Intelligent construction method and system for bridge model based on BIM

Abstract

The embodiment of the application provides a kind of based on BIM bridge model intelligent construction method and system, belong to based on BIM bridge model intelligent construction field;The method includes collecting bridge model construction related data and pre-processing, according to the bridge model construction related data after pre-processing, component library is established;The geometric linkage rule and topological connection logic of component in the component library are defined, and the initial BIM bridge model is generated through the generation of adversarial network;Lightweight finite element calculation is carried out on the initial BIM bridge model, and the optimized BIM bridge model is obtained;After the specification conflict detection and conflict parameter adjustment of the optimized BIM bridge model, the final bridge model is obtained.The application constructs parameter relation matrix and dynamic linkage mechanism, introduces the generation of complex bridge type in adversarial network, and introduces graph convolution network in adversarial network, to ensure that bridge topology is reasonable and parameters are dynamically linked.

Description

TECHNICAL FIELD

[0001] The application relates to an intelligent construction method and system for a bridge model based on BIM. BACKGROUND

[0002] In recent years, the informationization and intelligentization technology in the field of bridge engineering has developed rapidly, and the building information modeling (BIM) technology has become an important tool for the whole life cycle management of bridges. The traditional BIM modeling method mainly relies on manual operation, and the geometric model is manually constructed through CAD drawings, which has the problems of low efficiency and easy errors. With the popularization of three-dimensional laser scanning, unmanned aerial photography measurement and other technologies, the reverse modeling method based on point cloud data has significantly improved the collection accuracy of bridge geometric information, but the multi-source data fusion still faces challenges. For example, the fusion method based on feature point matching is easily disturbed by noise in complex scenes, and the iterative closest point (ICP) algorithm is sensitive to the initial registration, which may lead to local optimization problems. In the aspect of parametric modeling, the mainstream BIM platforms (such as Revit and Tekla) realize the dynamic adjustment of component form through family parameterization, but the existing parameter relationship expression is mainly based on linear logic, which is difficult to support the automatic update of complex nonlinear dependent parameters. In addition, the integration technology of finite element analysis and BIM has gradually matured, but the traditional finite element model has a large consumption of computing resources due to the fixed grid density, especially in the processing of large-span bridges, which is low in efficiency. SUMMARY

[0003] The purpose of the embodiments of the application is to provide an intelligent construction method and system for a bridge model based on BIM, which is used for efficient collaboration and intelligent optimization in the whole process of bridge design.

[0004] In order to achieve the above-mentioned purpose, the embodiments of the application provide an intelligent construction method for a bridge model based on BIM, which comprises the following steps: collecting bridge model construction related data and preprocessing, establishing a component library according to the preprocessed bridge model construction related data; defining the geometric linkage rules and topological connection logic of the components in the component library, and generating an initial BIM bridge model through a generative adversarial network; performing lightweight finite element calculation on the initial BIM bridge model to obtain an optimized BIM bridge model; and performing standard conflict detection and conflict parameter adjustment on the optimized BIM bridge model to obtain a final bridge model.

[0005] Optionally, the bridge model construction related data comprises spatial coordinate data, design input data, historical bridge BIM model data and construction method case library data.

[0006] Optionally, the defining the geometric linkage rules and topological connection logic of the components in the component library comprises: extracting bridge parameters from historical bridge BIM model data and construction method case library data, and dividing the bridge parameters into independent parameters and dependent parameters; establishing a parameter relationship matrix based on the independent parameters and the dependent parameters; performing mathematical relationship conversion according to the parameter relationship matrix to obtain mathematical relationship expressions of bridge components; and embedding the mathematical relationship expressions of the bridge components into a BIM platform to obtain parameterized component families with dynamic linkage.

[0007] Optionally, the defining the geometric linkage rules and topological connection logic of the components in the component library further comprises: classifying connection types of the components, and performing priority sorting on the classified connection types, wherein the priority sorting comprises design priority, calculation priority, conflict resolution priority, and maintenance priority; performing constraint condition modeling according to the priority sorting to obtain component connection constraint conditions; and embedding the component connection constraint conditions into a BIM platform and performing dynamic connection verification.

[0008] Optionally, the generating an initial BIM bridge model by using a generative adversarial network comprises: constructing a training data set of the generative adversarial network; constructing a generator and a discriminator of the generative adversarial network; designing a loss function and formulating a training strategy, and training the generator and the discriminator in stages using the training data set according to the formulated training strategy; mapping and converting bridge design parameters output by the generator, and using BIM to construct a bridge model to obtain an initial BIM bridge model.

[0009] Optionally, the constructing the generator and the discriminator of the generative adversarial network comprises: introducing a graph convolutional network into both the generator and the discriminator, and using the graph convolutional network in the generator to encode input graph structure data to obtain embedded representations of parameterized components and connection types of the components; regenerating geometric parameters and topological structures of a bridge type according to the embedded representations; and the discriminator processes graph structure data of the bridge type by using the graph convolutional network, and learns feature representations of the parameterized components and the connection types of the components.

