An optimal design method and system of a detachable highway toll island

Abstract

The application provides an optimal design method and system of a detachable highway toll island, and relates to the technical field of the highway toll island, and comprises the following steps: obtaining first information, second information, third information, fourth information and fifth information; obtaining a preliminary function framework according to the first information; performing segmentation processing according to the second information and the preliminary function framework to obtain a module segmentation result; obtaining a module optimization result according to the second information and the module segmentation result; obtaining a universal wheel arrangement scheme according to the fourth information and the module optimization result; performing connection design according to the fifth information and the universal wheel arrangement scheme to obtain a module connection scheme; and obtaining a toll island design scheme according to the module optimization result, the universal wheel arrangement scheme and the module connection scheme. The application adopts parameterized modeling to generate a preliminary function framework of the toll island, combines a component segmentation positioning algorithm and a topological optimization algorithm to design a segmentation and splicing mode of the module, and ensures the safety and stability of the modular structure.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of highway toll island, in particular to an optimization design method and system of a detachable highway toll island. BACKGROUND

[0002] Highway toll islands are indispensable infrastructure in modern traffic management systems and are widely used at highway entrances and exits to achieve standardized and safe management of vehicle traffic. Existing toll islands mostly use fixed cast-in-place base structures. Although this design can meet the basic needs of daily toll collection and vehicle guidance, it still has significant shortcomings in practical application. Due to the limited width of the lane, super-wide vehicles (such as rescue vehicles and emergency material transport vehicles) are often restricted when passing through, and even need to be removed in part to complete special tasks, which not only consumes a lot of manpower and resources, but also seriously affects traffic efficiency. In addition, the fixed toll island has poor crashworthiness and reusability. Once damaged, it often needs to be replaced or repaired in its entirety, further increasing costs and resource waste.

[0003] Currently, some simple improvement techniques are used to try to solve the above problems, such as using steel structures instead of cast-in-place structures and installing universal wheels at the bottom of the toll island to facilitate movement. However, these improvements mostly stay at the primary level of modular disassembly and lack systematic optimization design for mechanical properties, crashworthiness and energy absorption properties, module splicing methods, and dynamic movement schemes, making it difficult to meet the multiple demands of flexibility, functionality, and economy of modern traffic management. SUMMARY

[0004] The purpose of the present application is to provide an optimization design method and system of a detachable highway toll island to improve the above problems. In order to achieve the above purpose, the technical solutions adopted by the present application are as follows:

[0005] In a first aspect, the present application provides an optimization design method of a detachable highway toll island, comprising:

[0006] Obtaining first information, second information, third information, fourth information, and fifth information, the first information including spatial occupancy data and geometric morphology of the toll island, the second information including functional requirements and mechanical requirements, the third information including module splicing and connection requirements, the fourth information including universal wheel arrangement requirements, and the fifth information including crashworthiness performance requirements and economic targets;

[0007] Designing the basic morphology of the toll island according to the first information, controlling the growth direction, number, load capacity, and crashworthiness performance of the module through parameterized modeling, and obtaining a preliminary functional framework;

[0008] According to the second information and the preliminary function framework, the function framework is segmented by using a component segmentation positioning algorithm, a module splicing mode is determined through a topological optimization algorithm, and a connection node and an error tolerance zone are set according to the third information, so as to obtain a module segmentation result;

[0009] According to the second information and the module segmentation result, the mechanical performance of the module is optimized by using a graph neural network, so as to obtain a module optimization result;

[0010] According to the fourth information and the module optimization result, a chain simulation is performed, the number, position and load capacity of the universal wheels are dynamically adjusted, and a universal wheel arrangement scheme is obtained;

[0011] According to the fifth information and the universal wheel arrangement scheme, a connection design is performed, the distribution and size of the pins are optimized through finite element analysis, and a module connection scheme is obtained;

[0012] According to the module optimization result, the universal wheel arrangement scheme and the module connection scheme, a multi-objective optimization process is performed, the overall design of the toll island is weighted and iteratively optimized, and a toll island design scheme is obtained.

[0013] In a second aspect, the application further provides an optimization design system of a detachable highway toll island, comprising:

[0014] An acquisition module is configured to acquire first information, second information, third information, fourth information and fifth information, wherein the first information comprises spatial occupation data and geometric morphology of the toll island, the second information comprises functional requirements and mechanical requirements, the third information comprises module splicing and connection requirements, the fourth information comprises universal wheel arrangement requirements, and the fifth information comprises anti-collision performance requirements and economic targets;

[0015] A modeling module is configured to design a basic morphology of the toll island according to the first information, control the growth direction, number, load capacity and anti-collision performance of the module through parameterized modeling, and obtain a preliminary function framework;

[0016] A segmentation module is configured to segment the function framework by using a component segmentation positioning algorithm according to the second information and the preliminary function framework, determine a module splicing mode through a topological optimization algorithm, set a connection node and an error tolerance zone according to the third information, and obtain a module segmentation result;

[0017] An optimization module is configured to optimize the mechanical performance of the module by using a graph neural network according to the second information and the module segmentation result, and obtain a module optimization result;

[0018] A simulation module is configured to perform a chain simulation according to the fourth information and the module optimization result, dynamically adjust the number, position and load capacity of the universal wheels, and obtain a universal wheel arrangement scheme.

[0019] The design module is used for connection design according to the fifth information and the universal wheel arrangement scheme, and the distribution and size of the pin are optimized through finite element analysis, so as to obtain a module connection scheme.

[0020] The output module is used for multi-objective optimization processing according to the module optimization result, the universal wheel arrangement scheme and the module connection scheme, weight distribution and iterative optimization are performed on the overall design of the toll island, and a toll island design scheme is obtained.

[0021] The beneficial effects of the present application are:

[0022] The present application generates a preliminary functional framework of the toll island by using parameterized modeling and morphological optimization algorithm, combines component segmentation positioning algorithm and topological optimization algorithm to design the fine segmentation and splicing mode of the module, and ensures the safety and stability of the modular structure. Further, the mechanical properties of the module are optimized by using the graph neural network, the material distribution and the connection strength are dynamically adjusted, and the overall impact resistance is improved. In terms of mobile design, the quick switching of the toll island module between the static and mobile states is realized by combining the chain simulation technology and the universal wheel arrangement optimization algorithm, and the applicability and flexibility are significantly improved. In addition, the distribution and size of the pin are optimized through finite element analysis, the impact energy absorption performance between the modules is enhanced, and the modular degree, the crashworthiness, the mobility and the economy of the toll island are comprehensively improved by using the multi-objective optimization algorithm. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0024] Figure 1 A flow chart of an optimal design method of a detachable highway toll island according to an embodiment of the present application is shown.

[0025] Figure 2 A structure diagram of an optimal design system of a detachable highway toll island according to an embodiment of the present application is shown.

[0026] Figure 3 A structure diagram of an optimal design device of a detachable highway toll island according to an embodiment of the present application is shown.

[0027] Label in the figure: 800, an optimal design device of a detachable highway toll island; 801, a processor; 802, a memory; 803, a multimedia assembly; 804, an I / O interface; 805, a communication assembly; 901, an acquisition module; 902, a modeling module; 903, a segmentation module; 904, an optimization module; 905, a simulation module; 906, a design module; 907, an output module. DETAILED DESCRIPTION

[0028] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some but not all of the embodiments of the present application. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art without creative work based on the embodiments in the present application belong to the scope of protection of the present application.

[0029] It should be noted that: similar labels and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second" and the like are only used to distinguish description, and cannot be understood as indicating or implying relative importance.

[0030] Embodiment 1

[0031] The embodiment provides an optimal design method of a detachable highway toll island.

