Mountainous area limited terrain bridge construction site planning method

Abstract

The invention discloses a mountainous area limited terrain bridge construction site planning method. The method comprises the steps that a three-dimensional building information model of a construction area is constructed based on terrain and geological data; determining an optimization target and a constraint condition; constructing a multi-factor collaborative optimization model by taking the space coordinates and geometric parameters of the center of the construction site as design variables; constructing a comprehensive fitness function based on the optimization target and the penalty function; performing iterative optimization on the multi-factor collaborative optimization model by adopting a starfish algorithm to obtain a solution with the maximum comprehensive fitness; and displaying the optimal construction site arrangement scheme through the building information model. According to the method, scheme optimization comprehensively considering site safety, engineering economy and construction convenience is realized, an efficient and accurate optimization method is provided for mountain rugged terrain bridge construction site planning, and construction cost can be effectively reduced while safety is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of bridge construction, and more particularly to a method for planning bridge construction sites in mountainous, restricted terrain. Background Technology

[0002] In the construction of bridges in mountainous areas, the rational layout of the construction site plays a crucial role in construction efficiency, cost control, and safety. Mountainous terrain is complex and varied, often featuring rugged terrain, towering slopes, and narrow valleys, posing significant challenges to the layout of the construction site. Traditional methods of site layout rely heavily on manual experience; however, this approach has many limitations when facing complex mountainous environments.

[0003] On the one hand, the construction of prestressed concrete beam bridges in mountainous areas involves large precast components, requiring large storage areas and high-standard temporary transport roads. However, mountainous areas have limited site resources, and the construction of temporary roads is difficult, making it hard to meet these needs. Furthermore, bridge construction sites in mountainous areas often interfere with the construction of nearby tunnels, leading to frequent construction disruptions. Manual experience alone is insufficient to coordinate these two aspects, easily resulting in unreasonable construction arrangements. On the other hand, traditional manual experience-based site layout methods cannot fully consider factors such as complex terrain, geological safety, and construction costs. Complex terrain affects the layout of construction equipment and the safety of material transportation; geological safety issues may lead to geological disasters during construction; and construction costs are difficult to optimize and control. Moreover, the layout of bridge construction sites in mountainous areas needs to comprehensively consider multiple interdependent objectives such as economy, safety, and applicability. Manual experience alone cannot find a balance, easily leading to unreasonable site layouts. In addition, the construction process of bridges in mountainous areas is subject to significant dynamic changes, and the timeliness and accuracy of traditional manual experience-based adjustments are difficult to guarantee, easily causing delays and resource waste. Against this backdrop, the development of a method for the layout of bridge construction sites in mountainous areas that can comprehensively consider multiple factors and achieve efficient and intelligent optimization is particularly urgent.

[0004] For the planning of construction sites in confined spaces, transforming the multivariate parameter optimization problem under constraints into a mathematical optimization model and introducing intelligent algorithms for iterative optimization has already seen some research and application. Compared with traditional optimization methods, intelligent optimization algorithms have a natural advantage in solving such complex engineering problems. However, the construction of multiple types of design variables, multiple optimization objectives, constraints, and fitness functions for complex optimization problems in bridge construction sites in mountainous and rugged terrain still requires further improvement and refinement. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention proposes a site planning method for bridge construction in mountainous and restricted terrain. This method can efficiently find the optimal multi-site planning scheme in rugged mountainous terrain and can also adjust the optimization objectives according to the characteristics of different sites and the needs of different construction stages, so as to better optimize the layout of different types and numbers of sites under different terrain and geological conditions.