[0010] Optionally, the performing lightweight finite element calculation on the initial BIM bridge model to obtain an optimized BIM bridge model comprises: performing geometric simplification and feature extraction on the initial BIM bridge model; performing adaptive mesh optimization on the simplified initial BIM bridge model to obtain a lightweight BIM bridge model; performing solver acceleration on the lightweight BIM bridge model and extracting mechanical indexes; and performing multi-objective optimization on the solver-accelerated lightweight BIM bridge model to obtain an optimized BIM bridge model.

[0011] Optionally, the multi-objective optimization comprises: using the non-dominated sorting genetic algorithm II generation, randomly generating an initial population, each individual representing a parameterized combination of a bridge design scheme; calculating the multi-objective function value of each individual, and performing non-dominated sorting; calculating the crowding distance of the individuals in each non-dominated sorting level, and using the tournament selection method to select individuals from the current population as parent individuals; using the simulated binary crossover operator to perform crossover operation on the selected parent individuals; and using the polynomial mutation operator to perform mutation operation on the offspring individuals, and checking whether the new individuals meet the constraint conditions.

[0012] Optionally, after the optimization BIM bridge model is subjected to specification conflict detection and conflict parameter adjustment, a final bridge model is obtained, wherein the conflict detection comprises geometric conflict detection and specification conflict detection, and the adjustment comprises geometric adjustment, parameter correction and iteration checking.

[0013] In another aspect, the application provides a BIM-based bridge model intelligent construction system for implementing the BIM-based bridge model intelligent construction method, which comprises a control module comprising a memory, a processor and a computer program stored on the memory and executable on the processor, and the processor executes the computer program to implement the BIM-based bridge model intelligent construction method.

[0014] The above technical solution improves the point cloud preprocessing algorithm, constructs a parameter relationship matrix and a dynamic linkage mechanism, introduces a generative adversarial network to generate a complex bridge type, introduces a graph convolution network in the generative adversarial network, guarantees the rationality of bridge topology and the dynamic linkage of parameters, realizes automatic compliance, combines adaptive grid and CUDA accelerated lightweight finite element analysis to improve the calculation efficiency and shorten the calculation time, and realizes efficient collaboration and intelligent optimization of the whole bridge design process through the specification conflict adjustment strategy driven by multi-objective optimization.

[0015] Other features and advantages of the embodiments of the application will be described in detail in the following specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0016] The accompanying drawings are included to provide a further understanding of the embodiments of the application, and constitute a part of the specification, and are used together with the following specific embodiments to explain the embodiments of the application, but do not constitute a limitation on the embodiments of the application. In the drawings:

[0017] Figure 1 is a BIM-based bridge model intelligent construction flowchart.

[0018] Figure 2 is a non-dominated sorting genetic algorithm II generation optimization flowchart. DETAILED DESCRIPTION

[0019] The following description of the application will be made with reference to the accompanying drawings. Figure 1 - the accompanying drawings Figure 2 The specific embodiments of the present application are described in detail. It should be understood that the specific embodiments described herein are merely intended to illustrate and explain the embodiments of the present application, and are not intended to limit the embodiments of the present application.

[0020] It should be noted that the acquisition, transmission, storage, use, processing, etc. of data in the technical solutions of the present application comply with relevant provisions of national laws and regulations. In the embodiments of the present application, some existing industry solutions such as software, components, models, etc. may be mentioned, which should be considered as exemplary, and the purpose is only to illustrate the feasibility of the implementation of the technical solutions of the present application, but it does not mean that the applicant has or will necessarily use the solutions.

[0021] The present inventors found in the process of realizing the present application that the existing bridge BIM modeling technology has the disadvantages of weak dynamicity of parameterized logic, low intelligent level of complex bridge type generation, and low finite element calculation efficiency.

[0022] Embodiment 1

[0023] With reference to Figures 1-2 For the first embodiment of the present application, the embodiment provides a bridge model intelligent construction method based on BIM, comprising:

[0024] S100: Collect bridge model construction related data and perform preprocessing, and establish a component library according to the preprocessed bridge model construction related data.

[0025] Specifically, different devices such as unmanned aerial vehicles, ground three-dimensional laser scanners, total stations, GPS, etc. are used to collect spatial coordinate data, and the collected spatial coordinate data is preliminarily preprocessed and fused. The preliminary preprocessing and fusion is geographical related data, and the fusion method can use methods based on feature point matching, ICP algorithm (Iterative Closest Point), etc. to obtain unified spatial coordinate point cloud data.

[0026] Preferably, the spatial coordinate point cloud data is denoised, a statistical filtering algorithm is used to calculate the neighborhood point cloud density of each point in the spatial coordinate data, and the density abnormal points are removed; a radius filtering algorithm is used to remove isolated points with insufficient point number in the local area.

[0027] Preferably, the spatial coordinate point cloud data after denoising is subjected to point cloud downsampling to reduce the number of point clouds and retain key geometric features.

[0028] Further, collect design input data, including collecting CAD (Computer-Aided Design) drawings related to bridge model construction. The collected CAD drawings should include detailed geometric information such as bridge span, bridge width, pier location, etc. Obtain load demand data, including dead load (such as bridge self-weight), live load (such as vehicle load, crowd load), wind load, earthquake load, etc. Collect specification documents, which should include various standards and requirements for bridge design, including structural size, material performance, safety factor, etc.

[0029] Further, organize historical bridge BIM model data, from which existing bridge component parameters, structural systems, design experience, etc. can be extracted to provide reference and reference for new bridge design, and improve design efficiency and quality.