[0032] Referring to Figure 1 , the method includes steps S100 to S700.

[0033] Step S100, acquiring first information, second information, third information, fourth information and fifth information, the first information including spatial occupancy data and geometric morphology of the toll island, the second information including functional requirements and mechanical requirements, the third information including module splicing and connection requirements, the fourth information including universal wheel arrangement requirements, and the fifth information including crashworthiness performance requirements and economic targets;

[0034] It can be understood that the first information explicitly charges the physical space and shape characteristics of the island, which not only determines the segmentation scheme and growth direction of the module, but also needs to adapt to the specific size requirements of the toll island in different scenes, such as the width limit of the highway entrance or the complex terrain; The second information ensures that the design of the toll island can meet the vehicle guidance, protection function, and withstand various dynamic and static loads, including the impact force of vehicle impact and the static pressure of self-weight; The third information provides constraint conditions for the modular disassembly and assembly design, especially the strength of the connection node, the splicing form and the error tolerance range, which are crucial for the stability and reusability of the modular structure; The fourth information combines the stress distribution at the bottom of the module to define the position, number and load-bearing capacity of the universal wheel, ensuring the flexible switching of the module in the moving and stationary states; The fifth information directly affects the material selection, energy absorption structure optimization, and cost control of the overall scheme of the module design.

[0035] Step S200, according to the first information, the basic form of the toll island is designed, the growth direction, number, load capacity and crashworthiness of the module are controlled through parameterized modeling, and a preliminary functional framework is obtained;

[0036] It should be noted that this step converts the toll island from abstract functional requirements to specific structural design basis. Taking the highway toll island as an example, its space is long and narrow and needs to withstand complex traffic dynamic loads. Parameterized modeling not only quickly generates a feasible module form, but also can flexibly adjust the design according to different space and functional requirements. The unique effect of this method is that it changes the design from static to dynamic through algorithm-driven modeling, providing a highly flexible and accurate basic framework for the subsequent optimization steps.

[0037] Step S300, according to the second information and the preliminary functional framework, the functional framework is segmented and processed by using component segmentation positioning algorithm, the module splicing mode is determined by using topology optimization algorithm, and the connection node and error tolerance zone are set according to the third information, and the module segmentation result is obtained;

[0038] It can be understood that this step improves the disassembly and reusability of the toll island through modular segmentation, and the combination of topology optimization and graph theory optimization ensures the strength and economy of the module splicing.

[0039] Step S400, according to the second information and the module segmentation result, the mechanical properties of the module are optimized by using graph neural network, and the module optimization result is obtained;

[0040] It can be understood that, through the optimization of the graph neural network, the material distribution and connection strength inside the module are comprehensively improved, and the anti-deformation ability of the module under dynamic load and complex stress conditions is significantly enhanced. In addition, the graph neural network optimization can also reduce unnecessary material use while ensuring mechanical performance, thereby balancing economy and engineering applicability.

[0041] Step S500, according to the fourth information and the module optimization result, carry out chain simulation, through dynamically adjusting the number, position and bearing capacity of the universal wheel, obtain the universal wheel arrangement scheme;

[0042] It should be noted that the optimized universal wheel arrangement scheme can effectively disperse the stress at the bottom of the module, avoid local overload, and ensure quick switching between dynamic movement and static fixation, providing strong support for multi-scene application of the toll island.

[0043] Step S600, according to the fifth information and the universal wheel arrangement scheme, carry out connection design, through finite element analysis to optimize the distribution and size of the pin, obtain the module connection scheme;

[0044] It can be understood that the optimized module connection scheme can ensure structural safety and durability under the conditions of multiple disassembly and assembly of the highway toll island and long-term use, while significantly reducing the maintenance and replacement cost of the connection node.

[0045] Step S700, according to the module optimization result, the universal wheel arrangement scheme and the module connection scheme, carry out multi-objective optimization processing, through weight distribution and iterative optimization of the overall design of the toll island, obtain the toll island design scheme.

[0046] It should be noted that, through comprehensive consideration of the optimization results of the module, the universal wheel and the connection, this step organically combines local optimization with overall design, ensuring that the toll island performs best in various complex scenarios (such as high traffic flow impact, frequent disassembly and assembly requirements, cost constraints, etc.) and economy.

[0047] Further, step S200 includes step S210 to step S250.

[0048] Step S210, according to the first information, initialize the shape of the toll island space data, divide the target space into multiple functional areas through Voronoi segmentation algorithm, combine the geometric shape of the toll island to limit boundary conditions, and obtain the initial functional area division result;

[0049] Specifically, the first step in morphological initialization is to construct a mathematical model of the toll island space, representing the target area as a geometric domain in two-dimensional or three-dimensional space. For example, in a toll island scenario, lane width, entrance / exit locations, and surrounding geometric features are used as input conditions to generate a basic spatial framework describing the toll island. Next, the target space is divided into multiple functional regions using the Voronoi segmentation algorithm. The Voronoi algorithm uses several initial points (called generation points) as a basis to divide the space into multiple polygonal regions, where the distance from all points within each region to their corresponding generation point is less than the distance to other generation points. The selection of generation points is dynamically adjusted based on the functional requirements of the toll island (such as the number of lanes and traffic flow distribution) and modular design goals to ensure that each functional region has a reasonable area and shape. Furthermore, to ensure that the segmentation results match the geometric shape of the toll island, boundary conditions need to be applied to constrain the Voronoi segmentation results. For example, in the boundary region of the toll island, irregular geometric shapes (such as arcs and hypotenuses) need to be used as constraints to ensure that the boundaries of the segmented regions are consistent with the actual shape of the toll island. Meanwhile, in functional areas near toll booths or entrances / exits, the segmentation results need to consider higher traffic density, dynamically adjusting the location of the generation points to optimize the area division. This step transforms abstract spatial data into specific functional area distributions, providing a geometric basis for optimizing the subsequent module growth direction.

[0050] Step S220: Based on the initial functional area division results, optimize the growth direction. Use gradient descent and morphological optimization algorithms to determine the growth path of the module within the division area. With minimizing functional area overlap and gap area as the objective function, adjust the direction vector of the module so that the growth direction of the module matches the force direction and functional requirements of the toll island, and obtain the module growth direction result.

[0051] The objective function formula is as follows:

[0052]

[0053] in, Indicates the direction of module growth The overall evaluation index; i represents the module number; n represents the total number of modules; j represents the number of overlapping or gap regions; m represents the total number of overlapping and gap regions; k represents the constraint condition number; p represents the total number of constraints; This represents the growth direction vector of the i-th module; A represents the principal stress direction vector of the i-th module in the stress analysis; j This represents the area of ​​the j-th overlapping or gap region. A function representing the degree of deviation between the kth constraint condition and the boundary limit condition, used to limit the matching degree of the module growth direction and the boundary shape; α, β and γ represent the weight coefficients respectively.

[0054] Objective function The comprehensive evaluation module growth direction is divided into three parts:

[0055] The first term is Used to measure the matching degree of the module growth direction And the principal stress direction By minimizing the difference between the two, ensure that the module growth direction is consistent with the actual stress state, improve the mechanical properties of the module. Suitable for processing highway toll island dynamic load and complex shape of the multi-stress distribution.

[0056] The second term is Used to minimize the area A j Of the overlapping and void regions. This optimization function area division compactness, avoid waste space, while ensuring that the module boundary does not interfere with each other.

[0057] The third term is Used to limit the deviation degree of the module growth direction and the boundary shape of the toll island, by restricting the growth direction of the module near the irregular boundary, ensure the global consistency and geometric constraint adaptability of the segmentation result.

