Modularized building generation and optimization method and system based on BIM graphic data model

Abstract

The embodiment of the invention provides a modular building generation and optimization method and system based on a BIM graphic data model. The method is applied to the technical field of building information models and computer aided design, and comprises the following steps: acquiring spatial layout information of a modular building based on a building information model to obtain model data; converting the extracted BIM model data into a graph data model, and generating graph data containing space nodes and connecting edges; based on the optimization target, a multi-target optimization method is adopted to optimize the graphic data model, and a plurality of alternative design schemes are obtained; and scoring the alternative design schemes by adopting the fitness evaluation model, converting the design scheme with the highest score into an operable BIM format file, and outputting the operable BIM format file through the cloud data to complete the design process. Through deep fusion of the BIM and the graph data model, the dynamic optimization algorithm and the cloud collaboration technology, the technical problems that a three-dimensional modeling method is complex in process and insufficient in precision are solved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of building information modeling and computer-aided design, and in particular to a modular building generation and optimization method and system based on a BIM graphic data model. Background Art

[0002] In traditional architectural design, especially for complex modular buildings, designers are often required to manually arrange functional areas according to requirements and building standards, and manually adjust and optimize spaces. This approach is not only time-consuming and labor-intensive, but also susceptible to subjective judgment when dealing with multiple constraints. This makes it difficult to arrive at the optimal design within a short timeframe, potentially leading to design flaws or implementation difficulties. Therefore, traditional design methods cannot meet the dual requirements of efficiency and precision in modern architectural design, especially during the rapid iteration and optimization of modular building design. Traditional methods are clearly insufficient.

[0003] Building Information Modeling (BIM) technology can provide spatial layout and geometric information of buildings, and can also support the integration of information such as materials, structures, and equipment, and can realize design visualization, simulation, and optimization. In modular building design, BIM is usually used for pre-construction planning and design, but BIM still lacks effective solutions for how to efficiently generate design plans, how to optimize among multiple alternative plans, and how to achieve automated design optimization. Existing modular building design methods usually rely on manual design, and designers need to manually arrange building modules based on multiple factors such as functional requirements and construction standards. Although in some cases, architectural design software can provide certain design assistance functions, there are still many shortcomings.

[0004] Therefore, there is an urgent need for a new design scheme that can automatically generate multiple design schemes that meet the design goals and constraints based on Building Information Modeling (BIM) data, and can quickly screen out the optimal scheme through optimization algorithms. Summary of the Invention

[0005] The present disclosure provides a modular building generation and optimization method and system based on BIM graphic data model, which combines building information model (BIM) and geographic information system (GIS) technology to realize three-dimensional geological data display, and at least solves the technical problems of complex process and insufficient accuracy of three-dimensional modeling methods.

[0006] According to a first aspect of the present disclosure, a modular building generation and optimization method based on a BIM graphic data model is provided, comprising the following steps: Acquiring spatial layout information of the modular building based on the building information model to obtain model data, wherein the model data includes the spatial position, spatial area, connection relationship of each module, and the dimensions of passages and doors between adjacent modules; Convert the extracted BIM model data into a graph data model to generate graph data containing spatial nodes and connecting edges; Based on the optimization goal, a multi-objective optimization method is used to optimize the graph data model and obtain several alternative design schemes; The fitness evaluation model is used to score the alternative design schemes, and the design scheme with the highest score is converted into an operable BIM format file and output through the cloud to complete the design process.

[0007] According to the above aspects and any possible implementation, there is further provided an implementation, wherein the modular building specifically includes a structural form, a shear wall structure, a podium wall structure, a cast-in-place component, and a volume module; The shear wall structure is a tubular structure, the pier wall structure is a supporting structure, the cast-in-place components include beams, columns, walls, slabs, stairs, cores and platforms, and the volume module includes prefabricated components.

[0008] According to the above aspects and any possible implementation, there is further provided an implementation, wherein the attributes of the nodes of the graph data model include unit space size, volume, and functional requirements; The edges of the graphical data model include connection relationships between module units, wherein the weights of the edges are determined according to doors, windows, passages and streamline factors between modules.

[0009] According to the above aspects and any possible implementation, a further implementation is provided, in which the specific generation process of the graphic data model is as follows: Extract the graphs of vertices and edges of the shear wall structure, tubular structure, and supporting structure to form an adjacency graph and a connectivity graph. Combined with the topological relationship of the modular building structure, the graph structure data is expanded to obtain an extended graph containing the hierarchical relationship of the structural system. Extract the geometric boundaries of precast and cast-in-place components in the BIM model, define their topological relationships, and associate component types, connection methods, and expansion diagrams; Extract the geometric and semantic information of the BIM model and store them in a matrix according to component type to obtain matrix data. Then calculate and update the weights between the B-rep model and modules. Extracting geometric constraints and semantic constraints between components based on the matrix data and the updated weights, and combining them with the extended graph to generate a graph structure including geometric accuracy, structural type, and construction logic; A modular building graphic data model is generated based on the graph structure, wherein the output form of the graphic data model includes the combination relationship of the structural system and the spatial distribution and connection node coordinates of prefabricated and cast-in-place components.

[0010] According to the above aspects and any possible implementation, there is further provided an implementation, wherein the optimization objectives include: maximizing space utilization, minimizing connection redundancy, and satisfying structural safety constraints; The multi-objective optimization method adopts the NSGA-II genetic algorithm, in which the selection probability of individuals in the algorithm population is based on non-dominated sorting and crowding distance, specifically: ; in, Represents the non-dominated ranking level of individual i, where the smaller the level value, the higher the priority of the individual in the Pareto front, corresponding to the comprehensive advantages of the building scheme in space utilization, connection redundancy and structural safety constraints. For individuals The crowding distance measures the distribution density of adjacent individuals in the solution set, which is used to avoid excessive aggregation of optimization results and homogenization of modular solutions. is a minimum constant, The population size to set for the genetic algorithm.