[0030] Further, collect construction method case library data to understand the characteristics, difficulties and solutions of different bridge types in the construction process, which helps to fully consider the construction feasibility in the design stage, optimize the bridge design scheme, and reduce changes and risks in the construction process.

[0031] Preferably, after collecting all data, further preprocessing is performed, and the data used in the further preprocessing is the data after the preliminary preprocessing and other data except the geographic related data. The bridge design rule library is established based on the historical bridge BIM model data and the construction method case library data.

[0032] Further, classify the components according to function and form. The common bridge model component classification is shown in Table 1.

[0033]

[0034] It should be noted that the bridge model component classification is not limited to the content of Table 1.

[0035] Further, each bridge model component is controlled by key size parameters (length, width, height, thickness, etc.) to control the form. For example, for a box girder, its cross section is defined by the top plate thickness, web spacing, beam height, etc. Changing these size parameters can quickly adjust the geometric shape of the box girder to adapt to different design requirements. When modeling, use parameterized tools such as Revit (Autodesk) family editing function or Tekla (Tekla Structures) macro function to set the geometric parameters of the component as variable variables, and realize the dynamic adjustment of the component by adjusting the geometric parameters.

[0036] Further, material attribute binding is performed on the component, which automatically associates material properties (such as concrete strength, steel grade, etc.) and automatically calculates physical properties (mass, moment of inertia) based on material mechanical properties. For example, for a concrete girder, the mass per unit volume and the moment of inertia of the cross section are automatically calculated according to the concrete strength grade, providing basic data for subsequent structural analysis and design.

[0037] Preferably, a material database is established, and the characteristic parameters of common materials are recorded therein, facilitating calling and binding during modeling, while ensuring the accuracy and consistency of material properties.

[0038] Further, the pre-processed spatial coordinate point cloud data is unified with the coordinates of the CAD drawing, and the ICP algorithm is used to optimize the alignment accuracy (position accuracy, direction accuracy, scale accuracy) of the point cloud and the CAD.

[0039] It should be noted that the component properties need to strictly follow the IFC (Industry Foundation Classes) standard, and the component library is established after verifying that the component properties strictly follow the IFC standard. During the establishment of the component library, the category, parameters, and properties of the components are defined and stored according to the IFC standard, so that accurate information can be transmitted without loss or format incompatibility when data is exchanged and shared between different software.

[0040] Preferably, a component modification log (such as "box girder web thickness from 0.6m→0.8m") is recorded, and information such as the time, modification content, and modification reason of each modification is recorded in detail to trace the evolution process of the component. Support for historical version backtracking and preservation of optimization process data. When comparing the performance of different versions of components or restoring to a previous version, it can be easily operated, and through the analysis of historical version data, it provides a reference for subsequent design optimization.

[0041] Preferably, through multi-device fusion sampling, the accuracy and reliability of the spatial coordinate point cloud data are ensured, providing high-quality basic data for subsequent modeling. Based on historical BIM models, construction method case libraries, and CAD drawings, combined with parameterized modeling and material attribute binding, component reuse and dynamic adjustment are realized, and design efficiency is improved. The component properties follow the IFC standard, supporting cross-platform data exchange; recording modification logs and version backtracking facilitates design optimization iteration and responsibility tracking.

[0042] S200: Define the geometric linkage rules and topological connection logic of the components in the component library, and generate an initial BIM bridge model through a generative adversarial network.

[0043] Further, the bridge parameters are extracted from the historical bridge BIM model data and construction method case library data, and the bridge parameters are divided into independent parameters and dependent parameters; a parameter relationship matrix is established based on the independent parameters and the dependent parameters; the mathematical relationship conversion is performed according to the parameter relationship matrix, and the mathematical relationship formula of each component of the bridge is obtained; the mathematical relationship formula of each component of the bridge is embedded into the BIM platform, and a dynamically linked parameterized component family is obtained.

[0044] Specifically, geometric parameters such as cross-sectional size, length, height, etc. are extracted from the historical bridge BIM model; material parameters such as elastic modulus, yield strength, etc. are extracted; construction parameters such as construction sequence, prestress value, etc. are extracted; construction process parameters such as pouring segmentation, assembly sequence, etc. are extracted from the construction method case library data.

[0045] Further, the independent parameters are values that do not depend on other parameters, such as the total length of the bridge, the height of the main beam cross-section, and the material type; the dependent parameters are values that depend on the independent parameters or other dependent parameters, such as the distance between the piers, which depends on the total length of the bridge and the number of spans; the number of prestressed reinforcement, which depends on the cross-sectional size and the stress requirement.

[0046] Preferably, the upstream independent parameters and intermediate dependent parameters of each dependent parameter are analyzed to ensure that the relationship formula is complete and the logic is clear.

[0047] Further, a parameter relationship matrix is established, the rows in the parameter relationship matrix represent independent parameters, and the columns represent dependent parameters, and the mathematical relationship formula between the parameters is filled in the matrix to obtain the parameter relationship matrix. After obtaining the parameter relationship matrix, a comprehensive logical review is performed to ensure that all related parameters and their dependent relationships are correctly included.