[0058] The gradient descent method update formula is:

[0059]

[0060] Where, Represents the growth direction vector of the ith module in the tth iteration; Represents the growth direction vector of the ith module after the t+1th iteration; η represents the learning rate, used to control the update amplitude of each iteration; Represents the gradient of the objective function to the growth direction vector Each round of iteration, adjust the growth direction vector of the module, so that it gradually approaches the target direction, while optimizing the global overlap and void problem.

[0061] Step S230, according to the module growth direction result, optimize the number of modules, determine the number of modules based on the genetic algorithm module division and configuration model, take the module load capacity and space utilization efficiency as the optimization target, through the crossover, mutation and selection operation in genetic algorithm, get the module number configuration result;

[0062] It can be understood that, through the results of the module growth direction, the preliminary division range of the functional area and the stress distribution characteristics can be determined. On this basis, the genetic algorithm is optimized, a plurality of module configuration schemes are randomly generated as an initial population, and the performance of each scheme in load capacity balance and space utilization is evaluated by using a fitness function. The high fitness scheme is reserved by selection operation, the new scheme is generated by combining the characteristics of the excellent scheme by crossover operation, and the number or distribution of the modules is randomly adjusted by mutation operation to increase the diversity of the population, and finally the optimal configuration scheme is iteratively converged. This process can dynamically adapt to complex geometric shapes and stress distribution requirements, such as increasing the number of modules in high stress areas to improve the bearing capacity, and reducing the number of modules in low stress areas to reduce material cost and splicing complexity. The optimized number of modules can achieve load balance, reduce void area, improve space utilization efficiency, and provide a more reasonable segmentation basis for module splicing and shape design.

[0063] Step S240, according to the module number configuration result, simulates by adjusting the edge shape and size of the module to maximize the impact energy absorption and minimize the structure deformation as the optimization target, simulates the stress distribution and energy absorption performance under different vehicle impacts, and obtains a preliminary functional framework.

[0064] Specifically, this step first combines the module number configuration result, takes the edge shape and size of each module as adjustable variables, and uses finite element analysis to dynamically simulate the stress distribution of the module under impact load. Impact simulation not only needs to consider static load, but also needs to combine dynamic load (such as vehicle impact) to comprehensively evaluate the structure of the module. By adjusting the edge shape of the module, the contact surface distribution between the module and the impact area is optimized, the stress concentration area is reduced, and the energy absorption efficiency of the module is improved; and the adjustment of the size of the module aims to ensure that the geometric shape of each module meets the structural strength requirement and does not cause unnecessary material waste.

[0065] In the simulation process, the edge shape and size are adjusted step by step, the optimization target is to maximize the impact energy absorption capacity of the module, and at the same time minimize the deformation degree under load. For example, the edge shape of the module near the edge of the toll island can be designed as a nested style with a buffer structure, so as to better disperse the impact force; and in the internal module area, the shape can be simplified to reduce material use. This method can ensure that each module has excellent impact resistance performance and meets the requirements of economy and manufacturing feasibility.

[0066] Further, step S300 includes step S310 to step S340.

[0067] Step S310, according to the second information and the preliminary function framework, the module partition region is demarcated, through the clustering analysis to the geometric region of the function framework, the stress demand is combined as the clustering feature, the minimum function unit constraint condition of the module is set, and the preliminary module partition region is obtained;

[0068] It should be noted that the clustering analysis uses the geometric shape, stress distribution and functional requirements of the function framework as multi-dimensional data as feature input, and the stress uniformity and functional consistency of each module as the target, the framework region is grouped and divided. In this process, the module partition needs to meet the specific minimum function unit constraint condition, that is, each module must independently bear its functional requirements, such as bearing traffic load or energy absorption anti-collision. Combined with the mechanical demand, the clustering features mainly include stress concentration area, boundary geometric feature and logical relationship of functional partition. For example, in the stress concentration area, the module partition needs to ensure that the module size is moderate to avoid excessive stress concentration; in the area with single function (such as the central passing area), larger modules can be divided to reduce the number of splicing points. This step divides the complex function framework into reasonable preliminary module regions through the clustering algorithm, and ensures that each module retains its geometric and mechanical properties in the partition process, avoiding performance loss or increasing splicing complexity due to improper partition.

[0069] Step S320, according to the preliminary module partition region, the boundary shape of the module is optimized, the smoothness and structural strength of the module boundary are adjusted iteratively by dynamically adjusting the edge shape of the module, the optimization result of the module boundary shape is obtained, with the minimum complexity of the module boundary and the maximum stability of the mechanics as the target;

[0070] Firstly, according to the boundary shape characteristics of the preliminary module partition region, combined with the functional requirements and stress analysis of the toll island, the key feature points in the boundary shape and the parts that need to be optimized are identified. For example, in the area with high boundary complexity or sharp corners, these features may cause stress concentration and affect the overall strength of the module. Then, the gradient descent method is used to iteratively optimize the boundary, by gradually adjusting the position and curvature of the boundary points, the module boundary is made smoother and unnecessary complex shape is reduced. During the boundary optimization process, the smoothness of the boundary shape is defined as part of the objective function to limit the mutation of the boundary curvature, while ensuring the geometric continuity of the boundary shape and the manufacturability. At the same time, in order to improve the mechanical stability of the module, the boundary shape is combined with the mechanical distribution in the module during the optimization process. For example, in the high stress area, the boundary shape is optimized to disperse the load and prevent local stress concentration from damaging the overall performance of the module; while in the low stress area, the boundary can be simplified to reduce the amount of material used and the complexity of splicing. Through the gradient descent method, the partial derivative of the objective function is iteratively updated, and the optimal boundary shape that meets the complexity minimization and stability maximization is gradually converged.

[0071] The objective function formula is:

[0072] F(B) = F smooth (B) + F stress (B) + F thickness (B) ;

[0073]

[0074] wherein F(B) represents the objective function of boundary optimization; B represents the boundary shape vector set of the module; F smooth (B) represents the boundary smoothness optimization formula; F stress (B) represents the stress distribution uniformity optimization formula; F thickness (B) represents the material thickness smoothness optimization formula; μ represents the weight of boundary smoothness; B i represents the shape function of the i-th module boundary; represents the boundary of the i-th module; u represents the parameter coordinate on the boundary; v represents the weight of stress distribution uniformity optimization; l represents the region number on the module boundary; r represents the total number of boundary regions divided in all modules; σ l represents the stress value on the l-th module boundary; represents the boundary of the l-th module; represents the total length of the l-th module boundary; w represents the number of material thickness variation regions on the module boundary; s represents the total number of regions involving material thickness variation in all module boundaries; ξ represents the weight of material thickness smoothness optimization; z represents the boundary normal coordinate; H w represents the material thickness distribution function of the w-th module boundary; represents the boundary of the w-th module.

[0075] In step S330, the module splicing mode is designed according to the module boundary shape optimization result and the second information, the best connection scheme of the module is determined through a topology optimization algorithm, the number and distribution of the connection points are adjusted to minimize the stress concentration at the splicing position and the module splicing volume, and the module splicing mode result is obtained.

[0076] Specifically, the design of module splicing needs to consider two aspects: first, in combination with the boundary shape optimization result of the module, identify the areas suitable for connection, such as areas with smoother boundaries and uniform stress, which are more suitable as the basis for the distribution of splicing points; second, according to the mechanical requirements in the second information, ensure that the splicing points can effectively transfer the load and avoid failure caused by local stress concentration. On this basis, the number and position of splicing points are dynamically adjusted through a topology optimization algorithm. Topology optimization takes minimizing the stress concentration and module splicing volume at the splicing position as the objective function, and adjusts the layout of the connection points through repeated iteration to find the best solution. For example, in the module boundary area with greater stress, the density of splicing points can be increased to disperse the load; while in the low stress area, the splicing points can be reduced to reduce material use and splicing cost.

[0077] In this process, the core of the topology optimization algorithm is that it can realize the optimization of the global layout of the splicing points based on the actual geometric shape and mechanical distribution of the module, and dynamically adapt to the changes in the splicing area in combination with the complexity of the boundary shape. This method can minimize the splicing volume and material waste while ensuring the structural strength.

[0078] Step S340, according to the module splicing mode result and the third information, the connection nodes and error tolerance area of the module are designed, the global distribution and redundancy of the connection nodes are determined through graph theory algorithm, and the module segmentation result is obtained.

[0079] It should be noted that this step first extracts the information of the connection area between the modules according to the module splicing mode result, including the initial distribution of the splicing points, the boundary shape characteristics and the stress transmission path between the modules. In combination with the requirements of the third information on the strength, number and error tolerance of the connection nodes, the basic design constraints of the connection nodes are determined. The key of error tolerance area design is to ensure that the connection strength and stability can be guaranteed even if there is certain processing error or installation deviation in the assembly process. Next, the global distribution of the connection nodes is optimized through graph theory algorithm. The module connection is modeled as a graph structure, where the nodes represent the connection points of the module and the edges represent the splicing relationship and mechanical transmission path between the modules. By analyzing the characteristics of the graph, the distribution and redundancy of the connection nodes are determined, such as the minimum spanning tree, path optimality, etc. Redundancy optimization is to ensure that the module can still maintain the overall structural integrity in the case of failure of some connection nodes.

[0080] Further, step S400 includes step S410 to step S440.

[0081] Step S410, according to the second information and the module segmentation result, the mechanical properties of the module are preliminarily modeled, the stress condition of each module is simulated and analyzed by finite element analysis, the stress-strain properties of the module are mapped to node attributes, the connection strength and the force transmission direction are mapped to edge attributes, and an initial mechanical graph structure model is generated;

[0082] Specifically, first, the stress condition of each module is simulated in detail by using finite element analysis. By inputting the module segmentation result into the finite element software, the stress-strain distribution of the module under static and dynamic loads (such as traffic impact force and dead load) is simulated, so as to capture the mechanical properties of the module in the actual application scenario. In the finite element analysis, the key stress area and structural deformation characteristics of the module are accurately calculated, and specific data are provided for subsequent modeling. Then, the results of the finite element analysis are mapped to the initial mechanical graph structure model. The node attributes of the model represent the local mechanical properties of the module (such as stress, strain, compressive strength, etc.), which reflect the response capability of the module under different stress states; the edge attributes represent the connection characteristics and force transmission path between modules, including connection strength, directionality and force transmission efficiency. Through the abstraction of nodes and edges, the complex mechanical behavior of the module is converted into a graph structure that is easy to analyze and optimize. This step abstracts the mechanical behavior of the module by using the graph structure, so that the mechanical correlation between modules can be intuitively presented in a global perspective.

[0083] Step S420, according to the initial mechanical graph structure model, a structure optimization process is performed, the mechanical graph is partitioned by using a spectral clustering algorithm, the mechanical graph is divided into a plurality of subgraphs, and the node and edge attributes in the subgraphs are redefined to obtain a mechanical subgraph structure;

[0084] It should be noted that the initial mechanical graph structure model is composed of nodes (local mechanical properties of the module, such as stress and strain) and edges (force transmission properties of the module connection area, such as connection strength and transmission efficiency). Since the mechanical behavior of the module may have unnecessary connection redundancy or uneven distribution of force transmission paths, which directly affects the performance and cost of the overall structure, it is necessary to optimize the graph structure. The spectral clustering algorithm is applied in this process, and its core idea is to find the similarity measure between nodes by eigenvalue decomposition of the Laplacian matrix of the mechanical graph, and accordingly divide the graph into several subgraphs. In actual processing, the spectral clustering algorithm first calculates the association matrix of the nodes and edges of the mechanical graph, and combines the edge weight (such as force transmission strength) and node attribute (such as local stress concentration degree) to construct the Laplacian matrix of the graph. By performing eigenvalue decomposition on the Laplacian matrix, the closely related node sets in the graph are identified, and these sets are divided into independent subgraphs. The divided subgraphs have the following characteristics: the nodes and edges within each subgraph have higher correlation (such as similar stress distribution and force transmission direction). The connection between subgraphs is reduced, optimizing the edge weight distribution and avoiding unnecessary mechanical transmission paths. After the subgraph is generated, the node and edge attributes within each subgraph are redefined. For example, the edge weight within the subgraph can be adjusted to the average force transmission strength, and the node attribute is aggregated to the overall stress distribution characteristics of the region. In this way, each subgraph not only independently describes the local mechanical properties of the module, but also reduces the complexity of the global graph. The optimized mechanical subgraph structure can effectively reduce the redundancy in module connection design, improve force transmission efficiency and structural stability

[0085] Step S430, according to the mechanical subgraph structure, the node material distribution and edge connection strength of the module are predicted and optimized through the graph neural network, taking the minimization of the module stress uniformity and edge connection strength under dynamic load as the target, using the message passing network to iteratively update the attributes of the nodes and edges, to obtain the material distribution-connection strength optimization result;

[0086] It can be understood that the core of the graph neural network in this step is to use a message passing network (MPN) to gradually update the material distribution and connection strength attributes through the interaction of nodes with neighbor nodes and edges. The node update optimizes the local material properties to adapt to dynamic loads, and the stress is more uniform; the edge update adjusts the force transmission path and strength to avoid redundancy. In multiple rounds of iteration, the graph neural network gradually propagates global and local information, making the mechanical graph converge to an optimal state. Through this optimization, the stress uniformity within the module is significantly improved, reducing the risk of local stress concentration; the connection material consumption is reduced, reducing the cost; the optimization result can flexibly adapt to the dynamic load demand of complex traffic scenarios, providing higher safety, stability and economy for the toll island module design. This optimization method based on graph neural network fully reflects the global and local balance of the mechanical properties of the toll island module, and is suitable for the actual deployment of complex modular structures.

[0087] Step S440, according to the material distribution-connection strength optimization result, the overall mechanical performance of the module is comprehensively optimized, and the global distribution of the module form and material is adjusted through a multi-objective genetic algorithm to generate a module optimization result.

[0088] It should be noted that the multi-objective genetic algorithm integrates various performance requirements such as dynamic load, static load and structural stability into the optimization framework through global search of the form and material distribution. The optimization objectives include maximizing the impact resistance of the module to make it have higher stability in complex stress environment; minimizing the unevenness of the module material distribution to ensure that the material distribution is more in line with the stress demand; and reducing the amount of material used to control the cost.

[0089] The optimization process of the genetic algorithm starts with randomly generating multiple sets of module form and material distribution schemes as the initial population, each scheme consisting of geometric parameters (such as size, curvature) and material parameters (such as density, elastic modulus) of the module. The performance of each scheme in terms of impact resistance, stress uniformity and material utilization is evaluated by the fitness function. Individuals with higher fitness are selected for crossover operation to combine the advantages of two excellent schemes to generate new candidate schemes; at the same time, mutation operation randomly adjusts the parameters of the individuals to introduce diversity and avoid falling into local optimum. After multiple iterations, the algorithm gradually converges to the global optimal solution, and finally generates the optimization result of the module.

[0090] Further, step S500 includes steps S510 to S540.

[0091] Step S510, according to the fourth information and the module optimization result, the stress distribution at the bottom of the module is analyzed, the preliminary result of stress analysis is obtained by identifying the high stress area and uniform bearing area at the bottom, combining the edge and center symmetry constraint conditions to delimit the preliminary area of the universal wheel arrangement, and the preliminary result of stress analysis is obtained.

[0092] It can be understood that, based on the module optimization results, the stress distribution at the bottom of the module is calculated by using the finite element analysis method to identify the high stress area and the uniform load area at the bottom. The high stress area usually appears at the position where the load is concentrated or the structure is mutated, and the uniform load area reflects the relatively stable stress state of the module. Through this classification, the mechanical properties of the module bottom can be effectively divided. On the basis of stress distribution analysis, combined with the symmetry constraint conditions of the module edge and center, the arrangement area of the universal wheel is further optimized. The symmetry constraint aims to ensure the balance of the universal wheel arrangement in geometry and mechanics, and avoid tilting or load imbalance of the module during movement due to uneven arrangement. For example, in the edge area of the module, the universal wheel can be arranged near the high stress point to disperse the load; in the center area of the module, the symmetry of the arrangement needs to be ensured to maintain the overall stability.

[0093] Step S520, according to the preliminary results of stress analysis, the arrangement points are dynamically distributed in the candidate area by using the iterative arrangement algorithm, and the arrangement position of the universal wheel is optimized in combination with the distribution of the module connection nodes to obtain a preliminary universal wheel arrangement scheme;

[0094] It can be understood that the candidate area is delimited by the identified high stress area and uniform load area. The iterative arrangement algorithm takes the candidate area as the search space, and dynamically adjusts the position of the arrangement points to preferentially distribute the universal wheels in the high stress area to disperse the load and reduce stress concentration, while ensuring that the arrangement points in the low stress area do not affect the overall mechanical properties. In the process of arrangement, the distribution of the module connection nodes is taken into account in the optimization. The connection nodes are usually the key parts of the module stress, and the arrangement of the universal wheels in these areas should avoid interfering with the splicing, while providing additional support to improve the stability of the connection nodes. The iterative arrangement algorithm dynamically optimizes the arrangement scheme by gradually adjusting the arrangement points in the candidate area. In each iteration, the position and number of arrangement points are re-evaluated according to the objective function of stress dispersion and load balance. The arrangement points that contribute most to the structural stability are preferentially selected, and redundant points are eliminated to improve the arrangement efficiency.

[0095] Step S530, according to the preliminary universal wheel arrangement scheme, the number and load capacity of the universal wheels are discretely optimized, the load proportion is allocated by using the integer linear programming algorithm, and a universal wheel load allocation scheme is generated;

[0096] Specifically, the local stress and load requirements at each caster location are obtained through a preliminary caster arrangement scheme. These locations are treated as discrete variables, each corresponding to a caster and its possible load-bearing capacity options. The load-bearing capacity is represented by discrete values, such as standard load ranges (e.g., 200kg, 400kg, 600kg, etc.). The objective function is optimized to maximize the stress uniformity at the bottom of the module while minimizing the total number of casters and the risk of overload, ensuring that the load distribution of each caster meets both local requirements and achieves global balance. During the optimization process, an integer linear programming algorithm dynamically adjusts the number and distribution of casters by selecting the optimal load-bearing capacity combination, making the overall load more uniform and reducing local stress concentration and vibration risks during dynamic movement.

[0097] Step S540: Simulate the kinematic performance of the caster wheel connection according to the caster wheel load distribution scheme. Simulate the movement path and load uniformity of the Lagrange dynamics modeling module and optimize the connection structure of the caster wheels to obtain the final caster wheel arrangement scheme.

[0098] Understandably, Lagrangian dynamics modeling is used to describe the motion behavior of the module and its caster system. This model comprehensively considers the module's inertia, the supporting force of the casters, the forces at connection points, and external forces acting on the module along its movement path (such as steering and impact forces). By introducing the Lagrangian function, the system's kinetic and potential energy are combined within a unified mathematical framework, establishing a set of overall dynamic equations for the module. During dynamic simulation, the model focuses on analyzing the smoothness of the movement path and the uniformity of the load. The smoothness of the movement path indicates whether the casters can maintain a stable trajectory under different terrain and load conditions, avoiding module tilting or offset due to uneven force distribution. Load uniformity indicates whether the instantaneous load changes of each caster during dynamic movement are uniform, especially whether the casters in high-load areas can effectively distribute the force. During the simulation, the design of the caster connection structure is also included in the optimization objectives. By adjusting the distribution of connection points, the flexibility of the connection devices, and the connection angles, it is ensured that the casters can quickly adapt to changes in force at the bottom of the module and smoothly transition between different directions and load distributions during movement. The optimization results are obtained through step-by-step iterations to ensure that each adjusted solution improves both the smoothness of movement and the uniformity of load.

[0099] Further, step S600 includes steps S610 to S640.

[0100] Step S610: Analyze the fifth information and the universal wheel arrangement scheme, extract the main force transmission paths between modules based on the stress path analysis algorithm, and determine the initial distribution position of the pins in combination with the boundary conditions and mechanical requirements of the modules to obtain the preliminary pin layout scheme.

[0101] It can be understood that the stress path analysis captures the main paths of force transmission from one module to another by identifying the key stress chains in the module structure, which are usually concentrated in the splicing areas with large stress or the module boundary conditions with strong constraints. This process can clearly identify the areas that need to focus on the distribution of the pins, such as high stress concentration points under dynamic load and splicing areas with frequent force transmission. Under the constraints of the module boundary conditions, the pins need to be distributed in the areas that can maximize the dispersion of stress, while avoiding interference with the geometric characteristics of the module edge or reducing the connection strength of the module. For example, at the module splicing, the pins should be arranged at the stress balance point to reduce the connection failure caused by excessive local stress. Inside the module, the pin layout needs to match the mechanical requirements to optimize the load transmission efficiency between modules.

[0102] Step S620, according to the preliminary scheme of pin distribution, the size and shape of the pin are optimized, a plurality of groups of pin design parameter combinations are dynamically generated, and the simulation results of the performance of the pin under different stress states are combined to screen the pin size optimization result;

[0103] Through this optimization process, the pin design can more accurately adapt to the dynamic mechanical requirements of the toll island module, providing key protection for the efficiency and durability of the overall modular design.

[0104] Step S630, according to the pin size optimization result, the strength, energy absorption and distribution density of the pin are iteratively optimized through a multi-objective particle swarm optimization algorithm to obtain a pin distribution optimization result;

[0105] It can be understood that the multi-objective particle swarm optimization algorithm dynamically searches and iterates the pin distribution scheme by constructing a fitness function. Each particle represents a pin distribution scheme, including specific position, strength and energy absorption parameter distribution information. The algorithm simulates the movement of particles in the search space, combines the multi-objective constraints of strength, energy absorption and density, and continuously adjusts the position and velocity of the particles. In each iteration, the fitness of the particles is evaluated according to the weight of the objective function, the best scheme is retained, and other particles are guided to approach the optimal solution through global and local optimal information.

[0106] Step S640, according to the pin distribution optimization result, the overall connection performance of the module is verified, the dynamic stress performance of the pin is simulated through finite element analysis, dynamic impact and static load are taken as test conditions, the overall impact resistance and long-term stability of the module connection are evaluated, and the final module connection scheme is obtained.

[0107] It should be noted that in the dynamic impact working condition, the high dynamic load scene such as vehicle impact and sudden vibration is simulated, and through the analysis of the stress change, deformation and energy absorption performance of the pin under the impact force, it is verified whether the anti-impact ability meets the requirements. In the static load working condition, the self-weight and long-term stress condition of the module in daily use are simulated, and the stress distribution and fatigue life of the pin are evaluated to ensure the long-term stability of the connection.

[0108] Further, step S700 includes step S710 to step S740.

[0109] Step S710, according to the module optimization result, the universal wheel arrangement scheme and the module connection scheme, the overall structure performance is modeled, the performance index system of the toll island design is established through the weighted objective function model, the impact resistance performance, the mobility, the durability and the economy are taken as the target and different weight coefficients are allocated, the overall optimization objective function of the toll island is generated, and the comprehensive performance model is obtained;

[0110] It can be understood that the quantification of the impact resistance performance ensures that the modular design of the toll island can maintain stability under high dynamic load scenarios; the mobility evaluation improves the applicability of the toll island in disassembly and complex terrain; the durability analysis prolongs the service life of the module and the connecting piece; the economic optimization reduces the overall cost. Through the model, the toll island design not only meets the diversified application requirements, but also realizes a high degree of coordination among performance, cost and reliability, providing a solid foundation for the global optimization of the overall design.

[0111] Step S720, according to the comprehensive performance model, parameter optimization is carried out, the module size, the universal wheel position and the connection node distribution are adjusted iteratively through the nonlinear programming method, and the initial structure optimization result is obtained;

[0112] It should be noted that the optimization of the module size reduces the material waste while maintaining the mechanical performance, the adjustment of the universal wheel position improves the dynamic mobility performance and the load distribution uniformity of the module, and the optimization of the connection node improves the strength and impact resistance of the splicing place. The overall optimization result provides an accurate basic configuration for the subsequent global design.

[0113] Step S730, according to the initial structure optimization result, the module splicing sequence and arrangement mode are optimized through the simulated annealing algorithm to minimize the connection stress and installation angle deviation, and the target value change of the load distribution evaluation is combined to obtain the global arrangement optimization result of the module;

[0114] It can be understood that the simulated annealing algorithm adjusts the splicing order and arrangement position of the module by random disturbance, and gradually reduces the probability of large adjustment according to the decrease of the annealing temperature. The target function value is calculated after each adjustment, and if the target value is improved, the new scheme is accepted; if the target value is deteriorated, it is decided whether to accept according to the annealing temperature to jump out of the local optimum. The finally generated module global arrangement optimization result provides a more efficient and reliable splicing scheme for the toll island modular design, ensures the long-term stability of the module in the complex mechanics and dynamic environment, and at the same time simplifies the actual construction process and reduces the manufacturing and installation cost.

[0115] In step S740, according to the module global arrangement optimization result, the overall design scheme of the toll island is comprehensively optimized by a multi-objective genetic algorithm, taking the impact resistance performance, movement performance and cost control as the target, taking the module material distribution, structure form and movement configuration as the optimization variable, generating the optimization solution through selection, crossover and mutation operations, and combining the fitness function to evaluate the scheme performance to obtain the final overall design scheme of the toll island.

[0116] Embodiment 2:

[0117] As shown in Figure 2 The embodiment provides an optimization design system of a detachable highway toll island, and the system comprises:

[0118] The acquisition module 901 is configured to acquire first information, second information, third information, fourth information and fifth information, the first information comprises space occupation data and geometric form of the toll island, the second information comprises functional requirements and mechanical requirements, the third information comprises module splicing and connection requirements, the fourth information comprises arrangement requirements of universal wheels, and the fifth information comprises impact resistance performance requirements and economic targets;

[0119] The modeling module 902 is configured to design a basic form of the toll island according to the first information, control growth directions, quantities, load capacities and impact resistance performances of the modules through parameterized modeling, and obtain a preliminary functional framework;

[0120] The segmentation module 903 is configured to segment and process the functional framework by using a component segmentation positioning algorithm according to the second information and the preliminary functional framework, determine a module splicing mode by using a topology optimization algorithm, set connection nodes and error tolerance zones according to the third information, and obtain a module segmentation result;

[0121] The optimization module 904 is configured to optimize mechanical performances of the modules by using a graph neural network according to the second information and the module segmentation result, and obtain a module optimization result;

[0122] The simulation module 905 is configured to perform linkage simulation according to the fourth information and the module optimization result, dynamically adjust quantities, positions and load capacities of the universal wheels, and obtain a universal wheel arrangement scheme.

[0123] The design module 906 is configured to perform connection design according to the fifth information and the universal wheel arrangement scheme, and to optimize the distribution and size of the pins through finite element analysis, so as to obtain a module connection scheme.

[0124] The output module 907 is configured to perform multi-objective optimization processing according to the module optimization result, the universal wheel arrangement scheme and the module connection scheme, to perform weight distribution and iterative optimization on the overall design of the toll island, and to obtain a toll island design scheme.

[0125] In one specific embodiment disclosed in the present application, the modeling module 902 comprises:

[0126] The first division unit is configured to perform morphology initialization on the spatial data of the toll island according to the first information, to divide the target space into a plurality of functional areas through a Voronoi division algorithm, to combine the boundary conditions limited by the geometric morphology of the toll island, and to obtain an initial functional area division result.

[0127] The first optimization unit is configured to perform growth direction optimization processing according to the initial functional area division result, to determine the growth path of the module in the division area through a gradient descent method and a morphology optimization algorithm, to take the minimization of the functional area overlap and the gap area as an objective function, to adjust the direction vector of the module, to make the growth direction of the module match the stress direction and the functional demand of the toll island, and to obtain a module growth direction result.

[0128] The second optimization unit is configured to optimize the number of modules according to the module growth direction result, to determine the number of modules through a module division and configuration model based on a genetic algorithm, to take the module load capacity and the space utilization efficiency as optimization objectives, to perform cross, mutation and selection operations in the genetic algorithm, and to obtain a module number configuration result.

[0129] The first simulation unit is configured to perform simulation according to the module number configuration result, to adjust the edge shape and size of the module, to take the maximization of impact energy absorption and the minimization of structural deformation as optimization objectives, to simulate the stress distribution and energy absorption performance under the impact of different vehicles, and to obtain a preliminary functional framework.

[0130] In one specific embodiment disclosed in the present application, the division module 903 comprises:

[0131] The first clustering unit is configured to perform module division area demarcation according to the second information and the preliminary functional framework, to perform clustering analysis on the geometric area of the functional framework, to combine the mechanical demand as a clustering feature, to set the minimum functional unit constraint condition of the module, and to obtain a preliminary module division area.

[0132] The third optimization unit is used to optimize the boundary shape of the module based on the initial module segmentation region. By dynamically adjusting the edge shape of the module, with the goal of minimizing the complexity of the module boundary and maximizing the mechanical stability, the smoothness and structural strength of the module boundary are iteratively adjusted by combining the gradient descent method to obtain the optimized result of the module boundary shape.

[0133] The fourth optimization unit is used to design the module splicing method based on the module boundary shape optimization results and the second information. It determines the optimal connection scheme of the module through the topology optimization algorithm. With the goal of minimizing stress concentration at the splicing position and module splicing volume, it adjusts the number and distribution of connection points to obtain the module splicing method result.

[0134] The first design unit is used to design the connection nodes and error tolerance areas of the modules based on the module splicing method and third information. It uses graph theory algorithms to determine the global distribution and redundancy of the connection nodes and obtain the module segmentation result.

[0135] Example 3:

[0136] Corresponding to the above method embodiments, this embodiment also provides an optimization design device for a detachable highway toll island. The optimization design device for a detachable highway toll island described below and the optimization design method for a detachable highway toll island described above can be referred to in correspondence.

[0137] Figure 3 This is a block diagram illustrating an optimized design device 800 for a detachable highway toll island, according to an exemplary embodiment. Figure 3 As shown, the optimized design device 800 for a detachable highway toll island may include: a processor 801 and a memory 802. The optimized design device 800 for a detachable highway toll island may also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0138] The processor 801 is configured to control overall operation of the optimization design device 800 of the detachable highway toll island. The memory 802 is configured to store various types of data to support operation of the optimization design device 800 of the detachable highway toll island. For example, the memory 802 can store instructions for any application or method operating on the optimization design device 800 of the detachable highway toll island, and application-related data, such as contact data, sent and received messages, pictures, audio, video, and the like. The memory 802 can be implemented by any type of volatile or nonvolatile memory device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The multimedia component 803 can include a screen and an audio component. The screen can be a touch screen, for example. The audio component is configured to output and / or input audio signals. For example, the audio component can include a microphone configured to receive external audio signals. The received audio signals can be further stored in the memory 802 or transmitted through the communication component 805. The audio component also includes at least one speaker configured to output audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, which can be a keyboard, a mouse, a button, and the like. The buttons can be virtual buttons or physical buttons. The communication component 805 is configured to enable wired or wireless communication between the optimization design device 800 of the detachable highway toll island and other devices. The wireless communication, such as Wi-Fi, Bluetooth, near field communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 805 can include a Wi-Fi module, a Bluetooth module, an NFC module.

[0139] In an exemplary embodiment, the optimization design device 800 of the detachable highway toll island can be implemented by one or more of Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor or other electronic elements to execute the optimization design method of the detachable highway toll island.

[0140] In another exemplary embodiment, a computer readable storage medium including program instructions is also provided, which when executed by a processor, implements the steps of the optimization design method of the detachable highway toll island. For example, the computer readable storage medium can be the memory 802 including program instructions as described above, which can be executed by the processor 801 of the optimization design device 800 of the detachable highway toll island to complete the optimization design method of the detachable highway toll island.

[0141] The above description is merely a specific implementation of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered by the scope of protection of the present application.

Claims

1. A method for optimal design of a removable highway toll island, characterized by, The application relates to a method for designing a toll island, and belongs to the field of computer-aided design. The method comprises the following steps: acquiring first information, second information, third information, fourth information and fifth information, wherein the first information comprises spatial occupancy data and geometric morphology of the toll island, the second information comprises functional requirements and mechanical requirements, the third information comprises module splicing and connection requirements, the fourth information comprises arrangement requirements of universal wheels, and the fifth information comprises requirements of anti-collision performance and economic targets; designing a basic morphology of the toll island according to the first information, controlling growth directions, quantities, load capacities and anti-collision performances of the modules through a parameterized modeling module, and obtaining a preliminary functional framework; segmenting and positioning the functional framework by using a component segmentation positioning algorithm according to the second information and the preliminary functional framework, determining a module splicing mode through a topological optimization algorithm, setting connection nodes and error tolerance zones according to the third information, and obtaining a module segmentation result; optimizing mechanical performances of the modules by using a graph neural network optimization module according to the second information and the module segmentation result, and obtaining a module optimization result; carrying out a chain simulation according to the fourth information and the module optimization result, dynamically adjusting quantities, positions and load capacities of the universal wheels, and obtaining a universal wheel arrangement scheme; carrying out connection design according to the fifth information and the universal wheel arrangement scheme, optimizing distribution and sizes of the pins through finite element analysis, and obtaining a module connection scheme; 2. The method of claim 1, wherein, carrying out multi-target optimization processing according to the module optimization result, the universal wheel arrangement scheme and the module connection scheme, carrying out weight distribution and iterative optimization on overall design of the toll island, and obtaining a toll island design scheme. The method comprises the following steps: designing a basic morphology of the toll island according to the first information, controlling growth directions, quantities, load capacities and anti-collision performances of the modules through a parameterized modeling module, and obtaining a preliminary functional framework, which comprises the following steps: initializing a morphology of spatial data of the toll island according to the first information, dividing the target space into a plurality of functional areas through a Voronoi segmentation algorithm, combining boundary conditions of the geometric morphology of the toll island, and obtaining an initial functional area segmentation result; optimizing growth directions according to the initial functional area segmentation result, determining growth paths of the modules in the divided areas through a gradient descent method and a morphology optimization algorithm, taking minimization of functional area overlap and gap area as an objective function, adjusting a directional vector of the modules, matching the growth directions of the modules with stress directions and functional requirements of the toll island, and obtaining a module growth direction result; optimizing the number of modules according to the module growth direction result, determining the number of modules through a module division and configuration model based on a genetic algorithm, taking module load capacity and spatial utilization efficiency as optimization targets, and obtaining a module number configuration result through crossover, mutation and selection operations in the genetic algorithm; carrying out simulation according to the module number configuration result, adjusting edge shapes and sizes of the modules, taking maximization of impact energy absorption and minimization of structural deformation as optimization targets, simulating stress distribution and energy absorption performance under impacts of different vehicles, and obtaining a preliminary functional framework.

3. The method of claim 1, wherein, According to the second information and the preliminary function framework, the function framework is segmented by a component segmentation positioning algorithm, the module splicing mode is determined by a topology optimization algorithm, and the connection nodes and error tolerance zones are set according to the third information, to obtain a module segmentation result, including: According to the second information and the preliminary function framework, a module segmentation region is delimited, a clustering analysis is performed on the geometric region of the function framework, the mechanical requirements are combined as clustering features, the minimum functional unit constraint condition of the module is set, and a preliminary module segmentation region is obtained; According to the preliminary module segmentation region, the boundary shape of the module is optimized, the module edge shape is dynamically adjusted, the smoothness and structural strength of the module boundary are iteratively adjusted by combining the gradient descent method, and a module boundary shape optimization result is obtained; According to the module boundary shape optimization result and the second information, a module splicing mode is designed, the best connection scheme of the module is determined by a topology optimization algorithm, the number and distribution of the connection points are adjusted to minimize the stress concentration and the module splicing volume, and a module splicing mode result is obtained; According to the module splicing mode result and the third information, the connection nodes and error tolerance zones of the module are designed, the global distribution and redundancy of the connection nodes are determined by a graph theory algorithm, and a module segmentation result is obtained.

4. The method of claim 1, wherein, According to the second information and the module segmentation result, the mechanical properties of the module are optimized by a graph neural network, to obtain a module optimization result, including: According to the second information and the module segmentation result, the mechanical properties of the module are initially modeled, the stress conditions of each module are simulated by finite element analysis, the stress-strain properties of the module are mapped as node attributes, the connection strength and force transmission direction are mapped as edge attributes, and an initial mechanical graph structure model is generated; According to the initial mechanical graph structure model, a structure optimization process is performed, the mechanical graph is partitioned by a spectral clustering algorithm to reduce node connection redundancy and optimize edge weight distribution, the mechanical graph is divided into several subgraphs, and the node and edge attributes in the subgraphs are redefined, to obtain a mechanical subgraph structure; According to the mechanical subgraph structure, the node material distribution and edge connection strength of the module are predicted and optimized by a graph neural network, to minimize the stress uniformity of the module under dynamic load and the edge connection strength, the attributes of the nodes and edges are iteratively updated using a message passing network, and a material distribution-connection strength optimization result is obtained; According to the material distribution-connection strength optimization result, the overall mechanical properties of the module are comprehensively optimized, the global distribution of the module shape and material is adjusted by a multi-objective genetic algorithm, and a module optimization result is generated.

5. The method of claim 1, wherein, According to the fourth information and the module optimization result, a chain simulation is performed, the number, position and load-bearing capacity of the universal wheel are dynamically adjusted, and a universal wheel arrangement scheme is obtained, including: According to the fourth information and the module optimization result, a module bottom stress distribution is analyzed, a preliminary gimbal wheel arrangement area is delimited by identifying a bottom high stress area and a uniform bearing area in combination with a module edge and a center symmetry constraint condition, and a stress analysis preliminary result is obtained; According to the stress analysis preliminary result, a candidate area is dynamically distributed by an iterative point distribution algorithm, a gimbal wheel arrangement position is optimized in combination with a module connection node distribution, and a preliminary gimbal wheel arrangement scheme is obtained; According to the preliminary gimbal wheel arrangement scheme, a gimbal wheel quantity and a bearing capacity are discretely optimized, a bearing proportion is distributed by an integer linear programming algorithm, and a gimbal wheel bearing distribution scheme is generated; According to the gimbal wheel bearing distribution scheme, kinematics of gimbal wheel connection is simulated, a module moving path and load uniformity are simulated by Lagrange dynamics modeling, and a connection structure of the gimbal wheel is optimized, and a final gimbal wheel arrangement scheme is obtained.

6. The method of claim 1, wherein, According to the fifth information and the gimbal wheel arrangement scheme, a connection design is performed, a pin distribution and size are optimized by finite element analysis, a module connection scheme is obtained, including: According to the fifth information and the gimbal wheel arrangement scheme, a preliminary pin layout scheme is obtained by extracting a main stress transmission path between modules based on a stress path analysis algorithm, and delimiting an initial pin distribution position in combination with a boundary condition and a mechanical demand of the module; According to the pin layout preliminary scheme, a size and a shape of the pin are optimized, a plurality of pin design parameter combinations are dynamically generated, and a pin size optimization result is obtained by screening in combination with a performance simulation result of the pin under different stress states; According to the pin size optimization result, a strength, energy absorption and distribution density of the pin are iteratively optimized by a multi-objective particle swarm optimization algorithm, and a pin distribution optimization result is obtained; According to the pin distribution optimization result, a whole connection performance of the module is verified, a dynamic stress performance of the pin is simulated by finite element analysis, a whole impact resistance and long-term stability of the module connection are evaluated under dynamic impact and static load test conditions, and a final module connection scheme is obtained.

7. The method of optimizing the design of a removable highway toll island of claim 1, wherein, According to the module optimization result, the gimbal wheel arrangement scheme and the module connection scheme, a multi-objective optimization processing is performed, a toll island design scheme is obtained by weight distribution and iterative optimization of a whole toll island design, including: According to the module optimization result, the gimbal wheel arrangement scheme and the module connection scheme, a whole structure performance is modeled, a performance index system of the toll island design is established by a weighted objective function model, impact resistance, mobility, durability and economy are taken as targets and are distributed with different weight coefficients, a toll island whole optimization objective function is generated, and a comprehensive performance model is obtained; According to the comprehensive performance model, a parameter optimization is performed, a module size, a gimbal wheel position and a connection node distribution are iteratively adjusted by a nonlinear programming method, and an initial structure optimization result is obtained; According to the initial structure optimization result, the module splicing sequence and arrangement mode are optimized through a simulated annealing algorithm to minimize the connection stress and installation angle deviation, and the load distribution evaluation target value is changed to obtain a module global arrangement optimization result; According to the module global arrangement optimization result, a multi-objective genetic algorithm is used to comprehensively optimize the overall design scheme of the toll island, taking the impact resistance performance, movement performance and cost control as the optimization variables, and taking the module material distribution, structure form and movement configuration as the optimization variables, to generate an optimization solution through selection, crossover and mutation operations, and to evaluate the performance of the scheme by combining the fitness function, to obtain the final overall design scheme of the toll island.

8. A system for optimal design of a removable highway toll island, characterized by, Comprise: An acquisition module is configured to acquire first information, second information, third information, fourth information and fifth information, the first information comprises spatial occupation data and geometric form of the toll island, the second information comprises functional requirements and mechanical requirements, the third information comprises module splicing and connection requirements, the fourth information comprises arrangement requirements of universal wheels, and the fifth information comprises impact resistance performance requirements and economic targets; A modeling module is configured to design a basic form of the toll island according to the first information, control growth direction, quantity, load capacity and impact resistance performance of the module through parameterized modeling, and obtain a preliminary functional framework; A segmentation module is configured to segment and position the functional framework by using a component segmentation algorithm according to the second information and the preliminary functional framework, determine a module splicing mode by using a topology optimization algorithm, set connection nodes and error tolerance zones according to the third information, and obtain a module segmentation result; An optimization module is configured to optimize mechanical performance of the module by using a graph neural network according to the second information and the module segmentation result, and obtain a module optimization result; A simulation module is configured to perform linkage simulation according to the fourth information and the module optimization result, dynamically adjust quantity, position and load capacity of the universal wheels, and obtain a universal wheel arrangement scheme; A design module is configured to perform connection design according to the fifth information and the universal wheel arrangement scheme, optimize distribution and size of the pins by using finite element analysis, and obtain a module connection scheme; An output module is configured to perform multi-objective optimization processing according to the module optimization result, the universal wheel arrangement scheme and the module connection scheme, perform weight distribution and iterative optimization on the overall design of the toll island, and obtain a toll island design scheme.

9. The system for optimized design of a removable highway toll island of claim 8, wherein, The modeling module comprises: A first division unit is configured to initialize a form of spatial data of the toll island according to the first information, divide the target space into a plurality of functional areas by using a Voronoi segmentation algorithm, limit boundary conditions in combination with a geometric form of the toll island, and obtain an initial functional area division result; A first optimization unit is configured to perform growth direction optimization processing according to the initial functional area division result, determine a growth path of the module in the division area by using a gradient descent method and a form optimization algorithm, take minimization of functional area overlap and gap area as an objective function, adjust a direction vector of the module, make the growth direction of the module match a stress direction and functional requirements of the toll island, and obtain a module growth direction result; A second optimization unit is configured to optimize the number of modules according to the module growth direction result, determine the number of modules based on a module division and configuration model of a genetic algorithm, take the module load capacity and space utilization efficiency as optimization objectives, and obtain a module number configuration result through crossover, mutation and selection operations in the genetic algorithm. A first simulation unit is configured to simulate according to the module number configuration result, adjust the edge shape and size of the module, take maximizing impact energy absorption and minimizing structural deformation as optimization objectives, simulate stress distribution and energy absorption performance under different vehicle impacts, and obtain a preliminary functional framework.

10. The system for optimized design of a removable highway toll island of claim 8, wherein, The segmentation module comprises: A first clustering unit is configured to perform module segmentation region demarcation according to the second information and the preliminary functional framework, perform clustering analysis on the geometric region of the functional framework, combine the mechanical demand as a clustering feature, set a minimum functional unit constraint condition of the module, and obtain a preliminary module segmentation region. A third optimization unit is configured to optimize the boundary shape of the module according to the preliminary module segmentation region, dynamically adjust the module edge shape, take minimizing the module boundary complexity and maximizing the mechanical stability as objectives, combine a gradient descent method to iteratively adjust the smoothness and structural strength of the module boundary, and obtain a module boundary shape optimization result. A fourth optimization unit is configured to perform module splicing mode design according to the module boundary shape optimization result and the second information, determine the best connection scheme of the module through a topology optimization algorithm, take minimizing the stress concentration at the splicing position and the module splicing volume as objectives, adjust the number and distribution of the connection points, and obtain a module splicing mode result. A first design unit is configured to design the connection nodes and error tolerance zones of the module according to the module splicing mode result and the third information, determine the global distribution and redundancy of the connection nodes through a graph theory algorithm, and obtain a module segmentation result.

Images