[0011] Regarding the above aspects and any possible implementation, a further implementation is provided, wherein the process of optimizing the graph data model using a multi-objective optimization method to obtain several alternative design solutions is as follows: Generate an adjacency matrix based on the spatial layout of the modular building, where the element value of the adjacency matrix is ​​a binary identifier. If module i is adjacent to module j, the corresponding element value is 1, otherwise it is 0; Establish a connection matrix, the element value of the connection matrix is ​​determined by the connection weight W between modules ij Represents, wherein the connection weight is calculated based on the extracted door and window sizes, streamline complexity, and structure type; Based on the matrix data, the NSGA-II algorithm is used to drive the Pareto frontier, and the objective function is defined as: ; in, Represents the module extracted based on the B-rep model Plane projected area, is the total building area, is the minimum permissible area of ​​a single module as determined by the building code; and Corresponding modules and The spatial position mapped by the vertex coordinates of the graph structure, The module spacing threshold is dynamically adjusted according to the structure type; Representation Module With adjacent modules The connection weight is calculated by multiplying the door and window area ratio and the streamline overlap rate. is the minimum connection strength based on the structure type; and Represents the module set and the graph structure vertex set respectively, which are used to constrain the topological connection rules between modules. f1 and f2 are both objective functions. pass Nonlinear mapping constrains the module area discreteness to ensure And evenly distributed.

[0012] According to the above aspects and any possible implementation, there is further provided an implementation, wherein the multi-objective optimization method is provided with constraint conditions; The constraints include module size constraints, connection strength constraints, and traffic safety constraints, specifically: The module size constraint stipulates that the area of ​​each module shall not be less than a preset minimum area value; The connection strength constraint stipulates that the connection weight between any two modules is not less than a minimum threshold dynamically set according to the building type; The passage safety constraint stipulates that the channel size reaches a preset safety standard value.

[0013] According to the above aspects and any possible implementation, there is further provided an implementation, wherein the fitness evaluation model includes sub-scoring items; the sub-scoring items include a space utilization score, a connection efficiency score, a comprehensive fitness score, and a constructability score; The space utilization score is the ratio of the total module area to the total building area; The space utilization score is the sum of the connection weights between all modules; The comprehensive fitness score is defined as the weighted sum of multiple objectives; The constructability score is obtained through construction simulation calculation, wherein the parameter system of the construction simulation includes material cost budget parameters, construction schedule parameters and construction difficulty parameters.

[0014] According to the above aspects and any possible implementation, a further implementation is provided, wherein the comprehensive fitness score is specifically: ; Among them, the comprehensive fitness score is the sum of the areas of each module Total construction area The ratio of , which reflects the efficiency of the module layout in utilizing the building space, and the connection efficiency score is obtained by summing the weights of the connections between modules. Quantization, weight Based on the dynamic calculation of door and window matching, traffic flow overlap rate and structure type, λ1, λ2, λ3 are weight coefficients, and ,γ is the attenuation coefficient, d ij is the actual distance between modules i and j, T cost is the total construction cost, T budget The budget cap.

[0015] According to a second aspect of the present disclosure, a modular building generation and optimization system based on a BIM graphic data model is provided, which is used to implement the modular building generation and optimization method based on a BIM graphic data model as described in the first method, comprising: a model data acquisition module, a graphic data model construction module, a model design module, and an evaluation module; The model data acquisition module is used to acquire spatial layout information of the modular building based on the building information model to obtain model data; The graphic data model construction module is used to convert the extracted BIM model data into a graphic data model to generate graph data including spatial nodes and connection edges; The model design module is used to optimize the graphic data model based on the optimization goal using a multi-objective optimization method to obtain several alternative design solutions; The evaluation module is used to score the alternative design schemes using the fitness evaluation model, and convert the design scheme with the highest score into an operable BIM format file, and output it through the cloud data to complete the design process.

[0016] Compared with the prior art, the present invention has the following technical effects: (1) The present invention can realize the automatic generation and optimization of architectural design through a design method based on the BIM diagram data model, significantly improving design efficiency, reducing manual intervention, and increasing the degree of automation in the design process. Secondly, the automatically generated design scheme can fully consider the connection and adaptability between modules, ensuring the rationality of the design and the integrity of the function. Unlike traditional design methods, the present invention can automatically generate multiple design schemes under multiple constraints and select the optimal scheme through evaluation and optimization algorithms, effectively improving the quality of architectural design and construction feasibility.

[0017] (2) The present invention has strong flexibility and scalability, and can adapt to modular building design tasks of different types, scales, and complexities. The system can optimize design solutions in real time, generate multiple design alternatives, and customize them according to project requirements, greatly improving the adaptability and efficiency of building design.

[0018] (3) This invention can effectively improve the automation and optimization level of modular building design, solve the problems in traditional design methods, and promote the development of building design towards intelligence and automation. This method has important application prospects and can be widely used in modular building design, urban building planning, architectural design software development and other fields, providing new technical support for the future development of the construction industry.

[0019] It should be understood that the contents described in the Summary of the Invention section are not intended to limit the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. The accompanying drawings are provided for a better understanding of the present disclosure and do not constitute a limitation of the present disclosure. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, among which: Figure 1 A schematic diagram of a modular building generation and optimization method based on a BIM graphic data model according to an embodiment of the present disclosure is shown; Figure 2 A framework diagram of a modular building generation and optimization method based on a BIM graphic data model according to an embodiment of the present disclosure is shown; Figure 3 A structural schematic diagram of the information required for modular building classes in a modular building generation and optimization method based on a BIM graphic data model according to an embodiment of the present disclosure is shown; Figure 4 A schematic diagram of a method for extracting required information from BIM using a modular building generation and optimization method based on a BIM graphic data model according to an embodiment of the present disclosure is shown; Figure 5 A schematic diagram of the structure of a modular building generation and optimization system based on a BIM graphic data model according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0021] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] To make the purpose, technical solutions, and advantages of the embodiments of the present disclosure more clear, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of the present disclosure.

[0023] In traditional architectural design, especially in the design process of complex modular buildings, designers are usually required to manually arrange various functional areas according to requirements and building standards, and manually adjust and optimize the space. This method is not only time-consuming and labor-intensive, but also when dealing with multiple constraints, the design plan is easily affected by human subjective judgment, resulting in the inability to come up with the best design plan in a short period of time. In addition, there is another important problem with traditional design, that is, it is difficult for designers to simultaneously consider the needs of multiple aspects such as building space function, structural safety, module connection, and convenience of passage, which may cause design defects or implementation difficulties in the design plan. Therefore, traditional design methods cannot meet the dual requirements of modern architectural design for efficiency and accuracy, especially in the rapid iteration and optimization process of modular building design, traditional methods are obviously insufficient.

[0024] Building Information Modeling (BIM), a revolutionary architectural design technology, has been widely adopted in the construction industry in recent years. Through digital models, BIM provides strong support for all phases of building design, construction, and operations. BIM not only provides spatial layout and geometric information about buildings but also supports the integration of information such as materials, structures, and equipment, enabling design visualization, simulation, and optimization. In traditional architectural design, BIM is primarily used for the visualization of buildings and the integrated management of information. However, existing BIM models primarily focus on the storage and display of static building information, failing to fully realize their potential for design optimization and generation.

[0025] In modular building design, BIM is often used for pre-construction planning and design. However, BIM still lacks effective solutions for efficiently generating design solutions, selecting optimal solutions among multiple alternatives, and achieving automated design optimization. Currently, most BIM-based modular building design solutions rely on manual intervention and are unable to effectively compare and optimize multiple design options in real time. Although some studies have proposed BIM-based modular building design methods, these methods still have significant limitations. First, they often rely on static design parameters and manual adjustments, lacking flexibility. Second, they often optimize design based on simple rules and constraints, failing to fully consider the complex spatial relationships and connection requirements between modules in modular buildings. As a result, the design results are often unsatisfactory.

[0026] Especially in modular buildings, designers must fully consider factors such as spatial layout, inter-module connectivity, and compatibility. The connectivity between modules not only influences the building's functional layout but also directly impacts the feasibility of construction and subsequent operations. Ignoring these connectivity and compatibility requirements can lead to problems during modular construction and even affect the building's overall performance. Therefore, modular building design requires not only a rational spatial layout but also optimal inter-module connectivity and compatibility to ensure the building's integrity and practicality.

[0027] Existing modular building design methods typically rely on manual design, requiring designers to manually arrange building modules based on multiple factors such as functional requirements and construction standards. Although architectural design software can provide certain design assistance functions in some cases, it still has many shortcomings. For example, the generation of design solutions often fails to fully consider the pros and cons of different design solutions and lacks an automated design optimization mechanism. Moreover, the design optimization process usually requires designers to make repeated adjustments, which is inefficient. Furthermore, existing design methods lack a mechanism for comprehensive optimization based on the particularities of modular buildings. For example, the spatial area of ​​the modules, the connection method between modules, the size of open channels, etc. have not been fully included in the scope of design optimization, resulting in significant limitations on the quality and efficiency of architectural design.

[0028] Therefore, in modular building design, there is an urgent need for a new design method that can automatically generate multiple design solutions that meet design goals and constraints based on Building Information Modeling (BIM) data, and quickly select the optimal solution through optimization algorithms. This method not only significantly improves design efficiency but also automatically generates the optimal design solution under multiple constraints, effectively solving the problems of excessive manual intervention and insufficient design optimization in traditional design methods.

[0029] To fill gaps in existing technology, many researchers have attempted to innovate based on BIM, proposing methods that use computer algorithms to optimize modular building design. These methods typically rely on algorithms to model the connections between building spaces and modules, automatically generating multiple design options. However, these methods often overlook the complex inter-module connections and the diversity of spatial layouts, as well as the impact of uncertainties in the design process. More importantly, traditional design methods are mostly based on predetermined rules and standards, making it difficult to flexibly adjust designs based on actual needs.

[0030] Therefore, to effectively improve the automation and optimization capabilities of modular building design, it is necessary to introduce a design method based on the BIM graphical data model. Leveraging the spatial layout information and graphical representations within the BIM model, an automated and optimized design generation system can be constructed. This system can rapidly generate multiple design alternatives while meeting design objectives, evaluate their suitability, and select the optimal solution. This approach not only improves design efficiency and shortens the design cycle, but also fully considers the complex relationships and compatibility between modules, thereby enhancing the quality and practical application value of modular building design.

[0031] The goal of this invention is to provide a method for automatically generating and optimizing modular building designs based on a BIM data model. This method effectively overcomes the shortcomings of existing technologies, addressing the low efficiency, poor quality, and inadequate optimization issues in modular building design, and provides a new approach and method for modular building design. This innovative method significantly improves designer efficiency, reduces manual intervention, and enhances the quality and rationality of building design solutions, ultimately promoting the intelligent and automated development of modular building design.

[0032] It's worth noting that to better capture and represent modular building information within a BIM environment, this invention leverages the proposed extensions of existing and new entities within the MVD (Model View Definition) presentation within the data model. This integration integrates IFC model view definitions and industry standards (such as the PPVC manual and code of practice) to refine and standardize the storage and interaction of building information. MVD supplements or expands traditional IFC files to encompass the unique components, connections, and construction requirements of modular buildings. Meanwhile, the PPVC manual and industry code of practice provide detailed requirements for dimensions, installation, and safety for prefabricated and assembled buildings, ensuring a sufficient basis for automated generation and optimization of solutions. This approach ensures data integrity and accuracy when identifying and capturing building entity information (such as spatial location, geometry, and module connection methods), laying a solid foundation for subsequent generation of graphical data models and multi-objective optimization.

[0033] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0034] Reference Figure 1 and Figure 2 As shown, this embodiment provides a modular building generation and optimization method based on a BIM graphic data model, comprising the following steps: S101. Acquire spatial layout information of a modular building based on a building information model (BIM) to obtain model data, wherein the model data includes the spatial position, spatial area, connection relationship of each module, and dimensions of passages and doors between adjacent modules.

[0035] In this example, BIM model data for the building project is first obtained through BIM design software. This data includes basic information such as the building's geometry, the spatial location of each module, dimensions, door and window openings, wall thickness, and floor layout. This process uses a standardized BIM model interface to export the relevant data into a standard format for subsequent data processing.

[0036] like Figure 3 As shown, in this embodiment, the modular building specifically includes: a structural form (analysis model), a shear wall structure (specifically a tubular structure), a podium wall structure (specifically a supporting structure), cast-in-place components, and a volume module; The shear wall structure is a tubular structure, the pier wall structure is a supporting structure, the cast-in-place components include beams, columns, walls, slabs, stairs, cores and platforms, and the volume module includes prefabricated components.

[0037] This embodiment inputs basic building project requirement information, including project scale, functional requirements, space allocation, budget range, etc. This input information will serve as the basis for subsequent design plan generation. The system will generate a building design plan that meets the requirements based on this input, including the proposed expansion of existing entities and new entities in the MVD presentation, an information requirements flow chart, and identification of entities required for modular building design, which are obtained through the Prefabrication Specification (PPVC) Manual and the Code of Conduct Manual.

[0038] S102: Convert the extracted BIM model data into a graphic data model to generate graph data including spatial nodes and connection edges.

[0039] like Figure 4 As shown, in this embodiment, the specific generation process of the graphic data model is: Graph structure expansion: Based on the spatial parameters of the BIM model, the vertices and edges of the shear wall structure, tubular structure, and support structure are extracted to form an adjacency graph and a connectivity graph. The graph structure data is expanded by combining the topological relationships of the structural forms (such as the vertical connection between beams and columns, and the horizontal overlap between thick plates and walls) to obtain an expanded graph that includes the hierarchical relationships of the structural system. Modular Part Association: Extract the geometric boundaries of precast components (stairs, cores, platforms) and cast-in-place components (podium walls, volume modules) in the BIM model, define their topological relationships (such as the bolted connection between precast stairs and cast-in-place cores), and associate component types and connection methods with expansion diagrams. Matrix data construction: Extract the geometric information (B-rep model parameterized data) and semantic information (structural type, material properties) of the BIM model, and store the data in a matrix by component type. The shear wall structure and supporting structure correspond to the row and column indexes of the matrix; the weights between modules are dynamically updated according to the structural form (such as the circumferential connection of the tubular structure and the node articulation of the supporting structure).

[0040] Data extraction and graph generation: Based on matrix data and updated weights, geometric constraints (such as column spacing thresholds) and semantic constraints (such as prefabricated components cannot be cut) between components are extracted. Combined with the extended graph structure, a graph structure is generated that includes geometric accuracy, structural type, and construction logic. Model output: Generate a modular building graphic data model based on the graph structure, and its output form includes the combination relationship of the structural system (a hybrid system of shear walls + supporting structures) and the spatial distribution and connection node coordinates of prefabricated and cast-in-place components.

[0041] In this embodiment, each node in the graph data model represents a single unit of a modular building (such as a room or corridor). Node attributes include the unit's spatial dimensions, volume, and functional requirements. Edges represent the connections between modular units, and edge weights are defined based on factors such as doors, windows, passageways, and streamlines between modules. This step uses a custom algorithm to map building entities to the graph data model, ensuring that the graph data model accurately represents the structural information of the building space.

[0042] This embodiment converts imported BIM model data into a graphical data model through an automated conversion algorithm. During the conversion process, the system extracts not only the geometric dimensions and functional information of the building units, but also the connectivity and relative positions between building modules, allowing for subsequent automated design optimization.

[0043] In this step, the present invention further refines the module classification and connection information within the BIM model through IFC model view definitions, MVD entity extensions, and the prefabricated component standards specified in the PPVC manual. Specifically, the system automatically identifies different types of components, such as shear walls, pier walls, cast-in-place assemblies, and volumetric modules, extracts their geometric attributes and semantic information, and converts them into nodes and their topological relationships within a graphical data model. By appropriately expanding and filtering the IFC file (i.e., MVD extension), the present invention is able to extract more precise modular building attributes (such as prefabricated component lifting points and connectable nodes), thus avoiding the shortcomings of traditional BIM models in describing modular details.

[0044] S103. Based on the optimization goal, a multi-objective optimization method is used to optimize the graphic data model to obtain several alternative design solutions.

[0045] In this embodiment, the optimization objectives include maximizing space utilization, minimizing connection redundancy, and satisfying structural safety constraints; The multi-objective optimization algorithm uses the NSGA-II genetic algorithm, in which the selection probability of individuals in the algorithm population is based on non-dominated sorting and crowding distance, specifically: (1) in, represents the non-dominated ranking level of individual i (the smaller the level value, the higher the priority of the individual in the Pareto front, corresponding to the comprehensive advantages of the building scheme in terms of space utilization, connection redundancy and structural safety constraints), For individuals The crowding distance (measures the distribution density of adjacent individuals in the solution set, used to avoid excessive aggregation of optimization results leading to homogenization of modular solutions), is a minimum constant, The population size to set for the genetic algorithm.

[0046] In this embodiment, the specific optimization process is: Adjacency matrix definition: During the model construction phase, an adjacency matrix is ​​generated based on the spatial layout of the modular building. The matrix element values ​​are binary identifiers. If module i is adjacent to module j, the corresponding element value is 1, otherwise it is 0. Connection matrix establishment: During the model optimization phase, a connection matrix is ​​established, whose element values ​​are determined by the connection weights W between modules. ij It indicates that the weights are calculated based on the door and window sizes, streamline complexity, and structure type (shear wall / tubular structure) extracted in S102; Objective function construction: Based on the matrix data generated by S102, the NSGA-II algorithm is used to drive the Pareto frontier. The objective function is defined as: (2) Among them, in the formula, Represents the module extracted based on the B-rep model in step S102 Plane projected area, is the total building area, The minimum permissible area of ​​a single module as determined by the building code (by Function constraint module area discreteness); and Corresponding modules and The spatial position mapped by the vertex coordinates of the graph structure, The module spacing threshold is dynamically adjusted according to the structure type (for shear wall structure, , tubular structure , For the floor height, is the pipe diameter); Representation Module With adjacent modules The connection weight is calculated by multiplying the door and window area ratio and the streamline overlap ratio in step S102. is the minimum connection strength based on the structure type (0.6 for shear wall structure, 0.4 for support structure); function pass Nonlinear mapping constrains the module area discreteness to ensure And the distribution is uniform; the objective function The weighted combination of space utilization (area ratio weighted term) and spacing compliance (dynamic threshold constraint term) drives the compact arrangement of modules; the objective function Then the weak connection is punished by the ReLU function ( ), minimize redundant connections to improve the efficiency of structure-function coupling.

[0047] Combination of the three, relying on and The topological rules of the system are used to form a multi-objective optimization framework that takes into account regulatory compliance, spatial economy and structural robustness.

[0048] The attribute value of the node also includes the functional priority of the module, which is dynamically assigned according to preset rules.

[0049] In this embodiment, an optimization algorithm is used to automate the design of spatial layouts based on a constructed graphical data model. This optimization algorithm considers multiple design factors, including building functional zoning, lighting, ventilation, flow patterns, and fire escape routes. Designers can set specific design criteria based on project requirements (e.g., maximizing space efficiency and minimizing energy consumption), and the system automatically adjusts the layout of building units to achieve optimal spatial allocation and functional organization.

[0050] During the design process, the system automatically adjusts the modular building's design parameters based on input requirements and constraints (such as functional areas, circulation routes, and sound insulation requirements). These solutions may involve different building unit configurations, module layouts, and connection methods, ensuring that multiple design options can meet project needs.

[0051] Furthermore, in this embodiment, the constraints of the multi-objective optimization method specifically include: The module size constraint requires that the area of ​​each module should not be less than the preset minimum area value; The connection strength constraint stipulates that the connection weight between any two modules must not be lower than a minimum threshold dynamically set according to the building type; Passage safety constraints ensure that all channel dimensions meet preset safety standard values.

[0052] S104. Use the fitness evaluation model to score the alternative design schemes, convert the design scheme with the highest score into an operable BIM format file, and output it through the cloud to complete the design process.

[0053] The evaluation model in this embodiment includes building function assessment, space utilization, lighting and ventilation, structural stability, construction difficulty, etc. The evaluation score of each design plan will affect the final selection of the plan, and the system will automatically select the best plan or provide improvement suggestions.

[0054] Secondly, the feasibility of the selected design is verified to ensure its smooth implementation during actual construction. The system uses simulation to verify potential construction issues, such as construction time, labor costs, and material usage. Based on the verification results, the system automatically adjusts the design to improve construction efficiency and building quality.

[0055] Finally, the optimized design solution is exported as executable construction drawings and BIM model data to support subsequent construction and supervision. All design documents and model data are shared via the cloud, facilitating collaboration among different teams and ensuring a smooth construction process.

[0056] Specifically, this embodiment evaluates the connectivity (edge ​​weights) between modular units in the calculation graph. Based on the connectivity between building units, the system evaluates the feasibility of different design solutions, ensuring that the modules in the modular building design can achieve good functional coordination through effective connection. This step also verifies the operability and safety of the design solutions through simulation.

[0057] Using multiple optimization algorithms, different modular building design options are generated. Each option is evaluated based on design criteria and constraints, such as structural safety and ease of construction. The system provides a feasibility assessment report for each design option, allowing designers to select the option that best meets project requirements or combine and adjust multiple alternative options.

[0058] The optimized design is exported as executable construction drawings and BIM model data, facilitating execution and monitoring during the subsequent construction phase. The design is presented through a visual interface, allowing designers to clearly see design details such as spatial layout, functional divisions, and module connections, ensuring accurate implementation during the construction phase.

[0059] In this embodiment, the sub-scoring item definitions of the fitness evaluation model include: The space utilization score is the ratio of the sum of the areas of each module to the total building area; The connection efficiency score is the sum of all inter-module connection weights; The comprehensive fitness score is defined as the weighted sum of multiple objectives: (3) Among them, the space utilization score is the sum of the areas of each module Total construction area The ratio of represents the efficiency of module layout in utilizing building space; the connection efficiency score is obtained by summing the weights of connections between modules. Quantization, weight Based on dynamic calculation of door and window matching, flow line overlap rate and structure type (e.g. shear wall rigid connection weight is 1.2, tubular structure flexible connection weight is 0.8), comprehensive fitness score Defined as the multi-objective weighted sum ,in is the weight coefficient ( ), cost adaptability is determined by the total construction cost and budget cap The ratio constraint ensures that the scheme is optimized within the economic feasibility boundary. γ is the attenuation coefficient, which controls the influence of the distance between modules on the connection weight. The larger γ is, the more drastic the weight drop is when the distance is farther. ijIt is the actual distance between modules i and j (such as the geometric center distance), which is used to quantify the layout compactness. The closer the distance, the higher the connection efficiency score.

[0060] The constructability score is obtained through construction simulation calculation, and the parameter system of the construction simulation includes material cost budget parameters, construction schedule parameters and construction difficulty parameters.

[0061] In this embodiment, the cloud collaboration platform supports multiple users to modify the design plan in real time and synchronously update the BIM file through the version control algorithm.

[0062] like Figure 5 As shown, this embodiment also provides a modular building generation and optimization system based on the BIM graphic data model, including: a model data acquisition module 1, a graphic data model construction module 2, a model design module 3 and an evaluation module 4.

[0063] The model data acquisition module 1 is used to acquire the spatial layout information of the modular building based on the building information model (BIM) to obtain model data; By synchronizing data through the cloud platform, designers upload BIM model data to the cloud system for real-time sharing and collaboration across different devices. In this way, multiple teams can modify and optimize the design plan simultaneously and obtain the latest design data in real time.

[0064] The graphic data model construction module 2 is used to convert the extracted BIM model data into a graphic data model and generate graph data containing spatial nodes and connection edges; wherein, the system can automatically parse the building information in the BIM model and convert the data into a graphic data model according to the set rules.

[0065] Model Design Module 3 optimizes the graphical data model using a multi-objective optimization method based on the optimization objective, generating several design alternatives. Using the graphical data model, the system analyzes the layout of the building space and assesses the connectivity and functional compatibility between modules. The system automatically calculates parameters such as the spatial utilization of building units and the rationality of flow lines, and automatically adjusts the spatial layout based on design requirements.

[0066] On a cloud platform, the system automatically generates modular building plans using a generative design approach based on a graph data model. The system optimizes building spaces based on multi-dimensional data, generating multiple design alternatives and making real-time adjustments based on feedback from different teams.

[0067] The evaluation module 4 is used to score the alternative design schemes using the fitness evaluation model, and convert the design scheme with the highest score into an operable BIM format file, and output it through the cloud data to complete the design process.

[0068] Based on the design requirements and constraints entered by the user, the system automatically generates multiple modular building design solutions. These solutions include different spatial configurations, module combinations, and connection methods to meet the diverse needs of the project.

[0069] Each generated design scheme will be optimized through an automatic optimization algorithm. The system will evaluate the space utilization, functionality, constructability and other aspects of the design scheme, automatically select the optimal design scheme, and generate a feasibility report.

[0070] The optimized design plan will be output as standard construction drawings and BIM data to facilitate subsequent work by the construction and supervision teams.

[0071] In Evaluation Module 4, generated design proposals are centrally evaluated via a cloud platform. The system scores and evaluates each proposal based on multiple factors, including building functionality, layout rationality, and modular connectivity. These evaluation results are updated in real time, allowing designers to review and modify them from any location and device.

[0072] The system allows multiple teams to optimize design solutions in parallel. Each team can modify the design based on their expertise (such as structural design, spatial layout, construction plan, etc.) and share the modified results in real time on the cloud platform. Other teams can immediately see the latest changes and build on them.

[0073] After multiple rounds of optimization and evaluation, the system outputs the final modular building design and generates relevant construction drawings and BIM data. All design files are shared via the cloud, allowing the construction and supervision teams to access data and provide feedback during project implementation.

[0074] In the present invention, the MVD presentation allows for the extension of specific entities or attributes required for modular buildings on top of the IFC standard. For example, the mounting holes, hanging points, or reinforcement nodes reserved in the volume module can all be represented in a custom manner in the IFC file. The PPVC manual and code of conduct provide the system with design and construction constraints for prefabricated and assembled buildings, including requirements such as component specifications, material types, and safety passage dimensions. When acquiring BIM data, the system will prioritize checking these extended entities and rules, and parse the corresponding information into a graphical data model, so that subsequent multi-objective optimization can take into account the special needs of prefabricated and assembled buildings.

[0075] It can be seen that, on the first aspect, the present invention provides a method for automatic generation and optimization design of modular buildings based on a BIM graphic data model, the method comprising: a BIM-based graphic data model, a data conversion algorithm, and generative design; on the second aspect, the present invention greatly improves the feasibility and accuracy in prefabricated and assembled building scenarios by performing MVD extension on the IFC standard and combining PPVC and industry codes of conduct.

[0076] The core technology of this invention lies in converting building entities and their attributes in a Building Information Model (BIM) into a graph-structured data model. This model then utilizes an optimization algorithm to rapidly generate and screen optimal design solutions, addressing the drawbacks of traditional design methods, which suffer from low efficiency, high manual intervention, and difficulty meeting multiple constraints. BIM models, as digital representations in architectural design, contain multidimensional data such as a building's geometry, spatial location, and material information. However, traditional BIM models typically exist as static information, lacking dynamic automatic generation and optimization capabilities. This invention, by converting BIM models into graph data models, facilitates more flexible and efficient automated processing of building spaces. Specifically, this invention represents building spaces using nodes and edges in graph theory. Each node in the graph represents a modular building unit, such as a room, corridor, kitchen, or bathroom. The size of each node is related to the area or volume of the building unit, while edges represent the adjacency and connectivity between modules. The connections between modules are represented by edge weights, which depend on the size of the connections between modules, such as door and window openings and passageways, and their impact on the flow of the building space. The combination of nodes and edges forms a graph data model with structured spatial information, which can provide accurate spatial relationship information for subsequent automated design generation.

[0077] In order to realize the conversion of BIM model to graph data model, the present invention has developed an efficient data conversion algorithm. The algorithm can automatically extract the building entity information in the BIM model, including spatial position, geometric dimensions of building modules, connection mode between modules, etc. The algorithm first classifies the building elements in the BIM model according to predetermined rules, converts each building unit into a node in the graph, and converts the connection relationship between the nodes into an edge. Then, the algorithm calculates the adjacency matrix and connectivity matrix of the graph according to the set rules to ensure the connection relationship between each module and the coordination of spatial functions. These data conversion algorithms also take into account the actual needs of architectural design, and can automatically adjust the conversion rules according to the type and complexity of different architectural designs, and ensure that the generated graph data model meets the actual design needs. Through this automated conversion process, the complex data in the traditional BIM model can be effectively converted into graph structure data that is convenient for further optimization and analysis, solving the technical problem that the traditional BIM model cannot be directly used for design generation and optimization.

[0078] Based on the BIM graph data model, the present invention provides a generative design method that automatically generates multiple design solutions for modular buildings. This method fully utilizes the spatial information, module connectivity, and functional requirements in the BIM graph data model to automatically generate multiple feasible design solutions based on set design criteria and constraints. Design criteria may include spatial layout optimization, inter-module connectivity, functional differentiation, and comfort requirements, while design constraints involve aspects such as the building's structural safety, seismic requirements, and lighting and ventilation. The key to this generative design method lies in its ability to automatically generate design solutions. By inputting information such as building functional requirements and spatial layout requirements, the system utilizes the connectivity between nodes in the graph data model to perform spatial layout, taking into account inter-module compatibility, functional requirements, and ease of use. For example, the system can automatically adjust the position and connectivity of each module by setting objectives (such as maximizing space utilization or optimizing building circulation) to ensure that the final design solution meets all functional requirements and constraints.

[0079] During the design generation process, the system evaluates multiple design options based on an optimization algorithm and scores each option's feasibility and optimization using pre-set evaluation criteria. These criteria may include space efficiency, design rationality, and construction feasibility. Ultimately, the system selects the optimal design based on the evaluation results, which the designer can then select or further modify. This generative design approach significantly improves design efficiency and avoids subjective bias and redundancy that can occur during the design selection process.

[0080] In its second aspect, the present invention provides a prototype system for implementing a method for automatically generating and optimizing modular building designs based on a BIM (Building Information Model) diagram data model. To validate the present invention's technical solutions, the present invention also developed a prototype system capable of implementing the aforementioned generative design method within existing BIM design software. This system seamlessly integrates with BIM design software, automatically converting diagram data based on the BIM model, generating design solutions, and providing real-time design optimization recommendations.

[0081] The prototype system boasts high computational efficiency, capable of generating thousands of design alternatives in a short period of time to accommodate the complex demands of modular building design. Utilizing a Core i7-8550U CPU as its core computing unit, the system boasts powerful computing capabilities, ensuring efficient operation despite complex design tasks. Users can set design criteria and constraints based on project requirements, and the system automatically adjusts spatial layouts, optimizes module connections, and generates multiple design alternatives based on these settings. Designers can review and evaluate these alternatives through an intuitive visual interface, ultimately selecting the optimal solution or further revising the design.

[0082] It should be noted that for the aforementioned method embodiments, for simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present disclosure is not limited by the order of the actions described, because according to the present disclosure, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present disclosure.

[0083] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not limited herein.

[0084] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A modular building generation and optimization method based on BIM graphic data model, characterized in that: The following steps are involved: Acquiring spatial layout information of the modular building based on the building information model to obtain model data, wherein the model data includes the spatial position, spatial area, connection relationship of each module, and the dimensions of passages and doors between adjacent modules; Convert the extracted BIM model data into a graph data model to generate graph data containing spatial nodes and connecting edges; Based on the optimization goal, a multi-objective optimization method is used to optimize the graph data model and obtain several alternative design schemes; The fitness evaluation model is used to score the alternative design schemes, and the design scheme with the highest score is converted into an operable BIM format file and output through the cloud to complete the design process.

2. The modular building generation and optimization method based on the BIM graphic data model according to claim 1 is characterized in that: The modular building specifically includes structural forms, shear wall structures, podium wall structures, cast-in-place components, and volume modules; The shear wall structure is a tubular structure, the pier wall structure is a supporting structure, the cast-in-place components include beams, columns, walls, slabs, stairs, cores and platforms, and the volume module includes prefabricated components.

3. The modular building generation and optimization method based on BIM graphic data model according to claim 1 is characterized in that: The attributes of the nodes of the graphical data model include unit space size, volume and functional requirements; The edges of the graphical data model include connection relationships between module units, wherein the weights of the edges are determined according to doors, windows, passages and streamline factors between modules.

4. The modular building generation and optimization method based on the BIM graphic data model according to claim 1 is characterized in that: The specific generation process of the graphic data model is as follows: Extract the graphs of vertices and edges of the shear wall structure, tubular structure, and supporting structure to form an adjacency graph and a connectivity graph. Combined with the topological relationship of the modular building structure, the graph structure data is expanded to obtain an extended graph containing the hierarchical relationship of the structural system. Extract the geometric boundaries of precast and cast-in-place components in the BIM model, define their topological relationships, and associate component types, connection methods, and expansion diagrams; Extract the geometric and semantic information of the BIM model and store them in a matrix according to component type to obtain matrix data. Then calculate and update the weights between the B-rep model and modules. Extracting geometric constraints and semantic constraints between components based on the matrix data and the updated weights, and combining them with the extended graph to generate a graph structure including geometric accuracy, structural type, and construction logic; A modular building graphic data model is generated based on the graph structure, wherein the output form of the graphic data model includes the combination relationship of the structural system and the spatial distribution and connection node coordinates of prefabricated and cast-in-place components.

5. The modular building generation and optimization method based on BIM graphic data model according to claim 1 is characterized in that: The optimization objectives include: maximizing space utilization, minimizing connection redundancy, and satisfying structural safety constraints; The multi-objective optimization method adopts the NSGA-II genetic algorithm, in which the selection probability of individuals in the algorithm population is based on non-dominated sorting and crowding distance, specifically: ； in, Represents the non-dominated ranking level of individual i, where the smaller the level value, the higher the priority of the individual in the Pareto front, corresponding to the comprehensive advantages of the building scheme in space utilization, connection redundancy and structural safety constraints. For individuals The crowding distance measures the distribution density of adjacent individuals in the solution set, which is used to avoid excessive aggregation of optimization results and homogenization of modular solutions. is a minimum constant, The population size to set for the genetic algorithm.

6. The modular building generation and optimization method based on BIM graphic data model according to claim 1 is characterized in that: The process of optimizing the graphic data model using the multi-objective optimization method to obtain several alternative design solutions is as follows: Generate an adjacency matrix based on the spatial layout of the modular building, where the element value of the adjacency matrix is ​​a binary identifier. If module i is adjacent to module j, the corresponding element value is 1, otherwise it is 0; Establish a connection matrix, the element values ​​of which are represented by the inter-module connection weights Wij, wherein the connection weights are calculated based on the extracted door and window sizes, streamline complexity, and structure type; Based on the matrix data, the NSGA-II algorithm is used to drive the Pareto frontier, and the objective function is defined as: ； in, Represents the module extracted based on the B-rep model Plane projected area, is the total building area, is the minimum permissible area of ​​a single module as determined by the building code; and Corresponding modules and The spatial position mapped by the vertex coordinates of the graph structure, The module spacing threshold is dynamically adjusted according to the structure type; Representation Module With adjacent modules The connection weight is calculated by multiplying the door and window area ratio and the streamline overlap rate. is the minimum connection strength based on the structure type; and Represents the module set and the graph structure vertex set respectively, which are used to constrain the topological connection rules between modules. f1 and f2 are both objective functions. pass Nonlinear mapping constrains the module area discreteness to ensure And evenly distributed.

7. The modular building generation and optimization method based on BIM graphic data model according to claim 1 is characterized in that: The multi-objective optimization method is provided with constraints; The constraints include module size constraints, connection strength constraints, and traffic safety constraints, specifically: The module size constraint stipulates that the area of ​​each module shall not be less than a preset minimum area value; The connection strength constraint stipulates that the connection weight between any two modules is not less than a minimum threshold dynamically set according to the building type; The passage safety constraint stipulates that the channel size reaches a preset safety standard value.

8. The modular building generation and optimization method based on BIM graphic data model according to claim 1 is characterized in that: The fitness evaluation model includes sub-scoring items; the sub-scoring items include space utilization score, connection efficiency score, comprehensive fitness score and constructability score; The space utilization score is the ratio of the total module area to the total building area; The space utilization score is the sum of the connection weights between all modules; The comprehensive fitness score is defined as the weighted sum of multiple objectives; The constructability score is obtained through construction simulation calculation, wherein the parameter system of the construction simulation includes material cost budget parameters, construction schedule parameters and construction difficulty parameters.

9. The modular building generation and optimization method based on the BIM graphic data model according to claim 8 is characterized in that: The comprehensive fitness score is specifically: ； Among them, the comprehensive fitness score is the sum of the areas of each module Total construction area The ratio of , which reflects the efficiency of the module layout in utilizing the building space, and the connection efficiency score is obtained by summing the weights of the connections between modules. Quantization, weight Based on the dynamic calculation of door and window matching, traffic flow overlap rate and structure type, λ1, λ2, λ3 are weight coefficients, and , γ is the attenuation coefficient, dij is the actual distance between modules i and j, T cost is the total construction cost, T budget The budget cap.

10. A modular building generation and optimization system based on a BIM graphic data model, for implementing the modular building generation and optimization method based on a BIM graphic data model according to any one of claims 1 to 9, characterized in that: include: Model data acquisition module (1), graphic data model construction module (2), model design module (3) and evaluation module (4); The model data acquisition module (1) is used to acquire spatial layout information of a modular building based on a building information model to obtain model data; The graphic data model construction module (2) is used to convert the extracted BIM model data into a graphic data model, and generate graph data including spatial nodes and connection edges; The model design module (3) is used to optimize the graphic data model based on the optimization target using a multi-objective optimization method to obtain several alternative design solutions; The evaluation module (4) is used to score the alternative design schemes using a fitness evaluation model, and convert the design scheme with the highest score into an operable BIM format file, and output it through the cloud data to complete the design process.

Images