[0048] Further, the mathematical relationship formula in the parameter relationship matrix is converted into a mathematical expression that can be embedded into BIM, and all mathematical relationship formulas are unified into a format suitable for BIM parameterized modeling. The converted mathematical relationship formula is verified to ensure that it is consistent with the actual bridge design logic. If there is a deviation, the mathematical relationship formula is adjusted.

[0049] Further, a parameterized component family is created in the BIM platform, the independent parameters are used as the parameters of the parameterized component family, the mathematical relationship formula is embedded into the parameters of the parameterized component family, the independent parameters are editable parameters; the dependent parameters are automatically calculated and assigned by the mathematical relationship formula, and are not exposed as editable parameters to avoid circular dependency.

[0050] Further, by modifying the independent parameters, it is verified whether the dependent parameters are automatically updated, and a multi-level dependent relationship test is performed, such as changes in the cross-sectional size of the main beam leading to changes in the self-weight, to obtain a dynamically linked parameterized component family that supports rapid design changes and optimization. The generated parameterized component family is saved for subsequent use.

[0051] Further, the connection types of the components are classified, the classified connection types are prioritized, the prioritization includes design priority, calculation priority, conflict resolution priority and maintenance priority; constraint condition modeling is performed according to the prioritization, component connection constraint conditions are obtained; the component connection constraint conditions are embedded in the BIM platform and dynamic connection verification is performed.

[0052] Specifically, the component connection types are classified (such as welding, bolt connection and hinged connection), and the connection types are prioritized according to design priority (such as structural safety), calculation priority (such as stress analysis), conflict resolution priority (such as collision detection) and maintenance priority (such as maintainability).

[0053] Further, constraint conditions are established for each connection type according to the prioritization of the connection types, and component connection constraint conditions are obtained. For example, welding connection requires that the components are aligned and have no gap, and bolt connection requires that the hole diameters are matched. The connection type constraint conditions are embedded in the model using the constraint function of BIM (such as the alignment constraint and the equidistance constraint of Revit), the effectiveness of the component connection constraint conditions is tested in BIM, after the test is completed, the connections that meet the component connection constraint conditions are retained, and the connections that do not meet the component connection constraint conditions are repaired, such as parameterized adjustment, topological relationship correction, material attribute rebinding and construction process optimization, and a BIM model of dynamic connection verification is obtained.

[0054] Preferably, if the bridge type belongs to the standardized bridges in the template library, such as simply supported bridges and T-shaped rigid frame bridges, based on the input parameters (span combination, load level, etc.), the bridge type template in the bridge design rule library is called, the bridge pier positions are automatically matched according to the spatial coordinate point cloud data, and a parameterized bridge type scheme is output, if the bridge type does not belong to the standardized bridges in the template library, the bridge type is determined as a complex bridge type.

[0055] Further, for the complex bridge type, such as special-shaped cable-stayed bridges and spatial curved bridge towers, corresponding geometric parameter data, topological relationship data and mechanical performance data are extracted from the pretreated historical bridge BIM model library data and construction method case library data as training data.

[0056] Further, a training data set of the generative adversarial network is constructed, a generator and a discriminator of the generative adversarial network are constructed, a loss function is designed and a training strategy is formulated, the generator and the discriminator are trained in stages according to the formulated training strategy using the training data set, the bridge design parameters output by the generator are mapped and converted, a bridge model is constructed using BIM, and an initial BIM bridge model is obtained.

[0057] Further, the constructing the generator and the discriminator of the generative adversarial network comprises: introducing a graph convolution network into both the generator and the discriminator, and using the graph convolution network to encode input graph structure data in the generator to obtain an embedding representation of a parameterized component and a connection type of the component; and regenerating geometric parameters and a topological structure of a bridge type according to the embedding representation; and the discriminator processes graph structure data of the bridge type through the graph convolution network and learns a feature representation of the parameterized component and the connection type of the component.

[0058] Specifically, the generator and the discriminator are designed using the generative adversarial network, and the graph convolution network is introduced into the generator, the training data is input into the generator, and the training data is arranged into topological graph structure data, nodes in the topological graph structure data represent various components (such as main girder, cable tower, cable, etc.) of the bridge, and edges represent the connection relationship between the components. The feature vector of the node contains geometric parameters and mechanical performance parameters, and the feature vector of the edge contains the type and attribute of the connection relationship. By introducing the graph convolution network to process the graph structure data, the graph convolution network can learn the feature representation of the node and the edge, and capture the complex relationship between the components of the bridge.

[0059] Further, in the generator, the input graph structure data is first encoded by the graph convolution network to obtain the embedding representation of the node and the edge. Then, the embedding representation of the node and the edge is used to generate the geometric parameters and the topological structure of a new bridge type. The output of the generator is a bridge type scheme, which contains geometric parameters such as main girder cross section shape, cable tower shape, cable arrangement, and topological relationship. The bridge type scheme generated by the generator and the collected real bridge type data are input into the discriminator to evaluate the similarity between the bridge type scheme generated by the generator and the real bridge type data, and to judge the rationality of the bridge type scheme. The discriminator also introduces the graph convolution network to process the graph structure data of the bridge type, learns the feature representation of the node and the edge, and compares the graph structure features of the generated scheme and the real data. The discriminator outputs the rationality probability of the generated bridge type scheme.

