for railway lines in complicated mountain areas |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110471218.5A CN113240163B (en) | 2021-04-29 | 2021-04-29 | Multi-objective optimization method for railway lines in complicated mountain areas |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113240163A CN113240163A (en) | 2021-08-10 |
| CN113240163B true CN113240163B (en) | 2024-02-27 |
Family
ID=77131428
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110471218.5A Active CN113240163B (en) | 2021-04-29 | 2021-04-29 | Multi-objective optimization method for railway lines in complicated mountain areas |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113240163B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114154382B (en) * | 2021-11-18 | 2022-11-15 | 中国公路工程咨询集团有限公司 | High-resolution remote sensing three-dimensional visual railway route selection auxiliary decision making system and method |
| CN117726196B (en) * | 2024-02-18 | 2024-05-14 | 中南大学 | Comprehensive evaluation method for environment suitability for railway route selection |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106529166A (en) * | 2016-11-04 | 2017-03-22 | 河海大学 | Method for regional water-source optimized configuration based on MAEPSO algorithm |
| CN112487695A (en) * | 2020-11-30 | 2021-03-12 | 中南大学 | Multi-target intelligent comprehensive line selection method for railway in complex environment |
-
2021
- 2021-04-29 CN CN202110471218.5A patent/CN113240163B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106529166A (en) * | 2016-11-04 | 2017-03-22 | 河海大学 | Method for regional water-source optimized configuration based on MAEPSO algorithm |
| CN112487695A (en) * | 2020-11-30 | 2021-03-12 | 中南大学 | Multi-target intelligent comprehensive line selection method for railway in complex environment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113240163A (en) | 2021-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhu et al. | Prediction of rockhead using a hybrid N-XGBoost machine learning framework | |
| Wu et al. | Optimal site selection for parabolic trough concentrating solar power plant using extended PROMETHEE method: A case in China | |
| CN113240163B (en) | Multi-objective optimization method for railway lines in complicated mountain areas | |
| Pillot et al. | An integrated GIS and robust optimization framework for solar PV plant planning scenarios at utility scale | |
| CN112487695B (en) | Multi-target intelligent comprehensive line selection method for railway in complex environment | |
| Abudeif et al. | Multicriteria decision analysis based on analytic hierarchy process in GIS environment for siting nuclear power plant in Egypt | |
| CN112417573B (en) | GA-LSSVM and NSGA-II shield tunneling multi-objective optimization method based on existing tunnel construction | |
| Aruga et al. | Heuristic planning techniques applied to forest road profiles | |
| CN109214449A (en) | A kind of electric grid investment needing forecasting method | |
| CN110046213B (en) | Power line selection method considering path distortion correction and cross crossing correction | |
| CN111639878B (en) | Landslide risk prediction method and system based on knowledge graph construction | |
| CN106600959A (en) | Traffic congestion index-based prediction method | |
| CN112906950A (en) | Sponge city planning and designing system | |
| CN113554467A (en) | Railway three-dimensional linear intelligent design method based on co-evolution | |
| CN117745466B (en) | Tailing pond counting intelligent operation and maintenance system | |
| Deng et al. | Fuzzy rule based classification method of surrounding rock stability of coal roadway using artificial intelligence algorithm | |
| Kharazi et al. | Delineation of suitable sites for groundwater dams in the semi-arid environment in the northeast of Iran using GIS-based decision-making method | |
| Ajrina et al. | Mining Zone Determination of Natural Sandy Gravel using Fuzzy AHP and SAW, MOORA and COPRAS Methods. | |
| CN105887593B (en) | The three-dimensional leading line method for selecting of Permafrost Area highway | |
| CN115310878B (en) | Power space planning design method based on homeland space element data | |
| Heyns et al. | An evaluation of the effectiveness of observation camera placement within the MeerKAT radio telescope project | |
| Dražić et al. | Evaluation of Morphometric Terrain Parameters and Their Influence on Determining Optimal Density of Primary Forest Road Network | |
| CN110457831A (en) | A kind of covering suitability evaluation methods of groundwater source field | |
| CN115526544A (en) | Green mine construction planning method and system based on multi-type data | |
| CN109508890A (en) | The calculation method that biomass thermal power plant raw material bleeding point influences transportation cost |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |