Multi-disciplinary collaborative optimization method for steel modular building structure based on improved MOEA / D algorithm

Abstract

The application discloses a steel modular building structure multi-specialty collaborative optimization method based on an improved MOEA / D algorithm and belongs to the technical field of physical calculation. In view of the problems of existing multi-specialty data fragmentation, index calculation relying on manual operation, time-consuming numerical simulation and strong subjectivity of optimal solution selection, the application firstly extracts geometric data based on drawings, fills and fuses information through professional coding, and constructs an integrated digital basis; further, building material cost is automatically calculated, structure performance indexes are extracted, and building cold and heat loads are calculated; then, isomorphic graph and heterogeneous graph data are constructed, a GCN structure agent model and a HGCN energy consumption agent model are trained; next, an optimization problem is constructed with material cost minimization and energy consumption minimization as optimization objectives and structure performance as constraints, an improved MOEA / D algorithm with a dynamic neighborhood strategy is adopted, optimization iteration is carried out in combination with the agent model, and a Pareto optimal solution set is output; and finally, an optimal design scheme is obtained through screening.

Description

Technical Field

[0001] This invention relates to the field of physical computing technology, and in particular to a multi-disciplinary collaborative optimization method for steel modular building structures based on an improved MOEA / D algorithm. Background Technology

[0002] With the deepening of industrialization in construction, steel modular building structures, due to their high degree of industrialization, systematization, and integration, have gradually become an important development direction in the construction industry. The design of such buildings requires comprehensive consideration of multiple conflicting indicators, such as building material costs, structural safety performance, and building energy consumption. Traditional design methods typically rely on repeated manual adjustments, which are inefficient and make it difficult to obtain a globally optimal solution.

[0003] In recent years, multidisciplinary collaborative optimization technology has been widely used in architectural design. By constructing an optimization mathematical model and combining iterative calculations with structural analysis software and energy consumption simulation software, a set of Pareto optimal solutions can be generated, providing designers with multiple trade-off options. However, existing technologies still have the following problems in practical engineering applications: First, there is a lack of unified data support from the design source to the optimization model. Data from multiple disciplines is fragmented, and index calculations rely on manual methods, resulting in low efficiency in optimization model construction and poor data consistency. Second, the optimization process requires repeated calls to finite element analysis and energy consumption simulation, which is computationally costly and time-consuming, making it difficult to support large-scale iterations. Third, after obtaining the Pareto solution set, there is a lack of an objective and quantitative decision-making mechanism to select the optimal solution for the final construction drawing design. Usually, it relies on the subjective experience of designers or only selects the extreme solution of a single objective, making it difficult to guarantee the comprehensive performance of the final solution.

[0004] The above problems are interconnected and urgently require an intelligent method that can run through the entire process of data integration, indicator calculation, optimization modeling, iterative solution and optimal decision-making, so as to realize the automated and integrated selection from design drawings to the optimal solution. Summary of the Invention

[0005] This invention provides a multi-disciplinary collaborative optimization method for steel modular building structures based on an improved MOEA / D algorithm. It aims to solve the technical problems in the existing multi-disciplinary collaborative optimization design of steel modular building structures, which are difficult to construct optimization models, have low solution efficiency, and are subject to strong subjectivity in the selection of the final scheme and are difficult to guarantee the overall performance due to the lack of a unified data foundation, reliance on manual calculation of indicators, low optimization iteration efficiency, and lack of objective decision-making mechanism.

[0006] To address the aforementioned technical problems, this invention provides a multi-disciplinary collaborative optimization method for steel modular building structures based on an improved MOEA / D algorithm, comprising the following steps:

[0007] E1. Extract geometric data from DXF drawings, fill in the code and information fusion of the three disciplines of architecture, structure and energy consumption, and construct integrated digital basic JSON data;

[0008] E2. Based on the integrated digital foundation JSON data, automatically calculate the cost of building materials, automatically build a structural finite element analysis model and extract structural performance indicators, and automatically build an energy consumption analysis model and calculate the building's heating and cooling loads.

[0009] E3. Based on the integrated digital foundation JSON data, construct homogeneous graph data and heterogeneous graph data, and use the structural performance index and building heating and cooling load calculated in step E2 as supervision signals to construct and train a structural proxy model based on graph convolutional network and an energy consumption proxy model based on heterogeneous graph convolutional network, respectively. The structural proxy model is used to predict whether the structural performance exceeds the limit, and the energy consumption proxy model is used to predict the building heating and cooling load.

[0010] E4. Using the minimization of building material costs and building heating and cooling loads calculated in step E2 as optimization objectives, and the structural performance indicators extracted in step E2 as constraints, structural components and enclosure structures are selected as core design variables to construct the optimization problem.

[0011] E5. The improved MOEA / D algorithm is used to perform optimization iteration to solve the optimization problem. The improved MOEA / D algorithm introduces a dynamic neighborhood strategy and calls the structural proxy model and energy consumption proxy model constructed in step E3 to replace finite element analysis and energy consumption simulation for rapid prediction during the iteration process, and outputs the Pareto optimal solution set.

[0012] E6. Filter the Pareto optimal solution set to obtain the optimal design variables.

[0013] This invention provides a multi-disciplinary collaborative optimization method for steel modular building structures based on an improved MOEA / D algorithm. Step E1 constructs an integrated digital foundation, using DXF drawings as the sole input. Through three-disciplinary coding and tree-structure integration, it breaks down the barriers of fragmented multi-disciplinary data and solves the problem of relying on manual integration for optimization model construction. Step E2 automatically completes material cost accounting, structural performance extraction, and building thermal load simulation based on this foundation, achieving end-to-end automated calculation from data to underlying indicators, significantly improving modeling and evaluation efficiency. Step E3 constructs isomorphic and heteromorphic graph data, trains the GCN structural proxy model and the HGCN energy consumption proxy model, and uses the graph... Neural networks accurately capture the heterogeneous relationship between structural topology and energy consumption, replacing traditional time-consuming finite element analysis and energy consumption simulation, effectively solving the technical bottleneck of excessively long numerical simulation time in optimization iteration. By constructing a multi-disciplinary collaborative optimization problem through steps E4 to E6, an improved MOEA / D algorithm with dynamic neighborhood strategy is adopted and combined with a surrogate model to assist optimization. Finally, the Pareto optimal solution set is objectively screened using the TOPSIS ideal solution method, overcoming the shortcomings of traditional methods that rely on subjective experience to select solutions. This forms a complete technical closed loop from data integration, intelligent optimization to optimal decision-making, systematically solving the problems of difficulty in multi-disciplinary collaboration, low computational efficiency, and strong subjectivity in optimization decision-making in existing technologies. Attached Figure Description

[0014] Figure 1 A flowchart illustrating a multi-disciplinary collaborative optimization method for steel modular building structures based on an improved MOEA / D algorithm, provided in an embodiment of the present invention.