[0006] The specific technical solution is as follows: A method for planning bridge construction sites in mountainous, confined terrain includes the following steps: S1: Construct a three-dimensional building information model of the construction area based on terrain and geological data; S2: Determine the optimization objectives and constraints; the optimization objectives include: maximizing the area of ​​the construction site within the construction area, while minimizing the ruggedness of the construction site, the amount of leveling work, and the steepness of the site slope; the constraints include: the type, quantity, and size requirements of the construction site, as well as restricted terrain that cannot be crossed and terrain with geological risks. S3: Using the spatial coordinates and geometric parameters of the construction site center as design variables, construct a multi-factor collaborative optimization model; construct a comprehensive fitness function based on the optimization objective, constraints, and penalty function. S4: The starfish algorithm is used to iteratively optimize the multi-factor collaborative optimization model to obtain the solution with the highest comprehensive fitness, which is the optimal construction site layout scheme. The scheme includes the spatial coordinates and geometric parameters of the center of the construction site. S5: Display the optimal construction site layout scheme through building information modeling.

[0007] Furthermore, the expression for the multi-factor collaborative optimization model is as follows: ; In the formula, X i ={x i ,y i ,z i ,l i ,b i ,θ i} T For the design variables of the i-th site, (x i ,y i ,z i ) represents the center coordinates of the i-th site, and the geometric parameters (l) i ,b i ,θ i Let W be the length, width, and orientation angle of the i-th field, n be the number of fields, and W be the coordinates of the field. i Let A be the weight of the i-th site; A is the area of ​​the site. min,i A max,i Let G be the minimum and maximum area of ​​the i-th site, respectively;s R represents the comprehensive geological safety coefficient of the site, E represents the ruggedness of the site, and S represents the steepness of the site slope. min,i l max,i b represents the minimum and maximum lengths of the i-th site, respectively; min,i b max,i Let P be the minimum and maximum values ​​of the width of the i-th field, respectively; res For the restricted terrain distance penalty, P imp This is a penalty for terrain that cannot be crossed.

[0008] Furthermore, the value of the restricted terrain distance penalty term is the average distance from each uniformly distributed point on the edge of the site to the coordinates of the nearest restricted terrain geometric center point; If the site area overlaps with the uncrossable terrain area, the value of the uncrossable terrain penalty term is positive infinity; otherwise, the value of the uncrossable terrain penalty term is 0.

[0009] Furthermore, the comprehensive geological safety coefficient G of the site s The value is the average distance from each uniformly distributed point on the edge of the site to the coordinates of the nearest geological safety control point; The ruggedness R of the site is the sum of the absolute values ​​of the height differences between all adjacent uniformly distributed points within the site area; The value of the site leveling work volume E is the amount of cut and fill material leveled to the center elevation of the site. The steepness S of the site slope is taken as the reciprocal of the slope corresponding to the location of the maximum slope at the edge of the site.

[0010] Furthermore, the expression for the comprehensive fitness function is as follows: ; ; In the formula, Fitness represents the overall adaptability to the terrain. i R represents the fitness of the i-th site, where w1~w6 are weighting coefficients set manually; avg E represents the average ruggedness within the site area. dis S represents the initial site leveling work volume. avg This indicates the average steepness of the slope.

[0011] Furthermore, S4 is specifically implemented through the following sub-steps: S401: Set the basic parameters of the starfish algorithm, including: population size, maximum number of iterations, and number of generations of stagnation for each individual; randomly generate several starfish individuals based on the basic parameters to complete the initialization of the starfish population; each starfish individual represents a construction site layout scheme; S402: Using the comprehensive fitness function, calculate the comprehensive fitness value of each starfish individual, and sort the starfish individuals from high to low according to the comprehensive fitness value; S403: Simulate the movement behavior of starfish, perform a global search, and determine the optimal solution area; S404: Simulate the predatory behavior of starfish, perform a local search of the area surrounding the current optimal solution, and update the position of individual starfish using a parallel bidirectional search strategy; S405: Simulates the regeneration behavior of starfish, and performs local redevelopment on individuals with the lowest overall fitness to avoid the algorithm getting stuck in local optima; Meanwhile, if an individual in the population has exceeded the threshold for the number of generations of stagnation but its fitness has not improved, it is determined that it has entered a stagnation state and is forcibly triggered to carry out local redevelopment. S406: Repeat the above steps until any one of the iteration termination conditions is met, and output the optimal construction site layout scheme; the iteration termination conditions include: reaching the maximum number of iterations, the change in the population position of the starfish population being less than the set change threshold, and the optimal individual comprehensive fitness value converging and stabilizing.

[0012] Furthermore, in step S401, feasibility repair of the randomly generated population includes: If a construction site within an individual starfish falls into an insurmountable terrain, then the entire site is pushed outward along the nearest gradient direction to the boundary of the insurmountable terrain. If the site geometry does not meet the site size requirements, it should be scaled down to the required range.

[0013] Furthermore, in S404, the O starfish individuals with the highest overall fitness are selected as dominant individuals, and multiple sub-starfish individuals are generated within a set range around each dominant individual. The distances between the dominant individual and other starfish across multiple design variable dimensions are calculated. Two of these distances are randomly selected, and starfish predation behavior is simulated based on a parallel bidirectional search strategy to update the position of the starfish individual. The multiple design variables include: spatial coordinates x, y, z, and geometric parameters such as site length l, site width b, and site orientation angle θ. Calculate the overall fitness of all updated starfish offspring, compare each starfish offspring with its corresponding original dominant offspring, and retain the O starfish offspring with the highest overall fitness as the new dominant offspring.

[0014] Furthermore, the optimization objectives also include: distance between sites and transportation distance of components or equipment.

[0015] The beneficial effects of this invention are: This invention constructs an MFCO model and a CFF function, and introduces a biomimetic algorithm for scheme optimization. It fully utilizes the global exploration and local development capabilities of the biomimetic algorithm to achieve scheme optimization that comprehensively considers site safety, engineering economy and construction convenience. It provides an efficient and accurate optimization method for site planning of bridge construction in mountainous and rugged terrain, and can effectively reduce construction costs while ensuring safety. Attached Figure Description

[0016] Figure 1 This is a flowchart of a method for planning construction sites for bridges in mountainous, restricted terrain, according to an embodiment of the present invention.

[0017] Figure 2 This is a schematic diagram of the design variables for the site planning scheme in an embodiment of the present invention.

[0018] Figure 3 This is a topographical diagram of a mountainous bridge construction area in an embodiment of the present invention.

[0019] Figure 4 This is a schematic diagram of the site layout optimization results in an embodiment of the present invention.

[0020] Figure 5 This is a diagram illustrating the iterative convergence process of the optimization algorithm in an embodiment of the present invention. Detailed Implementation

[0021] The present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments. The objectives and effects of the present invention will become clearer as a result. The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0022] like Figure 1 As shown, a method for planning bridge construction sites in mountainous, confined terrain includes the following steps: S1: Construction Area Modeling. Based on high-precision topographic maps, satellite images, and engineering geological survey data, determine the topography and geological information of the construction site, and construct a three-dimensional building information model (BIM) of the construction area containing geological risk information; and determine the location coordinates and range of unfavorable geological factors that need to be avoided, the coordinates of geological safety control points, and the location coordinates and range of insurmountable topographic constraints.

[0023] S2: Determination of optimization objectives and constraints (based on the actual situation of bridge construction in mountainous areas and relevant specifications).

[0024] The optimization objectives include maximizing the area of ​​each construction site while minimizing the overall terrain ruggedness, leveling work, and slope steepness.

[0025] The constraints include design parameter constraints and topographic and geological constraints. Specifically, design parameter constraints are determined based on the bridge design, specifying the type and quantity of construction sites required, as well as the dimensional requirements for each type of construction site (including minimum area, maximum area, length range, and width range). Topographic and geological constraints include insurmountable terrain and restricted terrain. Insurmountable terrain, such as steep cliffs, deep canyons, and rapid rivers, is designated as a non-overlapping area constraint for the construction site. Restricted terrain, such as landslide-prone areas, debris flow-risk areas, and areas with weak foundations, is determined based on geological surveys and assessments, and its location is a geological safety constraint that the construction site should avoid.

[0026] S3: Using the spatial coordinates and geometric parameters of the centers of each construction site in the construction area as design variables, a multi-factor collaborative optimization model (MFCO) is constructed. Based on the MFCO model, a comprehensive fitness function (CFF) is proposed for planning bridge construction sites in mountainous and restricted terrain to address the conflict between objectives during the optimization process.

[0027] The MFCO model is constructed based on design variables, optimization objectives, and constraints to solve multi-factor optimization problems involving site area, flatness, leveling work volume, and slope steepness. The expression of the MFCO model is as follows: ; In the formula, X i ={x i ,y i ,z i ,l i ,b i ,θ i} T Let i be the design variable for the i-th construction site (hereinafter referred to as the site), such as Figure 2 As shown, (x i ,y i ,z i Let be the center coordinates of the i-th site, and the remaining parameters are geometric parameters. i Let b be the length of the i-th site. i Let θ be the width of the i-th field. i W represents the orientation angle of the i-th field, and n represents the number of fields. i Let G be the weight of the i-th site (set manually based on actual conditions). A is the total area of ​​the site, and G is the weight of the i-th site. sis the comprehensive geological safety coefficient for the site, used to reflect the geological stability of the site. It is the average distance from each uniformly distributed point on the site edge to the coordinates of the nearest geological safety control point. R is the ruggedness of the site, taken as the sum of the absolute values ​​of the height differences between all adjacent uniformly distributed points within the site area. The points are spaced 0.5m apart, and can be further densified if necessary. E is the site leveling volume, taken as the cut and fill volume required to level the site to the elevation of the site center. S is the steepness of the site slope, taken as the reciprocal of the slope corresponding to the location of the maximum slope at the site edge. A min,i Let A be the minimum area of ​​the i-th site. max,i The maximum value of the area of ​​the i-th site; l min,i Let l be the minimum length of the i-th site. max,i b is the maximum value of the length of the i-th site; min,i b is the minimum value of the width of the i-th site. max,i P represents the maximum width of the i-th site. res The distance penalty term for restricted terrain is the average distance from each uniformly distributed point on the site edge to the nearest geometric center point of the restricted terrain; P imp This is a penalty term for uncrossable terrain. If the site area overlaps with the uncrossable terrain area, the value is +∞ (the solution is directly eliminated), otherwise it is 0.

[0028] The MFCO model, based on the optimization objective set in S2—maximizing the total area A of all construction sites while minimizing the site ruggedness R, leveling work E, and slope steepness S—adds a penalty term P for maximizing the distance from restricted terrain. res Site geological safety comprehensive coefficient G s The optimization objective of this MFCO model is to consider multiple key factors such as the area of ​​the construction site, terrain adaptability, and construction volume. It is also specifically designed for the site characteristics of bridge construction in mountainous areas, taking into account both geological and construction risk factors. In the optimization process, it can identify and avoid potential geological risk areas, and provide a scientific, reasonable and safe optimization solution for bridge construction sites in mountainous areas.

[0029] Based on the MFCO model, the CFF function is constructed using the lower bound construction method and the penalty function method. Specifically, the lower bound construction method, based on the optimization model, sets reasonable lower limits for the six parameters of the design variables to ensure the feasibility and basic quality of the construction site layout scheme. The penalty term incorporates geological safety risks from insurmountable and restricted terrain into the CFF function. When the model's solution violates the constraints, the penalty term significantly reduces its fitness, guiding the algorithm to explore the feasible solution space. The CFF function expression is as follows: ; ; In the formula, Fitness represents the overall adaptability to the terrain. i R represents the fitness of the i-th site, where w1~w6 are weighting coefficients used to balance the importance of various factors. Their values ​​are set manually based on the actual situation. avg This represents the average ruggedness within the site area, and is numerically equal to the site ruggedness R. i The quotient of E and the site area is used for normalization; dis S represents the initial site leveling work volume, used for normalization; avg This represents the average slope steepness, and its value is the reciprocal of the average slope of all adjacent uniformly distributed points within the site area, used for normalization.

[0030] This fitness function integrates multiple optimization objectives and constraints in the established MFCO model through the next-boundary construction method and the penalty function method, ensuring that the optimization process comprehensively considers all conditions. It also introduces weight coefficients for each optimization objective, allowing adjustments based on specific problems or stage requirements, greatly increasing the model's flexibility and adaptability. Furthermore, to address the comprehensive consideration of optimization objectives of different data types, this fitness function employs a normalization strategy, enabling optimization objectives with different dimensions to be compared and optimized under the same standard.

[0031] S4: Solving the optimization model based on the starfish algorithm. The starfish algorithm is used to iteratively optimize the established MFCO model to obtain the solution with the highest overall site fitness, i.e., the optimal site planning parameters under the current construction conditions. This algorithm solves the problem of traditional solution methods easily getting trapped in local optima, and finally obtains the globally optimal site plan through iterative optimization. S4 is specifically implemented through the following sub-steps: S401: Initialize the starfish population. First, set the basic parameters of the starfish algorithm, including population size N, maximum number of iterations K, and the threshold M for the number of stagnant generations (referring to the number of iterations). Using the Latin hypercube sampling method, N sets of site planning parameters (N being the initial population size, with a value not less than 20) are randomly generated within the defined construction area boundary. Each starfish individual includes a potential construction site layout scheme, containing the center coordinates and geometric parameters of all sites.

[0032] The system also performs feasibility repairs on randomly generated populations, including: if a starfish individual's site falls into a prohibited area of ​​an insurmountable terrain, the entire site is pushed outward along the nearest gradient direction to the boundary of the insurmountable terrain; if the site's geometric dimensions do not meet the size requirements of the construction site, it is scaled up to the required range (e.g., if the site's geometric dimensions are smaller than the lower limit of the site area requirement, it is scaled up to the lower limit value).

[0033] S402: Overall Fitness Calculation and Ranking. The overall fitness of each starfish individual is evaluated using the CFF function (i.e., overall fitness is calculated), and the individuals are ranked according to their overall fitness to identify the current optimal solution and guide subsequent optimization. Individuals with higher overall fitness values ​​represent better construction site layout schemes.

[0034] S403: Simulates the random movement behavior of starfish on rugged terrain, performing a global search to find potential better locations and obtain the current optimal solution. The starfish's movement is simulated using a multi-dimensional search mode, expressed as follows:

[0035] In the formula, This represents the position of the j-th starfish individual after the k-th iteration. This represents the position of the j-th starfish individual before the k-th iteration, where j=1,2,3,…,N and k=1,2,3,…,K; and α1 and α2 are the positions of two other starfish individuals randomly selected from the starfish population in the optimization space. α1 and α2 are random numbers in the range (-1, 1) that are not all zero, used to control the search step size.

[0036] S404: Simulate starfish predation behavior and perform a local search of the area surrounding the current optimal solution. Specifically, select the top O starfish individuals with high overall fitness as dominant individuals, and generate multiple child starfish individuals within a certain range around each dominant individual. Specifically, the center coordinates of the child starfish individuals are randomly offset within a small range from the center coordinates of the dominant individuals, and the length and width are randomly adjusted within a certain proportion range of the size of the dominant individuals (artificially set, it is recommended to take a value of ±20%).

[0037] Then, the multidimensional distance between each dominant individual and non-dominant starfish individuals (including sub-starfish individuals) is calculated (multidimensional means all site design variables included in the starfish individual). The distance expression between different starfish individuals is as follows (taking the optimization of a single construction site as an example):

[0038] When a starfish individual encompasses multiple sites, its multidimensional distance is similarly represented as the Euclidean distance encompassing all design variables across multiple sites. Starfish predation behavior is simulated based on a parallel bidirectional search strategy, and then the distance d between two randomly selected starfish individuals is used. m1 and d m2 Update the position of the starfish child individual using the following update expression:

[0039] In the formula, This indicates the updated position of the j-th starfish individual. This represents the position of the starfish individual before the update, where r1 and r2 are both random numbers between (0,1). m1 and d m2 This is the distance vector between two randomly selected different starfish individuals in the current population and the current starfish individual.

[0040] Calculate the overall fitness of all updated starfish offspring, compare each starfish offspring with its corresponding original dominant offspring, and retain the top O starfish offspring with the highest overall fitness as the new dominant offspring.

[0041] S405: Simulates the regeneration behavior of starfish. For the starfish with the lowest overall fitness, i.e., the last individual ranked in S402, local redevelopment is performed by simulating the regeneration process to avoid the algorithm getting trapped in local optima. The local redevelopment expression is as follows:

[0042] If the location obtained from local redevelopment exceeds the design variable boundary (i.e., the lower limit value), then it is set as the boundary value.

[0043] Furthermore, when an individual in the population has exceeded the individual stagnation generation threshold M without improving its fitness, it is determined to be in a stagnation state, and its regeneration behavior is forcibly triggered, i.e., local redevelopment is carried out.

[0044] S406: Determine if any of the pre-defined iteration termination conditions are met. These conditions include: reaching the maximum number of iterations, the change in the starfish population's location being less than a set threshold, and the optimal individual's overall fitness value converging and stabilizing. If yes, output the location and geometric parameters of the construction site represented by the optimal starfish individual, which represents the optimal construction site layout scheme on rugged terrain. If not, repeat steps S402-S405 to continue iterative optimization.

[0045] S5: Site optimization results display. Outputs the coordinates and geometric parameters of all planar sites represented by the optimal starfish individual and displays them on the BIM model.

[0046] Furthermore, as the bridge construction progresses and the topographic and geological conditions of the construction area change significantly, the BIM model of the construction area can be updated when necessary. Based on the site requirements of the new construction phase, the optimization objectives (which can be added include: distance between sites, transportation distance of components or equipment, etc.) and the MFCO model can be adjusted and optimized again to determine the optimal construction site layout scheme for the new construction phase.

[0047] The present invention will be specifically illustrated below through an embodiment.

[0048] The construction site for a bridge in a mountainous area is located in a river valley with complex terrain, including a trespassable debris flow gully (i.e., uncrossable terrain) and two geologically risky areas (i.e., restricted terrain). The proposed bridge is a prestressed concrete continuous beam bridge connected to a rigid frame box girder bridge, with one section of the proposed bridge closely connected to an adjacent tunnel entrance. According to the construction organization design, the site required for construction at this bridge location includes a precast component storage area, denoted as Site 1, and a concrete mixing plant, denoted as Site 2.

[0049] The parameter initialization steps first involve creating a BIM model based on the construction area information and exporting the terrain data. Then, the terrain grid data and quadrilateral grid data of the construction area are read, and a 3D terrain model is constructed by combining this data with geological survey information. A simplified diagram of the construction area terrain model is shown below. Figure 3 As shown. Due to the dual constraints of the maximum size of transport vehicles at the construction site and the limited layout of construction sites in mountainous areas, the geometric dimensions of both construction sites are set as follows: the length ranges from 50 to 70 m, and the width ranges from 30 to 50 m.

[0050] Then, an MFCO model was constructed. Based on the site characteristics of the example, there is a debris flow gully in the construction area that cannot be traversed. The coordinate range of this gully is used as a constraint condition for non-overlapping areas and as a penalty term P for non-crossable terrain. imp The calculation basis is as follows: There are two geologically unfavorable risk areas within the region. Their coordinate range is used as a geological safety constraint and as a restricted terrain distance penalty term P. res The calculation basis. The CFF function is constructed as follows:

[0051]

[0052] Based on the importance and substitutability of the two sites, the weights W1=0.7 and W2=0.3. Based on the site characteristics, construction plan and geologically unfavorable area characteristics, the fitness component weights w1~w6 are 0.3, 0.2, 0.15, 0.15, 0.1 and 0.1 respectively.

[0053] Based on the constructed optimization model, the starfish algorithm is used to find the optimal site planning scheme. First, the basic parameters of the starfish algorithm are set. In this embodiment, the starfish population size N=20, the maximum number of iterations K=100, and to avoid some individuals getting trapped in local optima, the individual stagnation generation threshold M=12 is set. The population is initialized using the Latin hypercube sampling method to generate an initial starfish population within the construction area. Each starfish individual is represented by a twelve-dimensional vector, corresponding to the spatial location and geometric dimensions of the two construction sites, denoted as [x1,y1,z1,l1,b1,θ1,x2,y2,z2,l2,b2,θ2]. The sampling range is determined based on the boundary values ​​and safety margin of the terrain data (taking the diagonal length of the site) to ensure that all initial solutions are within the feasible region.

[0054] Calculate the overall fitness of each starfish individual and rank them; the higher the value, the better the solution.

[0055] Based on the calculated fitness and ranking information of the population, the population is updated in the order of global exploration, local development, regeneration, and renewal. First, in the global exploration phase, starfish individuals are allowed to move randomly within the solution space. A multi-dimensional search method is used to simulate the process of starfish searching for food in an unknown environment. This allows the algorithm to quickly cover the entire construction area and identify several potential suitable sites. Next, local development is carried out. A parallel bidirectional search strategy is used to simulate predation behavior, identifying two areas with the highest fitness as potential better solutions. Finally, by simulating the gradual gathering of starfish individuals towards these two areas, the behavior of starfish gathering to prey after discovering food is simulated. Through a parallel bidirectional search strategy, starfish individuals make detailed position adjustments within these areas. For the updated starfish population, the convergence of the optimization results is determined based on a pre-set iteration termination condition. If the termination condition is met, the optimal site location and geometric parameters represented by the best starfish individual are output, which is the optimal construction site layout scheme on rugged terrain. The output result is placed in the BIM model rendering as shown below. Figure 4 As shown. If the iteration termination condition is not met, the aforementioned optimization steps are repeated to continue iterative optimization. In this embodiment, the fitness convergence process is as follows: Figure 5 As shown.

[0056] It will be understood by those skilled in the art that the above descriptions are merely preferred examples of the invention and are not intended to limit the invention. Although the invention has been described in detail with reference to the foregoing examples, those skilled in the art can still modify the technical solutions described in the foregoing examples or make equivalent substitutions for some of the technical features. All modifications and equivalent substitutions made within the spirit and principles of the invention should be included within the scope of protection of the invention.

Claims

1. A mountainous restricted terrain bridge construction site planning method, characterized by, The method comprises the following steps: S1: constructing a three-dimensional building information model of a construction area based on terrain and geological data; S2: determining an optimization target and a constraint condition; the optimization target comprises maximizing the size of a construction site in the construction area while minimizing the ruggedness of the construction site, the amount of flattening work, and the steepness of the site slope; the constraint condition comprises the type, quantity, and size requirements of the construction site, and the limited terrain that cannot be crossed and the limited terrain with geological risks; S3: constructing a multi-factor collaborative optimization model by taking the spatial coordinates and geometric parameters of the center of the construction site as design variables, and constructing a comprehensive fitness function based on the optimization target, the constraint condition, and a penalty function; S4: performing iterative optimization on the multi-factor collaborative optimization model by using a sea star algorithm to obtain a solution with the maximum comprehensive fitness, i.e., an optimal construction site layout scheme, which comprises the spatial coordinates and geometric parameters of the center of the construction site; S5: displaying the optimal construction site layout scheme through the building information model.

2. The mountainous confined terrain bridge construction site planning method according to claim 1, characterized in that, The expression of the multi-factor collaborative optimization model is as follows: ； In the formula, X i ={x i ,y i ,z i ,l i ,b i ,θ i} T is the design variable of the ith site, (x i ,y i ,z i ) is the central coordinate of the ith site, the geometric parameters (l i ,b i ,θ i ) are the length, width and direction angle of the ith site, n is the number of sites, W i is the weight of the ith site; A is the area of the site, A min,i ,A max,i are the minimum and maximum values of the ith site area respectively; G s is the comprehensive safety coefficient of the site geology, R is the ruggedness of the site; E is the site grading engineering quantity, S is the steepness of the site slope; l min,i ,l max,i are the minimum and maximum values of the length of the ith site respectively; b min,i ,b max,i are the minimum and maximum values of the width of the ith site respectively; P res is the limited terrain distance penalty term, P imp is the non-crossable terrain penalty term.

3. The mountainous confined terrain bridge construction site planning method according to claim 2, characterized in that, The value of the limited terrain distance penalty term is the average distance from each uniform point at the edge of the site to the nearest geometric center point coordinate of the limited terrain; If the site range overlaps the range of the terrain that cannot be crossed, the value of the terrain that cannot be crossed penalty term is positive infinity; otherwise, the value of the terrain that cannot be crossed penalty term is 0.

4. The mountainous confined terrain bridge construction site planning method according to claim 2, wherein, The value of the site geological safety comprehensive coefficient G s is the average distance from each uniform point at the edge of the site to the nearest geological safety control point coordinate; The value of the ruggedness R of the site is the sum of the absolute values of the height differences of all adjacent uniform points in the site range; The value of the site flattening work amount E is the amount of filling and excavation for flattening the site to the elevation of the site center coordinate; The value of the site slope steepness S is the reciprocal of the slope corresponding to the maximum slope position at the edge of the site.

5. The mountainous confined terrain bridge site planning method according to claim 2, wherein, The expression of the comprehensive fitness function is as follows: ； ； In the formula, Fitness is the comprehensive fitness of the site, fitness i is the fitness of the i th site, w1-w6 are weight coefficients, which are set by humans; R avg represents the average ruggedness within the site range, E dis represents the initial scheme site grading engineering quantity, S avg represents the average slope steepness.

6. The mountainous confined terrain bridge site planning method according to claim 1, wherein, S4 is specifically implemented through the following sub-steps: S401: setting the basic parameters of the sea star algorithm, including the population size, the maximum number of iterations, and the individual stagnation algebra; randomly generating a plurality of sea star individuals according to the basic parameters to complete sea star population initialization; each sea star individual represents a construction site layout scheme; S402: calculating the comprehensive fitness value of each sea star individual by using the comprehensive fitness function, and sorting the sea star individuals from high to low according to the comprehensive fitness value; S403: simulating the movement behavior of the sea star to perform global search and determine a relatively optimal solution region; S404: simulating the predation behavior of the sea star to perform local search on the region around the current relatively optimal solution, and updating the position of the sea star individual by using a parallel bidirectional search strategy; S405: simulating the regeneration behavior of the sea star to perform local redevelopment on the individual with the lowest comprehensive fitness, so as to avoid the algorithm from falling into a local optimal solution; Meanwhile, if the iteration number of an individual in the population exceeds the individual stagnation algebra threshold value but the fitness is not improved, it is determined that the individual is in a stagnation state, and the individual is forced to trigger local redevelopment. S406: repeat the iteration of the above steps until any one of the iteration termination conditions is met, and output the optimal construction site layout scheme; the iteration termination conditions include: reaching the maximum number of iterations, the group position change of the starfish population being less than a set change threshold, and the optimal individual comprehensive fitness value converging stably.

7. The mountain-restricted terrain bridge construction site planning method according to claim 6, characterized by, In the S401, the feasibility repair is performed on the randomly generated population, including: If the construction site range of the starfish individual falls into the non-crossable terrain, the site is extrapolated as a whole to the boundary of the non-crossable terrain along the nearest gradient direction; If the geometric size of the site does not meet the size requirement of the site, the corresponding scaling is performed to the required range.

8. The mountainous confined terrain bridge site planning method according to claim 6, wherein, In the S404, O starfish individuals with the highest comprehensive fitness are selected as dominant individuals, and a plurality of sub-starfish individuals are generated in the range around each dominant individual as the center; The distance between the dominant individual and other starfish individuals in multiple design variable dimensions is calculated, two distances are randomly selected, the starfish predation behavior is simulated based on the parallel bidirectional search strategy, and the position of the starfish individual is updated; the multiple design variables include: spatial coordinates x, y, z, and geometric parameters site length l, site width b, and site orientation angle θ; The comprehensive fitness of all updated sub-starfish individuals is calculated, each sub-starfish individual is compared with the corresponding original dominant individual, and O starfish individuals with the highest comprehensive fitness are retained as new dominant individuals.

9. The mountainous confined terrain bridge site planning method according to claim 1, wherein, The optimization target further includes: site distance, component or equipment transportation distance.

Images