[0060] Preferably, a real-time specification checking module is embedded in the discriminator to ensure that the generated bridge type scheme meets the design standards.

[0061] Further, if the discriminator considers that the scheme is not reasonable, the rationality probability of the generated bridge type scheme is less than a set rationality threshold, the generator updates the network parameters of the generator and the discriminator through a back propagation algorithm to optimize the generation effect of the generator. After multiple iterations of optimization, the final complex bridge type scheme is output.

[0062] Preferably, the loss function of the generative adversarial network of the present scheme uses binary cross entropy to measure the distribution difference between the generated data and the real data.

[0063] Further, according to the bridge type scheme, the corresponding parameterized component family is called, and the parameters of the components in the parameterized component family are adjusted according to the spatial coordinate point cloud data, such as adjusting the height parameter of the bridge pier according to the elevation change of the point cloud; according to the above topological connection logic, the component connection points are automatically positioned, the called components are assembled, and the material properties, load labels and specification versions are automatically filled into the metadata; the IFC attributes are automatically bound when the initial BIM bridge model is generated, and the initial BIM bridge model is obtained after completion, and the initial BIM bridge model is checked for compliance.

[0064] Preferably, for the connection type that cannot be automatically assembled (such as the hinge of the spatial curved surface component), the conflict position should be recorded in the log and a manual editing interface should be provided.

[0065] Preferably, the mathematical relationship matrix of independent parameters (such as the total length of the bridge) and dependent parameters (such as the distance between the bridge piers) is constructed, combined with the embedded dynamic parameterized component family of the BIM platform, to realize “one place modification, global linkage”, reduce the repeated modeling work. For non-standardized bridge types (such as special-shaped cable-stayed bridges), the generative adversarial network is used to learn the characteristics of historical data combined with the graph convolution network, to generate innovative bridge type schemes that meet the mechanical performance and topological logic, which can improve the modeling ability of complex topological relationships of bridges, enhance the structural rationality of generated complex bridge types, and effectively shorten the complex design cycle. Based on the point cloud data, the component position is automatically matched, the specification checking module is embedded, the initial model is ensured to meet the design standards, and the risk of later rework is reduced.

[0066] S300: performing lightweight finite element calculation on the initial BIM bridge model to obtain an optimized BIM bridge model.

[0067] Further, the lightweight finite element calculation is performed on the initial BIM bridge model to obtain an optimized BIM bridge model, including: performing geometric simplification and feature extraction on the initial BIM bridge model; performing adaptive mesh optimization on the simplified initial BIM bridge model to obtain a lightweight BIM bridge model; performing solver acceleration on the lightweight BIM bridge model and extracting mechanical indicators; performing multi-objective optimization on the solver-accelerated lightweight BIM bridge model to obtain an optimized BIM bridge model.

[0068] Specifically, the geometric parameters of key components such as the main girder, tower, and pier in the initial BIM bridge model are ignored, and minor details such as bolt holes and decorative lines are ignored. Using adaptive mesh technology, the grid density is dynamically adjusted according to the stress gradient. The initial BIM bridge model is adaptively meshed in high-stress areas with a density of 5mm and in low-stress areas with a density of 50mm. When the stress gradient is >50MPa / m, it is considered a high-stress area. When the stress gradient is <10MPa / m, it is considered a low-stress area. When the stress gradient is between 10MPa / m and 50MPa / m, the grid density is not adjusted. Unstructured mesh generation is supported, and complex geometries such as curved towers and cable anchorage areas are compatible, resulting in a lightweight BIM bridge model.

[0069] Further, the stiffness matrix assembly and solution of the lightweight BIM bridge model are accelerated using CUDA. An appropriate sparse matrix compression storage format, such as the compressed sparse row (CSR) format, is used to store the non-zero elements and their corresponding row and column indices in a one-dimensional array. Only non-zero elements are stored, reducing memory usage. Based on the compression storage format, the matrix operation algorithm in finite element analysis is adjusted. During matrix multiplication, solving linear equations, and other operations, the compressed matrix is directly operated, improving computational efficiency.

[0070] Further, multiple working conditions (such as dead load + live load, wind load extreme combination, etc.) that need to be calculated are sorted and classified, and the load conditions and boundary conditions of each working condition are described and stored in detail. A batch processing task is formed by combining multiple working conditions, defining the calculation order and related parameters of each working condition, and assigning a unique identifier to each working condition for easy identification and management of the calculation results. At the same time, multiple solver processes or threads are started to handle different working conditions. Using multi-threading or distributed computing technology, different working conditions are solved in parallel. During the calculation process, the calculation status and resource occupation of each working condition are monitored and managed to ensure smooth calculation. The results obtained from each working condition are integrated and sorted, and the results are stored according to the working condition identifier.

[0071] Further, multi-objective optimization is performed on the lightweight BIM bridge model, including defining continuous variables and discrete variables. Continuous variables are design parameters that can vary continuously, such as main girder height, pier diameter, and cable spacing. Discrete variables are discrete design options such as section type and material type.

[0072] Further, bridge modeling-related objective functions are constructed, including a structure total mass minimization objective function and a stress uniformity maximization objective function. Bridge design-related constraints are embedded in the lightweight BIM bridge model.

[0073] Preferably, lightweighting helps improve computational efficiency and shorten calculation time.

[0074] Further, the multi-objective optimization includes using the non-dominated sorting genetic algorithm II generation, randomly generating an initial population, each individual representing a parameterized combination of a bridge design scheme; calculating the multi-objective function value of each individual, and performing non-dominated sorting; calculating the crowding distance of the individual in each non-dominated sorting level, and using the tournament selection method to select individuals from the current population as parent individuals; using the simulated binary crossover operator to perform crossover operation on the selected parent individuals; using the polynomial mutation operator to perform mutation operation on the offspring individuals, and checking whether the new individual meets the bridge design related constraint conditions.

[0075] Specifically, the NSGA-II algorithm (Non-dominated Sorting Genetic Algorithm II) is used to optimize the lightweight BIM bridge model. The population is initialized (randomly generating a parameter combination of a bridge design scheme), and an initial population is randomly generated, each individual representing a parameterized combination of a bridge design scheme, including the values of design variables such as geometric parameters such as main beam height, pier diameter, and discrete parameters such as cross-section type and material type.

[0076] Further, the individuals in the population are non-dominated sorted (the individuals are divided into different levels), and the individuals are divided into different levels. The non-dominated individual (i.e. the individual that is not inferior to other individuals in all objective functions) is the first level, then the non-dominated individual is continuously searched from the un-dominated individual, and so on, until all individuals are classified.

[0077] Further, in each non-dominated level, the crowding distance of the individual is calculated (to maintain the diversity of the population), which represents the density of the objective function space around the individual. Initially, a larger crowding distance is assigned to the boundary individual (the individual at the extreme position after the objective function is sorted), and then for other individuals, the objective function value is sorted, and the crowding distance is calculated according to the difference of the objective function value of adjacent individuals. The crowding distance of the individual is the sum of the crowding distances of the individual in each objective function dimension.

[0078] Further, the tournament selection method (selecting parent individuals) is used to select individuals from the current population as parent individuals. In the selection process, the individual with a high non-dominated level is preferentially selected; if the levels are the same, the individual with a large crowding distance is selected to maintain the diversity of the population.

[0079] Further, the selected parent individuals are subjected to crossover operation (to generate offspring individuals), to generate new individuals (offspring). A simulated binary crossover operator is used for continuous parameters, such as the height of the main beam, and a conventional crossover is used for discrete parameters, such as material combination, and a corresponding topology constraint crossover is used. According to the crossover probability and distribution index, the corresponding genes of the parent individuals are recombined to generate new gene combinations, to form the gene encoding of the offspring individuals. If the generated new individuals exceed the value range of the design variables, boundary processing is performed to set them to the boundary values.

[0080] Further, the offspring individuals are subjected to mutation operation (to fine-tune the offspring individuals). A polynomial mutation operator is used to make small random changes to the genes of the individuals according to the mutation probability and distribution index. The constraint conditions are checked, and the mutation rate of the current generation is calculated according to the adaptive mutation rate formula, and the strength of the mutation operation is adjusted according to the mutation rate to balance the global search and local search capabilities. The adaptive mutation rate formula is as follows:

[0081]

[0082] wherein, represents the mutation rate, represents the maximum mutation rate, t represents the current generation number, and T represents the total number of generations.

[0083] Further, for the generated new individuals (offspring), it is checked whether they satisfy the bridge design related constraint conditions, and for the individuals that violate the bridge design related constraint conditions, the violation degree of the bridge design related constraint conditions is calculated, and the individuals are punished according to the bridge design related constraint condition violation degree penalty function, to reduce their fitness.

[0084] Further, the constraint violation degree penalty function formula is as follows:

[0085]

[0086] wherein, represents the constraint penalty value, represents the Lagrange multiplier, P represents the parameter vector, i is the count index, and n represents the total number of constraint conditions, represents the ith constraint function.

[0087] Further, the parent population and the offspring population are combined to form a new population, and the combined population is subjected to non-dominated sorting and crowding distance calculation again, and then a certain number of individuals are selected according to a certain selection strategy (such as retaining the individuals of the front ranks, and retaining the individuals with large crowding distances in the same rank) to form a new parent population (retaining the individuals of high ranks and high crowding distances).

[0088] Further, check if the termination condition is met, such as reaching the maximum number of iterations, the rate of change of the Pareto frontier being less than a given threshold, etc. If the termination condition is met, stop the algorithm and output the Pareto optimal solution set; otherwise, return to the "cross operation" to continue iteration until the Pareto optimal solution set that meets the termination condition is obtained, select one or more optimal solutions from the Pareto optimal solution set, and return the design variables of the selected optimal solutions to the parametric component family. Since the components have been dynamically linked through the parameter relationship matrix and mathematical expressions (such as automatic adjustment of the distance between piers with the total length of the bridge), the BIM model automatically reconstructs the geometry and properties after updating the parameters.

[0089] Further, the BIM model after multi-objective optimization is subjected to a second finite element analysis (based on adaptive mesh and CUDA acceleration) to verify whether it meets the structural performance requirements (such as stress, deformation, stability, etc.). Check if the model after multi-objective optimization meets the standards in the specification document and the design input. If there is a conflict, return to the multi-objective optimization process to adjust the relevant parameters of the BIM model after multi-objective optimization to obtain an optimized BIM bridge model.