[0015] Figure 2 This is a schematic diagram of the parameterized filling process provided in an embodiment of the present invention;

[0016] Figure 3 This is a schematic diagram of the initial IFC model provided in an embodiment of the present invention. Figure 3 (a) is the initial IFC model corresponding to a standard floor building drawing. Figure 3 (b) is the initial IFC model corresponding to another standard floor building drawing;

[0017] Figure 4 This is a schematic diagram of the collision-corrected IFC model provided in an embodiment of the present invention. Figure 4 (a) is the modified IFC model corresponding to the first-floor architectural drawings. Figure 4 (b) is the modified IFC model corresponding to the two- to four-story building drawings;

[0018] Figure 5 This is a tree structure fusion hierarchy framework diagram provided in an embodiment of the present invention;

[0019] Figure 6An example of an embodiment of the present invention is a year-round outdoor temperature curve of a building.

[0020] Figure 7 This is a year-round temperature curve of a certain hot zone provided in an embodiment of the present invention;

[0021] Figure 8 This is a heat load curve diagram of a certain hot zone provided in an embodiment of the present invention;

[0022] Figure 9 This is a cooling load curve diagram inside a certain hot zone provided in an embodiment of the present invention;

[0023] Figure 10 A schematic diagram of structural performance indicators provided for embodiments of the present invention;

[0024] Figure 11 This is a flowchart of the MOEA / DI-GNN algorithm provided in an embodiment of the present invention;

[0025] Figure 12 The test set prediction and actual value fitting curve of the HGCN energy consumption proxy model provided in the embodiment of the present invention;

[0026] Figure 13 A comparison chart of the Pareto fronts of various optimization algorithms provided in the embodiments of the present invention. Detailed Implementation

[0027] To make the technical problem to be solved, the technical solution, and the beneficial effects of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and examples. It should be understood that the specific examples described herein are merely illustrative and not intended to limit the scope of this application.

[0028] Specifically, such as Figure 1 As shown in the flowchart, this embodiment provides a multi-disciplinary collaborative optimization method for steel modular building structures based on an improved MOEA / D algorithm, which includes the following steps:

[0029] E1. Constructing an integrated digital foundation: Extracting geometric data from DXF (Drawing Exchange Format) drawings, filling in the data with codes from the three disciplines of architecture, structure, and energy consumption, and integrating the information to construct an integrated digital foundation JSON (JavaScript Object Notation) data;

[0030] E2. Automatic Calculation of Multiple Professional Indicators: Based on the integrated digital foundation JSON data, automatically calculate the cost of building materials, automatically build a structural finite element analysis model and extract structural performance indicators, and automatically build an energy consumption analysis model and calculate the building's heating and cooling loads.

[0031] E3. Construct a graph neural network proxy model for prediction: Based on the integrated digital foundation JSON data, construct homogeneous graph data and heterogeneous graph data, and use the structural performance index and building heating and cooling load calculated in step E2 as supervision signals to construct a structural proxy model based on a graph convolutional network and an energy consumption proxy model based on a heterogeneous graph convolutional network and train them respectively. The structural proxy model is used to predict whether the structural performance exceeds the limit, and the energy consumption proxy model is used to predict the building heating and cooling load.

[0032] E4. Construction of Multidisciplinary Collaborative Optimization Problem: Taking the minimization of building material costs and building heating and cooling loads calculated in step E2 as optimization objectives, and the structural performance indicators extracted in step E2 as constraints, structural components and enclosure structures are selected as core design variables to construct the optimization problem.

[0033] E5. Iterative solution of optimization problem: The improved MOEA / D algorithm is used to perform iterative optimization solution of the optimization problem. The improved MOEA / D algorithm introduces a dynamic neighborhood strategy and calls the structural proxy model and energy consumption proxy model constructed in step E3 to replace finite element analysis and energy consumption simulation for rapid prediction during the iteration process, and outputs Pareto optimal solution set.

[0034] E6. Selecting the optimal design variables: Construct a decision matrix for the objective function values ​​(building material cost, building heating and cooling load) of each design scheme in the Pareto optimal solution set, use vector normalization to eliminate the influence of dimensions, and determine the positive ideal solution and the negative ideal solution; calculate the Euclidean distance and relative proximity of each scheme to the positive and negative ideal solutions, and determine the scheme with the largest relative proximity as the design scheme with the best overall performance and output it.

[0035] This invention provides a multi-disciplinary collaborative optimization method for steel modular building structures based on an improved MOEA / D algorithm. Step E1 constructs an integrated digital foundation, using DXF drawings as the sole input. Through three-disciplinary coding and tree-structure integration, it breaks down the barriers of fragmented multi-disciplinary data and solves the problem of relying on manual integration for optimization model construction. Step E2 automatically completes material cost accounting, structural performance extraction, and building thermal load simulation based on this foundation, achieving end-to-end automated calculation from data to underlying indicators, significantly improving modeling and evaluation efficiency. Step E3 constructs isomorphic and heteromorphic graph data, trains the GCN structural proxy model and the HGCN energy consumption proxy model, and uses the graph... Neural networks accurately capture the heterogeneous relationship between structural topology and energy consumption, replacing traditional time-consuming finite element analysis and energy consumption simulation, effectively solving the technical bottleneck of excessively long numerical simulation time in optimization iteration. By constructing a multi-disciplinary collaborative optimization problem through steps E4 to E6, an improved MOEA / D algorithm with dynamic neighborhood strategy is adopted and combined with a surrogate model to assist optimization. Finally, the Pareto optimal solution set is objectively screened using the TOPSIS ideal solution method, overcoming the shortcomings of traditional methods that rely on subjective experience to select solutions. This forms a complete technical closed loop from data integration, intelligent optimization to optimal decision-making, systematically solving the problems of difficulty in multi-disciplinary collaboration, low computational efficiency, and strong subjectivity in optimization decision-making in existing technologies.

[0036] (1) Step E1: Building an integrated digital infrastructure

[0037] Step E1 specifically includes the following steps:

[0038] E11. Drawing Information Extraction and Processing: Input DXF drawing, extract geometric data of grid, walls, doors and windows, module frames and hot zone frames according to line type layer based on Dxfgrabber library, calculate center line coordinates and endpoints, and output initial component data in JSON format;

[0039] E12, Multi-disciplinary Information Processing: The initial JSON data is processed by three disciplines: the architecture discipline performs 19-bit encoding, parameter filling, collision correction and component classification; the structural discipline calculates room loads and fills in the mechanical materials of components; the energy consumption discipline performs 23-bit encoding, space identification and enclosure structure filling, and outputs JSON data for each discipline.

[0040] E13. Multi-disciplinary information integration: Based on the five principles of cross-disciplinary integration, multi-level, classification, master-slave dependency, and coding collaboration, a tree structure integration method is adopted to integrate the JSON data of the three disciplines according to the hierarchy and output integrated digital basic JSON data.

[0041] The integrated digital infrastructure built in this step effectively breaks down data barriers between multiple disciplines, realizes intensive data management and efficient collaboration, solves the problems of poor data reusability and low correlation, verifies the feasibility of data extraction, processing and fusion methods, and lays a solid foundation for the full life cycle collaboration and design optimization of steel modular building structures.

[0042] Step E11 can be divided into the following sub-steps:

[0043] E111. Obtain the floor plan file of the steel module building structure as the initial file;

[0044] E112. Preprocess the initial file;

[0045] E113. Extract and process drawing information from the preprocessed initial file.

[0046] In step E111, to ensure the efficient transfer of multidisciplinary information extracted from the drawings to subsequent multidisciplinary information processing and fusion stages, a unified data format needs to be defined. This embodiment selects JavaScript Object Notation (JSON) as the data format for integrating multidisciplinary information in steel modular building structures.

[0047] This embodiment uses the CAD (Computer-Aided Design) floor plan file of the steel module building structure as the initial file. All kinds of building components (walls, columns, doors and windows, dimensions, etc.) are stored in line layers according to industry standards. Therefore, the extraction method based on vector graphics is adopted to directly extract the structured data of the building drawings.

[0048] Currently, there are two main formats for CAD drawings: binary format with the .DWG extension and ASCII format with the .DXF extension. Because the encoding and decoding methods are completely open source, all data elements in the original drawing can be read without loss, accurately preserving all structured information such as geometric parameters, spatial coordinates, and layer classifications of building components. This aligns with the requirements of this embodiment for extraction accuracy, convenience, and data integrity. Therefore, this embodiment extracts building-related data based on CAD drawings in .DXF format.

[0049] In step E112, the drawings are preprocessed to facilitate subsequent information extraction. The drawings need to be divided according to standard floors. For example, the drawings for the first floor and floors 2-4 represent two standard floors, so two standard floor DXF vector drawings are created. Furthermore, the line types on the drawings need to be standardized: the grid lines should use the "DOTE" line type, walls the "WALL" line type, doors the "DOOR" line type, windows the "WINDOW" line type, and text the "PUB_TEXT" line type. This embodiment optimizes the layout based on a defined module layout. To facilitate subsequent information extraction, defined module frames are added to the steel module building structure DXF drawings, using the "MODULE" line type. To facilitate thermal zone information extraction, defined thermal zone frames are added, using the "ZONE" line type.

[0050] In step E113, Dxfgrabber is a lightweight DXF graphics file parsing library developed using the Python programming language. Its core function is to quickly read, parse, and extract data from DXF format files generated by professional design software such as AutoCAD, without relying on the runtime environment of commercial design software like AutoCAD. It is one of the commonly used tools for processing DXF vector files in the open-source field. This embodiment uses the Dxfgrabber library based on the Python programming language to automatically extract component information from DXF vector drawings, including grid line extraction, door and window line extraction and processing, wall line extraction and processing, module layout information, and thermal distribution information.

[0051] In the multi-disciplinary information processing in step E12, this embodiment specifically divides the information into three categories: architectural, structural, and energy consumption information, and processes each category of information using different processes.

[0052] Architectural information is the core of integrated data. This part of information processing includes four parts: component coding, architectural design information filling, module information processing, and component information filling.

[0053] 1) Component Coding: In this embodiment, architectural data is encoded into 19-digit long numeric codes. A hierarchical segmented interpretation rule is adopted to define the core attributes of the components one by one, thereby achieving accurate identification of steel module integrated building components. The coding rules are shown in Table 1:

[0054] Table 1. Architectural Professional Data Coding Rules

[0055]

[0056] For example, the code 1010101010101010101 represents the first column in the southwest direction of the first geometric type of column component in the first module of the first module in the first standard floor of the steel modular building structure.

[0057] 2) Architectural design information: Architectural design information is a comprehensive description of the building, including the building name, building area, building height, number of floors, etc. This part of the information is directly entered into the JSON field.

[0058] 3) Module Information Processing: The information extracted from the drawings only includes module frames and wall / door / window information. Further wall line cutting is required according to the module layout diagram to obtain module data information, including module location and the geometric position information of internal wall / door / window components. Since modules are produced according to standardized procedures, they are matched and classified according to module data. Modules with the same or equivalent geometric features are grouped into the same category, totaling 6 equivalent geometric cases. For example, mirror images of each other belong to the same category, and rotated versions of each other belong to the same category of features.

[0059] 4) Component information filling:

[0060] 1. Assignment of parameterized information and preliminary calculation

[0061] The information extracted from the drawings mainly includes the length and width dimensions of the modules, and the centerline information of walls, doors, and windows. Further parametric filling is required, primarily involving the cross-sections, materials, geometric information, and location distribution parameters of walls, doors, windows, and other components (floors, roofs, etc., filled according to actual needs). For the parametric data, the starting point position, stretching length, and stretching direction of each specific component are calculated based on predefined module construction rules and component distribution locations. Then, specific information about the component is assigned based on its cross-sectional shape, cross-sectional dimensions, material type, material name, and whether it has openings. The process is as follows: Figure 2 As shown.

[0062] 2. Collision Correction

[0063] Parametric calculations only preliminarily determine the positions of components. Since the physical relationships between components are not yet clearly established, collision problems may occur, thus failing to effectively guide actual production and construction. To solve this critical problem, this embodiment first performs systematic collision detection on the model to accurately identify collision areas between components. Subsequently, based on the detection results, the geometric parameters and spatial positions of the building components are adjusted to achieve adaptive adjustment of the model, ultimately generating conflict-free building data information that meets engineering requirements.

[0064] This embodiment employs an axis-aligned bounding box (AABB) collision detection algorithm. This algorithm creates an axis-aligned bounding box for each component and calculates the minimum and maximum coordinates of each bounding box. By projecting the bounding boxes of two components onto three coordinate planes, if the projections on all three coordinate axes overlap, the two components are determined to have collided.

[0065] To accelerate detection, a recursive method is used to construct a BVH (Bounding Volume Hierarchy) tree. This organizes the AABB bounding boxes of all components in the module into a tree structure based on parent-child hierarchies. Upper-level nodes are "large bounding boxes" containing multiple lower-level nodes, and lower-level nodes are "small bounding boxes" containing a single object. During collision detection, it first checks if the upper-level large bounding boxes intersect. If they do not intersect, the detection of all lower-level small bounding boxes is skipped, thus improving detection efficiency.

[0066] IfcOpenShell is a free and open-source code library that provides analysis and creation functions. It can be used to process industrial basic IFC files, parse and retrieve relevant data within them, and perform tasks such as reading, creating, and modifying IFC files. This embodiment uses the Python-based IfcOpenShell library to map the previously obtained initial building JSON information to IFC components, resulting in an initial IFC model as follows: Figure 3 As shown, Figure 3 (a) Figure 3 (b) shows the initial IFC models corresponding to one standard floor building drawing (floor 1) and another standard floor building drawing (floors 2-4), respectively.

[0067] After obtaining the initial solid model, the Geom function of the IfcOpenShell library is used to extract geometric mesh data from the IFC components to obtain the triangular facet information of the components. Then, AABB bounding boxes are constructed to achieve collision detection. After collision detection, the model returns detailed collision information, including the accurate location of the collision and related dimensional data. Based on this information, an adaptive adjustment method is proposed. This method consists of two stages. In the first stage, components are grouped according to their categories, and the starting point position, cross-sectional dimensions, and extension length of components such as columns, beams, walls, doors, roofs, and floors are adjusted respectively, and the module data is integrated and updated. In the second stage, collision calculations are performed between the adjusted walls and the original door and window IFC component entities, and door and window opening information is added to the walls. After the two-stage adaptive adjustment, JSON data is obtained. The collision-corrected IFC model based on the IfcOpenShell library mapping is shown below. Figure 4 As shown, Figure 4 (a) Figure 4 (b) shows the modified IFC models corresponding to the 1st floor building drawings and the 2nd to 4th floor building drawings, respectively.

[0068] 3. Component Classification

[0069] To classify the geometric features of building components within the module, this embodiment first extracts the three-dimensional vertex coordinates of the components using the Geom function in the IfcOpenShell library. Based on this, it iterates through the equivalent geometric data of the components' X, Y, and Z axes, including their mirror images and rotations. For each type of geometric data, the SHA256 algorithm is used to calculate a hash value. Finally, the minimum hash value among the components themselves and all their mirror images is taken as the component's geometric fingerprint. Components with the same hash calculation value are classified as belonging to the same category.

[0070] Structural information processing mainly includes three aspects: filling in structural design information, calculating module loads, and filling in structural material information for components.

[0071] 1) Structural design information filling: This part of the information is a comprehensive description of the building's structural characteristics, mainly including basic structural design information such as site category, seismic intensity, and functional room dead and live coefficients. This information is entered directly into JSON fields.

[0072] 2) Module load calculation: To calculate the load of the module, room identification is required in the architectural drawings. In this embodiment, a counterclockwise room outline detection method is used. This method first identifies the text and text coordinates in the grid area, emits rays to the left based on the coordinates, takes the first intersecting line as the initial outline, and searches for the next outline in a counterclockwise direction from the endpoint to the starting point. This process is repeated until the initial outline is found, thereby determining the room outline.

[0073] Since a module may contain multiple rooms, or a room may contain multiple modules, we first traverse the rooms to find the modules they belong to. For rooms that contain multiple modules, we traverse the modules to find the modules contained in the room, thereby determining the room type and area range contained in each module.

[0074] For different room types, the corresponding dead and live load parameters are searched using the functional room dead and live load factor table. To calculate the area of ​​each room, this embodiment uses the Polygon.area function based on the Shapely library to calculate the area of ​​the enclosed region. Based on the weighted calculation of the room area and the room dead and live load factor, the average dead and live load factor of the ground is calculated using the trapezoidal triangle load distribution method to calculate the edge load.

[0075] 3) Material information for structural components: The mechanical properties of components are key to structural information. Therefore, it is necessary to supplement the structural properties of structural components, such as columns and beams, with information such as Poisson's ratio and modulus of elasticity.

[0076] Energy consumption professional information processing mainly includes five aspects: energy consumption information coding, energy consumption condition information filling, spatial identification, thermal zone spatial classification, and building envelope material filling.

[0077] 1) Energy Consumption Information Encoding: In this embodiment, energy consumption data is encoded into a 23-bit long numeric code. A hierarchical segmented interpretation rule is adopted to define the core attributes of the object bit by bit, thereby achieving accurate identification of energy consumption information objects. The encoding rules are shown in Table 2 below.

[0078] Table 2 Energy Consumption Professional Data Coding Rules

[0079]

[0080] For example, the code 10101020101010101010101 indicates the first enclosure structure in the south direction of the first hot zone in the first hot zone in the first space in the first hot zone in the first hot zone in the first space in the first space in the first type of wall in the first type of wall in the first standard floor of the modular building.

[0081] 2) Energy consumption condition information filling: This part describes the information required for building energy consumption calculation, mainly including altitude, annual average temperature, city location, latitude and longitude, and other information required for energy consumption design. This information is directly entered as a JSON field.

[0082] 3) Spatial Identification: To simplify the spatial division of modular buildings, this embodiment defines the module space as the smallest space. Whether a module is an independent space is determined by whether it has walls in all four directions (north, south, east, and west). If it is an independent space, then that module is considered the space. For modules with fewer than four wall locations, the room to which the module belongs is queried, and that room is ultimately considered the space.

[0083] 4) Spatial Classification of Hot Zones: To classify and integrate hot zones, the hot zones are compared based on the module component information corresponding to each space within the hot zone. Hot zones with equivalent geometric characteristics belong to the same category. Similarly, for each space within the same hot zone, the module information corresponding to the space is compared, and spaces with equivalent geometric characteristics are grouped into the same category.

[0084] 5) Envelope Information Filling: Envelope information is the core of energy consumption. In the energy consumption calculation model, the main envelope structures are five types of building envelope structures: roof, ground, wall, door, and window. In this embodiment, combined with the standardized production characteristics of modular buildings, the materials of the envelope structure of the whole building are set in a consistent manner, and the materials and thermal performance are assigned according to the material names of each envelope structure corresponding to the module in the building profession.

[0085] In step E13, cross-disciplinary integration is the basic prerequisite for data integration. The core objective is to break down professional barriers and achieve unified integration and collaborative sharing of information from various disciplines.

[0086] This embodiment focuses on the integration of building models, structural mechanics models, and building energy consumption models. The integration principles include:

[0087] Cross-disciplinary integration: The mechanical performance information of structural components is integrated one-to-one with the building components in the architectural field that have completely overlapping geometric shapes and spatial positions; the thermal zone spatial division and thermal performance information of the building envelope are integrated in parallel with the data of the architectural and structural fields.

[0088] Multi-level integration: For architectural and structural engineering, integration is carried out in three levels: standard layer, module, and component; for energy consumption, integration is carried out in three levels: thermal zone, space, and building envelope.

[0089] Category-based fusion: Standard floor data is fused to achieve floor classification; for architecture or structural engineering, modules are classified according to the geometric distribution information of internal components, and components are classified according to their functional categories and geometric information; for energy consumption engineering, thermal zones are classified according to the spatial distribution of internal thermal zones and the information of the building envelope, spaces are classified according to the distribution of components within each space, and building envelopes are classified according to their types and geometric information.

[0090] Master-slave dependency integration: The overall building dimensions, module layout, and room functions are defined as the master model data; the geometric section parameters of components and the energy consumption model and building envelope materials are defined as slave model data, so that the slave model data is stored and managed in dependence on the corresponding master model data.

[0091] Coding Collaboration and Integration: A unified coding system is established, with building type identification, standard floor, and floor information as common components, professional categories, modules, hot zones, and spatial information as meso-level coding, and component and enclosure structure information as micro-level coding.

[0092] In the data from the architecture, structure, and energy consumption disciplines, the building type identifier, standard floor, and floor information are consistent, so they are designed as common areas. In subsequent information from each discipline, building and structural information is assigned based on components, while energy consumption information is assigned based on thermal zone space envelope, etc. Therefore, a tree structure is used for information fusion, with its hierarchical framework as follows: Figure 5 As shown.

[0093] Step E1 completes the construction of the integrated digital infrastructure, providing data support for the automatic calculation of indicators. Based on this infrastructure, the following steps will realize the automatic calculation of three major categories of indicators: economic, energy consumption, and structural indicators of the integrated steel structure module.

[0094] (2) Step E2: Automatic calculation of multiple professional indicators

[0095] Step E2 specifically includes:

[0096] E21. Automatic calculation of economic indicators: Based on the module classification characteristics of integrated digital foundation JSON data, a step-by-step accumulation method from module level, standard level to building level is adopted to multiply the volume and unit price of structural components, enclosure structural components and other fixed components and sum them to calculate the total cost of building materials.

[0097] E22. Automatic calculation of building energy consumption indicators: Extract the spatial geometry information of the thermal zone and the thermal parameters of the building envelope from the integrated digital foundation JSON data, automatically generate IDF files through Eppy mapping, call EnergyPlus to perform building heat balance simulation, and calculate the sum of the annual cumulative heating load and cooling load.

[0098] E23. Automatic Calculation of Structural Indicators: Extract geometric information and mechanical parameters of structural components from integrated digital foundation JSON data, automatically generate node elements, apply boundary conditions and seismic loads in OpenSees, solve node displacements and component internal forces through finite element analysis, and extract inter-story drift angles and component strength, stiffness and stability indices.

[0099] It should be noted that there is no restriction on the execution order of E21, E22, and E23.

[0100] In step E21, material cost, as a core component of the total construction cost, accounts for a significant proportion of the total construction price. Accurate calculation of material cost is not only a fundamental prerequisite for economic analysis of construction but also directly reflects the economic rationality of material input in construction. Therefore, this embodiment selects the material cost of building components as the core economic calculation indicator. Based on the production characteristics of steel modular building structures, the total material cost of its components covers three categories: structural components, enclosing structural components, and other fixed components. This embodiment calculates the cost of all components uniformly based on volume. That is, the total material cost of structural components in a steel modular building is equal to the sum of the total cost of structural components (columns, beams, longitudinal beams), the total cost of enclosing structural components (walls, doors, windows, roofs, floor slabs), and the total cost of other fixed components (corner fittings, connectors, etc.).

[0101] In step E22, the building's heating and cooling load is essentially the theoretical heating and cooling capacity required to maintain a stable indoor thermal environment. Its calculation is based on unsteady-state heat balance theory and building envelope heat transfer theory, reflecting only the building's thermal performance and not involving the energy efficiency of air conditioning equipment or system losses. For any building thermal zone, at any given time... The building's heating and cooling load equals the building's total heat gain minus the building's total heat loss.

[0102] The total heat gain consists of solar radiation heat gain from the building envelope (determined by solar radiation intensity, solar radiation absorption coefficient of the building envelope, and shading coefficient), internal heat gain (heat gain from personnel, lighting, equipment, etc.), and air infiltration heat gain (calculated from indoor and outdoor temperature difference, infiltration air volume, and specific heat of air). The total heat loss is the sum of heat loss from building envelope transfer (calculated using the conduction transfer function method) and heat loss from air infiltration (calculated using the same formula as air infiltration heat gain, only the direction of substitution for temperature difference is reversed).

[0103] Under controlled indoor temperature conditions, in order to maintain the set heating temperature With cooling set temperature The hourly heating load and cooling load calculation expressions are as follows:

[0104] (1),

[0105] In formula (1): air density; The specific heat of air at constant pressure; For the volume of the hot zone; For time step; For a moment Heating load is only output when the indoor temperature is lower than the set heating temperature; For a moment Cooling load is only output when the indoor temperature is higher than the set cooling temperature; For a moment Indoor air temperature.

[0106] The hourly loads are accumulated over the entire year to obtain the annual heating load, annual cooling load, and total annual heating and cooling load. During the IDF design process, data such as the annual outdoor temperature, indoor temperature, and heating and cooling load of hot zones can be output. Visualization is performed using the Matplotlib library within the Python programming language environment. The example building's annual outdoor temperature is shown below. Figure 6 As shown in the analysis results of a certain thermal region, the annual temperature inside the thermal region is as follows: Figure 7 As shown. The heat load results within the hot zone are as follows. Figure 8 As shown. The cooling load results within the hot zone are as follows. Figure 9 As shown.

[0107] In step E23, for steel structure integrated buildings, this embodiment uses the finite element method to solve for the structural response. The essence of calculating the static response of a structure is to solve the finite element equilibrium equations. Its core logic is based on the displacement method, using nodal displacements as basic unknowns to construct the force equilibrium relationship. For a linear elastic structure, the overall equilibrium equation can be expressed as: , The overall stiffness matrix of the structure. The nodal displacement vector; This represents the nodal load vector.

[0108] For the steel modular building structure in this embodiment, the base shear method in the "Code for Seismic Design of Buildings" is used to calculate the horizontal seismic action. This method is applicable to buildings with a height of [missing information]. For regular structures where shear deformation is dominant, the core principle is to convert seismic forces into equivalent horizontal loads using the seismic influence coefficient. The specific calculation process is as follows: The equivalent total gravity load of the structure is the basis for seismic force calculation, obtained by weighting the representative gravity load values ​​of each floor. Horizontal seismic influence coefficient. Reflecting the coupling relationship between seismic intensity and structural dynamic characteristics, the calculation is performed in segments according to the period:

[0109] (2),

[0110] In formula (2): This represents the maximum value of the horizontal earthquake influence coefficient. The fundamental natural period of the structure; For the period of time; , This is the damping adjustment coefficient. The structural damping ratio; This is the decay index.

[0111] Standard value of total horizontal seismic action When the natural period of the structure When this is the case, additional seismic action at the top needs to be considered to correct for deviations in the high-rise response. , Add a coefficient to the top ( , ).

[0112] The standard value of horizontal seismic action for each floor is weighted by floor weight and height.

[0113] Finally, the calculated floor loads are distributed to specific nodes, ultimately forming the load vectors in the equilibrium equations. This completes the coupling between seismic load and finite element solution.

[0114] When designing modular integrated buildings, the safety and reliability of the structure must be considered. Because the structural load-bearing frame of a modular building mainly consists of modular beams, modular columns, and corner fittings connecting them, it is necessary to verify the performance indicators of bending members (beams) and bending-compression members (columns). According to the "Steel Structure Design Standard," the verification of components mainly considers three important indicators: strength, stiffness, and stability (overall stability and local stability). Whether the entire structure can meet the requirements under load is determined by the inter-story drift angle. Structural performance indicators include... Figure 10 As shown.

[0115] Steps E21, E22, and E23 use the integrated digital foundation for steel modular building structures established in step E1 as a unified data source. They focus on the automated calculation of three core professional indicators: building economics, energy consumption, and structure, and conduct systematic research. This opens up the channel between the digital foundation data and the calculation of various professional indicators, effectively solving the industry pain points of cumbersome traditional indicator calculation processes, low efficiency, disconnect between professional data and large amounts of manual intervention. This provides solid technical support and data guarantee for the digital analysis, performance evaluation and subsequent optimization design of steel modular building structures.

[0116] (3) Step E3: Construct a graph neural network agent model

[0117] Graph structures are the core data carriers of graph neural network models, and their mathematical definition is a triple. ,in Represents a set of nodes. Denotes the set of edges. Represents the node feature matrix ( For the number of nodes, , where R represents the set of real numbers (where R is the node feature dimension). Based on the differences in the types of nodes and edges, graph structures can be divided into two main categories: isomorphic graphs and heteromorphic graphs. The core difference between the two lies in whether there is a type distinction between nodes and edges.

[0118] A homogeneous graph is a graph structure containing only one type of node and one type of edge. All nodes have a unified feature dimension and physical meaning, and all edges have the same association logic. It is suitable for scenarios where node attributes are uniform and the association relationships are simple. The adjacency matrix of an isogeneous graph... Satisfying symmetry, that is ,in Represents a node With nodes There is a connection. This indicates no connection relationship. To preserve the characteristics of each node, self-loops are usually added to the adjacency matrix, resulting in an adjacency matrix with self-loops. ( (as the identity matrix), and through the degree matrix (used for normalizing the adjacency matrix) To avoid excessively large feature values, normalization is performed to prevent numerical deviations during feature propagation.

[0119] A heterogeneous graph is a graph structure containing multiple node types and edge types. Different node types have different feature dimensions and physical meanings, and different edge types correspond to different association logics (such as inclusion, connection, and belonging). It is suitable for scenarios with diverse node types and complex association logics. The mathematical representation of a heterogeneous graph needs to be extended to quadruples. ,in These represent the set of node types and the set of edge types, respectively. Unlike isomorphic graphs, heterogeneous graphs require independent feature processing and propagation rules for each type of node and edge. By distinguishing different types of neighborhood information, more accurate feature extraction can be achieved.

[0120] Graph Neural Networks (GNNs) are a class of deep learning models specifically designed for processing graph-structured data. Their core idea stems from the fusion of traditional graph theory and neural networks. They aim to learn the feature representations of nodes and the global structure in a graph by simulating information interaction and propagation between nodes, thus solving the learning challenges of non-Euclidean space data. GNNs are widely used in fields such as social networks, recommender systems, bioinformatics, and architectural engineering.

[0121] The core framework of GNN can be summarized as a neighborhood aggregation-feature propagation loop. Its basic process is as follows: In the initial state, each node only contains its own original features; through multi-layer aggregation operations, each node merges the features of its neighboring nodes to achieve iterative feature updates; after multiple rounds of propagation, the features of each node not only contain its own attributes, but also integrate global graph structure information, which can ultimately be used for downstream classification, regression and other tasks at the node level, edge level or graph level.

[0122] Graph Convolutional Networks (GCNs) are the most classic and widely used specific models in GNNs. Their core idea is to use neighborhood aggregation and feature propagation to allow each node to integrate its own features with the features of neighboring nodes, gradually generating an embedding vector that can represent the global graph structure.

[0123] The basic form of the inter-layer feature update formula for GCN is:

[0124] (3),

[0125] In formula (3): For the first The node feature matrix of the layer, the initial layer (Original node features); For the first The learnable weight matrix of the layer is used to map the feature dimensions of the nodes; Non-linear activation functions (such as ReLU) are used to introduce non-linear features and improve the expressive power of the model.

[0126] GCN uses multi-layer convolutional operations to gradually mine the first-order, second-order, and even higher-order neighborhood information of nodes, ultimately realizing the transformation of node features into global graph features, and adapting to various downstream tasks such as classification and regression.

[0127] To address the two core requirements of structural performance evaluation and energy consumption calculation in building design optimization, this embodiment uses Graph Convolutional Networks (GCN) to construct a structural analysis proxy model, which is suitable for feature extraction and binary classification tasks of homogeneous graph data; and uses Heterogeneous Graph Convolutional Networks (HGCN) to construct an energy consumption analysis proxy model, which is suitable for regression prediction tasks with multiple types of nodes and complex relationships. This ensures that the two types of proxy models can accurately and efficiently replace traditional numerical simulations and improve optimization efficiency.

[0128] The essence of a building structure is a whole composed of various components connected by specific relationships; therefore, a structural system can be abstracted as an isomorphic diagram. ,in For a set of nodes, The specific construction rules for the edge set are as follows:

[0129] ① Node definition: A node is a single component in a building structure. Each node corresponds to a structural member. A single node is denoted as _____. ( , (Total number of components). The core function of a node is to carry the component's own attribute information, providing a foundation for subsequent feature extraction.

[0130] ② Node Feature Construction: This embodiment constructs node features based on three aspects: basic component attributes, spatial pose, and cross-sectional dimensions. Basic attributes mainly include component type, cross-sectional type, material density, and tensile length; spatial pose mainly includes the three-dimensional coordinates (3D) of the starting point and the Euler angles (3D) of the spatial direction; cross-sectional dimensions mainly include five-dimensional cross-sectional dimension parameters, including cross-sectional length, width, wall thickness, and two-dimensional custom parameters to adapt to diverse cross-sectional shapes. To eliminate the influence of dimensional differences on model training, all features are normalized.

[0131] ③ Edge Relationship Construction: An edge represents a physical connection between two structural components. Edge construction uses the AABB algorithm, which determines the connection state between components by checking if their bounding boxes intersect, thus establishing a bidirectional connection. The specific steps are as follows: For each structural component... Construct an AABB bounding box whose boundary is determined by the extreme values ​​of the component's three-dimensional coordinates, denoted as . (min represents the minimum, max represents the maximum); for any two components and ( If the formula is satisfied:

[0132] (4),

[0133] Then determine the component and A connection exists between the two; establish a bidirectional edge between them. If an edge is created, its weight is set to 1; otherwise, no edge is created, and its weight is set to 0. The final adjacency matrix is ​​denoted as... ,in Representation of components and Connected, This indicates that they are not connected.

[0134] To construct the graph data for the steel modular building structure, the building classification data of the integrated digital foundation is expanded. First, the classification module information of the standard floors needs to be expanded. For a certain type of module, the module's location information and internal component information are calculated based on the transformation matrix of other modules. Second, the information for each floor of the standard floors is calculated to obtain a full model JSON file. Graph nodes are then built based on the full component JSON file. Using the Ifcopenshell library, the overall building JSON information is mapped to the IFC model and the component connection relationships are determined. The connection relationships are written into the full component JSON, and the graph node and edge relationships are built based on the full component JSON file.

[0135] ④ Prediction task setting: The prediction task of the structural analysis proxy model is set as binary prediction of structural performance. The core objective is to determine whether the key performance data of the building structure exceeds the preset limit, so as to quickly determine whether the structural performance index is feasible or not.

[0136] This embodiment sets the prediction categories to 10 dimensions, namely: whether the inter-story drift angle in the X-direction of the building exceeds the limit; whether the inter-story drift angle in the Y-direction of the building exceeds the limit; whether the bending strength of the beam exceeds the limit; whether the shear strength of the beam exceeds the limit; whether the deflection of the beam exceeds the limit; whether the overall stability of the beam exceeds the limit; whether the strength of the column exceeds the limit; whether the slenderness ratio of the column exceeds the limit; whether the in-plane stability of the column exceeds the limit; and whether the out-of-plane stability of the column exceeds the limit. Let the 10-dimensional structural performance data be... Each dimension corresponds to a different structural performance index; the preset limit vector is... , For the first Permissible limits for dimensional performance indicators ( The binary classification prediction rule is as follows: for each performance index... ,like If the performance of that dimension is satisfactory, the label 0 is output; otherwise... If the performance in that dimension is deemed unsatisfactory, label 1 is output. The final model output is a 10-dimensional binary classification label vector. ,in It is used to characterize the qualified status of each structural performance index, providing a direct basis for structural design optimization.

[0137] During model training, binary cross-entropy (BCE) is used as the basic loss function.

[0138] The construction of heterogeneous graphs for building energy consumption systems follows the principle of "multi-type nodes + targeted edge relationships." This embodiment abstracts the energy consumption system as a heterogeneous graph. ,in For a set of heterogeneous nodes, For a set of heterogeneous edges, A collection of node types This is a collection of edge types, and the specific construction rules are as follows:

[0139] ① Node Definition: Based on the constituent units of the building energy consumption system, five types of heterogeneous nodes are classified, namely, hot zone nodes (… ), spatial nodes ( ), Enclosing structure nodes ( ), ground nodes ( ), outdoor nodes ( ), covering the core elements of energy consumption analysis.

[0140] ② Node Feature Construction: Since this embodiment predicts the building's heating and cooling loads and uses ideal cooling and heating equipment, energy consumption is only related to the building's performance itself. The features of the hot zone are defined as: standard floor number, coordinate position (3D), and volume; the features of the spatial nodes are defined as: floor number, coordinate position (3D), and volume; the features of the building envelope nodes are defined as: coordinate position (3D), surface area, thermal conductivity, density, specific heat capacity, thickness, solar radiation absorptivity, infrared emissivity, visible light transmittance (for windows), and boundary type; the features of the ground are defined as soil thermal conductivity, soil density, soil specific heat capacity, solar radiation absorptivity, and infrared emissivity; and the features of the outdoors are defined as air thermal conductivity, air density, air specific heat capacity, solar radiation absorptivity, and infrared emissivity.