[0090] Preferably, through methods such as geometric simplification (ignoring minor details such as bolt holes), adaptive mesh (stress gradient driven densification / sparse), CUDA acceleration, etc., the computational resource consumption can be significantly reduced, and the finite element analysis efficiency can be improved. Using the optimized NSGA-Ⅱ algorithm to balance the minimization of structural mass, the maximization of stress uniformity, etc., combined with a constraint violation penalty function, the optimization scheme is ensured to be optimal in safety, cost, material, etc. dimensions, and the optimization results are directly returned to the parametric component family, and the BIM model is automatically reconstructed through dynamic linkage, avoiding manual adjustment errors and accelerating the iterative closed loop.

[0091] S400: After the specification conflict detection and conflict parameter adjustment of the optimized BIM bridge model, a final bridge model is obtained.

[0092] Further, after the specification conflict detection and conflict parameter adjustment of the optimized BIM bridge model, a final bridge model is obtained, which includes: the conflict detection includes geometric conflict detection and specification conflict detection; the adjustment includes geometric adjustment, parameter correction and iterative checking.

[0093] Further, the optimized BIM bridge model is subjected to rule checking, including geometric collision, load combination and specification requirements, using BIM platform plug-ins to automatically scan conflict items in the model and generate a conflict report.

[0094] Further, the conflict items in the conflict report are classified into geometric collision, load inconsistency, connection constraint violation, etc. according to the conflict type, and are sorted according to safety, constructability and maintainability to determine the adjustment priority.

[0095] Further, the call parameter relationship matrix is called to automatically locate conflict related parameters (such as beam height, pier distance, cable force), and multiple sets of adjustment candidate schemes are generated according to preset rules and optimization targets. Lightweight finite element analysis and specification checking are performed on each set of candidate adjustment candidate schemes, and the performance and specification compliance of the adjusted model are evaluated. Comprehensive multi-objective indicators and conflict priorities are adopted to select the optimal adjustment scheme by using a decision matrix or a weighted scoring method. The parameters of the optimal adjustment scheme are written back to the parameterized family, and the final bridge BIM model is regenerated and exported, and a conflict resolution report and a parameter modification log are attached.

[0096] Preferably, the BIM plug-in is used to scan geometric collision and load combination conflict, generate a structured conflict report, and reduce manual inspection cost; the conflict items are prioritized according to safety, constructability and the like, the associated parameters are located by using the parameter relationship matrix, multiple sets of adjustment schemes are generated and verified by finite element analysis, and the adjusted model is ensured to consider performance and specification.

[0097] The application further provides a BIM-based bridge model intelligent construction system for implementing the BIM-based bridge model intelligent construction method.

[0098] The application provides a storage medium having a program stored thereon, and the program is executed by a processor to implement the BIM-based bridge model intelligent construction method.

[0099] The application provides a processor for running a program, and the program is executed to implement the BIM-based bridge model intelligent construction method.

[0100] The application provides a device including a processor, a memory and a program stored in the memory and executable on the processor, and the processor is executed to implement the BIM-based bridge model intelligent construction method. The device herein can be a server, a PC, a PAD, a mobile phone or the like.

[0101] The application further provides a computer program product adapted to implement the BIM-based bridge model intelligent construction method when executed on a data processing device.

[0102] Those skilled in the art will appreciate that embodiments of the application can be practiced in a network computing environment. FIG. 1 illustrates an example of one embodiment of a network computing device 100. The computing device 100 includes one or more processors 102, a memory 104, an input / output interface 106, and a network interface 108. The computing device 100 can be a server, a desktop computer, a laptop computer, a handheld computer, a network appliance, a PDA, a cell phone, or any other computing device.

[0103] The present application can be described in the general context of method steps, which can be implemented in software, hardware, or a combination of software and hardware. The method steps can be stored on computer-readable storage media, which can include volatile and non-volatile, removable and non-removable media implemented in a method or technology suitable for storage of information such as computer readable instructions, data structures, program modules, or other data. Computer-readable storage media can include, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information. The computer-readable storage media can be accessed by a central processing unit (CPU) or other hardware device. Figure 1 Figure 1

[0104] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 Figure 1

[0105] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 Figure 1

[0106] In one typical configuration, the computing device includes one or more processors (CPU's), input / output interfaces, network interfaces, and memory.

[0107] The memory can include non-persistent memory and / or persistent memory, such as flash memory or other non-volatile memory, or a combination of both. The memory can be volatile, non-volatile, or a combination of both. The memory can be embedded, removable, or a combination of both. The memory is an example of computer-readable media. ​​​​​​​​​​​​​​​​​​​​​​

[0108] Computer-readable media includes permanent and non-permanent, movable and non-movable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device. According to the definition herein, computer-readable media does not include transitory media such as modulated data signals and carriers.

[0109] It should also be noted that the terms "comprising", "containing", or any other variant thereof are intended to cover non-exclusive inclusions, so that a process, method, article or apparatus that includes a list of elements does not only include those elements, but also includes other elements not explicitly listed, or further includes elements inherent in such a process, method, article or apparatus. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or apparatus that includes the element.

[0110] The above only is an embodiment of the present application, and is not used to limit the present application. For those skilled in the art, the present application can have various changes and variations. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the scope of claims of the present application.

Claims

1. A BIM-based bridge model intelligent construction method, characterized in that, The application relates to a bridge model construction method and device. Bridge model construction related data is collected and preprocessed, and a component library is established according to the preprocessed bridge model construction related data; Geometric linkage rules and topological connection logic of components in the component library are defined, and a generative adversarial network is used to design a generator and a discriminator, including introducing a graph convolutional network into the generator, extracting corresponding geometric parameter data, topological relationship data and mechanical performance data from preprocessed historical bridge BIM model library data and construction method case library data as training data, inputting the training data into the generator, and thus arranging the training data into topological graph structure data; Nodes in the topological graph structure data represent various components of the bridge, and edges represent the connection relationship between the components, the feature vector of the node includes the geometric parameters and mechanical performance parameters of the component, and the feature vector of the edge includes the type and attribute of the component connection relationship; bridge design parameters output by the generator are mapped and converted, and a BIM is used to build a bridge model to obtain an initial BIM bridge model; Lightweight finite element calculation is performed on the initial BIM bridge model to obtain an optimized BIM bridge model; After standard conflict detection and conflict parameter adjustment are performed on the optimized BIM bridge model, a final bridge model is obtained. 2.The BIM-based bridge model intelligent construction method according to claim 1, characterized in that, The bridge model construction related data includes spatial coordinate data, design input data, historical bridge BIM model data and construction method case library data. 3.The BIM-based bridge model intelligent construction method of claim 1, wherein, The definition of the geometric linkage rules and the topological connection logic of the components in the component library includes: Bridge parameters are extracted from historical bridge BIM model data and construction method case library data, and the bridge parameters are divided into independent parameters and dependent parameters; Based on the independent parameters and the dependent parameters, a parameter relationship matrix is established; According to the parameter relationship matrix, mathematical relationship conversion is performed to obtain bridge component mathematical relationship formulas; The bridge component mathematical relationship formulas are embedded into a BIM platform to obtain parameterized component families with dynamic linkage. 4.The BIM-based bridge model intelligent construction method of claim 1, wherein, The definition of the geometric linkage rules and the topological connection logic of the components in the component library also includes: The connection types of the components are classified, and the classified connection types are prioritized, The prioritization includes design priority, calculation priority, conflict resolution priority and maintenance priority; Constraint condition modeling is performed according to the prioritization to obtain component connection constraint conditions; The component connection constraint conditions are embedded into a BIM platform and dynamic connection verification is performed. 5.The BIM-based bridge model intelligent construction method of claim 1, wherein, The BIM is used to build a bridge model to obtain an initial BIM bridge model, including: A training data set of the generative adversarial network is constructed; The generator and the discriminator of the generative adversarial network are constructed; A loss function is designed and a training strategy is formulated, and the generator and the discriminator are trained in stages according to the formulated training strategy using the training data set; Bridge design parameters output by the generator are mapped and converted, a BIM is used to build a bridge model, and an initial BIM bridge model is obtained. 6.The BIM-based bridge model intelligent construction method of claim 5, wherein, The construction of the generator and the discriminator of the generative adversarial network includes: The graph convolution network is introduced in the generator and the discriminator, and the graph convolution network is used in the generator to encode the input graph structure data to obtain an embedding representation of the parameterized component and the connection type of the component; The geometric parameters and the topological structure of the bridge type are regenerated according to the embedding representation; The discriminator processes the graph structure data of the bridge type through the graph convolution network, and learns a feature representation of the parameterized component and the connection type of the component. 7.The BIM-based bridge model intelligent construction method of claim 1, wherein, The initial BIM bridge model is subjected to lightweight finite element calculation to obtain an optimized BIM bridge model, including: The initial BIM bridge model is subjected to geometric simplification and feature extraction; The simplified initial BIM bridge model is subjected to adaptive mesh optimization to obtain a lightweight BIM bridge model; The lightweight BIM bridge model is subjected to solver acceleration and extraction of mechanical indexes; The lightweight BIM bridge model subjected to solver acceleration is subjected to multi-objective optimization to obtain an optimized BIM bridge model. 8.The BIM-based bridge model intelligent construction method of claim 7, wherein, The multi-objective optimization includes: The non-dominated sorting genetic algorithm II is used to randomly generate an initial population, and each individual represents a parameterized combination of a bridge design scheme; The multi-objective function value of each individual is calculated, and non-dominated sorting is performed; The crowding distance of the individual in each non-dominated sorting level is calculated, and the tournament selection method is used to select individuals from the current population as parent individuals; The selected parent individuals are subjected to crossover operation by using a simulated binary crossover operator; The child individuals are subjected to mutation operation by using a polynomial mutation operator, and it is checked whether the new individual meets the constraint condition. 9.The BIM-based bridge model intelligent construction method of claim 1, wherein, After the optimized BIM bridge model is subjected to specification conflict detection and conflict parameter adjustment, a final bridge model is obtained, including that the conflict detection includes geometric conflict detection and specification conflict detection; and the adjustment includes geometric adjustment, parameter correction and iterative checking. 10.A BIM-based intelligent bridge model construction system, characterized in that, The system includes a control module, the control module includes a memory, a processor and a computer program stored on the memory and executable on the processor, and the processor executes the computer program to implement the BIM-based bridge model intelligent construction method according to any one of claims 1-9.

Images