[0141] ③ Edge Relationship Construction: The edge relationships of heterogeneous graphs are constructed based on the actual physical connections and spatial logic between various types of nodes. Combining the five types of nodes defined above—hot zone, space, enclosure structure, outdoor, and ground—with the calculated connection relationships of space, hot zone, enclosure structure, outdoor, and ground, a bidirectional edge relationship definition is adopted to fully characterize the interaction mechanism between nodes. The core of the edge relationship originates from the essential connection between nodes, specifically: hot zone contains space relationship, space contains enclosure structure relationship, enclosure structure connects to outdoor relationship, enclosure structure connects to ground relationship, enclosure structure connects to enclosure structure relationship, enclosure structure connects to hot zone relationship, and the reverse relationship of the above relationships, totaling 12 connection relationships.

[0142] The mathematical expression for edge relations can be uniformly described as follows: , Represents the set of edges in a heterogeneous graph. Indicates the first A set of edge relationships of different classes; each type of edge relationship is stored in the form of a two-dimensional tensor, i.e. ,in For the source node index set, The target node index set ensures that the model can accurately locate the relationships between nodes through edge indexes.

[0143] ④ Prediction Task Setting: The prediction task of the energy consumption analysis proxy model is set as the regression prediction of the total building heating and cooling load. The core objective is to accurately quantify the total building heating and cooling load over a certain period of time, providing a quantitative basis for building energy consumption optimization. The model output is one-dimensional continuous energy consumption data, i.e., the total building heating and cooling load. It covers the sum of the building's winter heating load and summer cooling load.

[0144] During model training, mean squared error (MSE) is used as the loss function to measure the deviation between the model's predicted values ​​and the actual simulation values. By minimizing this loss function, the HGCN model can fully learn the impact of various node characteristics and relationships on energy consumption, achieving accurate quantitative prediction of building heating and cooling loads, replacing the traditional and cumbersome energy consumption simulation process.

[0145] (4) Step E4: Multi-objective intelligent optimization

[0146] This embodiment focuses on temperate regions, selecting structural components and the building envelope as core design variables. It aims to achieve reasonable building costs, ensure the structural mechanical performance of the building, and simultaneously address energy consumption control requirements. In temperate regions, the building envelope is typically composed of thermally inert, high-density components. Different thicknesses and constructions of the envelope result in varying loads on the structure, indirectly affecting the selection of column and beam sections. This determines the cost of the envelope and, consequently, the overall structural cost, while also influencing the building's thermal load. Generally, increasing the thickness of the envelope leads to a lower heat transfer coefficient and lower building energy consumption, but also increases the load, resulting in higher costs for both the envelope and the structure. Designers must find the optimal balance between reducing material costs and reducing building energy consumption, while also meeting constraints such as inter-story drift angles and component performance. The core of this optimization problem lies in how to achieve comprehensive optimization of economy and energy consumption by rationally designing the geometric parameters and thermal performance of modular columns, beams, and the building envelope, while ensuring structural performance. Therefore, the multi-disciplinary collaborative optimization problem of steel modular integrated buildings is a typical multi-disciplinary collaborative optimization problem, which can be described as follows:

[0147] (5),

[0148] In equation (5): For n design variables ( Composed of ) Dimensional design variables; For the construction cost function, It is an energy consumption function; Here, m represents the i-th performance constraint specified in the design specifications, and m represents the total number of performance constraints. and The first Design variables The upper and lower boundaries are defined by find, min, and st, which represent finding, minimizing, and satisfying, respectively.

[0149] This optimization problem focuses on steel modular building structures with a given modular layout, using component dimensions and material parameters as the adjustment objects for optimization design. Its design variables mainly include the cross-sectional dimensions and materials of columns, beams, and longitudinal beams in various modules that affect structural performance; the number of variables dynamically changes with the building module type. Simultaneously, the design variables also cover the dimensions and materials of walls, doors, windows, roofs, and floors that affect building energy consumption. Other components and influencing factors, such as occupancy density and lighting start / stop times, are treated as constants.

[0150] To facilitate the selection of engineering components, a discretization selection scheme is adopted for all variables: columns and beams use fixed steel types, with box-section steel cross-section dimensions as discrete design variables; walls, doors, roofs, and floors maintain consistent material construction, with thickness as the basis for discrete selection; windows are selected according to different heat transfer coefficients corresponding to their structural forms, forming a discretization selection set. The design variables are shown in Table 3.

[0151] Table 3 Multidisciplinary Collaborative Optimization Design Variables

[0152]

[0153] For the optimization requirements of modular steel building structures (MiC), the construction cost function The smaller the better. This embodiment introduces a constraint penalty mechanism to transform the constrained optimization problem into an unconstrained optimization problem, defining the minimum material cost function considering structural performance constraints as:

[0154] (6),

[0155] In formula (6): For the material cost of the steel modular building structure, refer to step E2 for the calculation method. This is the penalty value when a single constraint is not satisfied; The number of individuals that do not meet the structural performance constraints includes the requirements stipulated in the "Steel Structure Design Standard" such as the inter-story drift angle limit, component strength, stiffness, and overall stability. To simplify the constraint dimensions, the maximum value of each dimension index among all components is used to determine the constraint conditions. The calculation method for each index is the same as in step E2.

[0156] The smaller the value, the better the material economy of the steel modular building structure while satisfying or minimizing violations of structural performance constraints. When all structural performance constraints are satisfied... ,at this time With material costs The values ​​are exactly the same.

[0157] The objective of minimizing building heating and cooling loads is expressed as:

[0158] (7),

[0159] In equation (7): This is the sum of the building's heating and cooling loads for the steel modular building structure. The calculation method is the same as step E2.

[0160] (5) E5: Iterative solution of optimization problems

[0161] MOEA / D, based on the idea of ​​decomposition, breaks down a multi-disciplinary collaborative optimization problem into a set of single-objective optimization subproblems. Through cooperation among neighboring subproblems, the optimal solution of the decomposed single-objective subproblem is essentially a solution in the Pareto solution set of the original multi-objective problem.

[0162] The MOEA / D algorithm generates Different weight vectors Decompose the multi-objective problem into There are several single-objective subproblems, with each weight vector corresponding to one subproblem. Within each single-objective subproblem, a comprehensive utility function represents the objective function of that single-objective optimization problem.

[0163] The MOEA / D algorithm posits that neighboring subproblems are similar and can co-evolve through cooperation, thus introducing the concept of neighborhood. For a given subproblem... its neighborhood This represents the set of subproblems most closely related to this subproblem, where the close relationship between subproblems can be measured by the Euclidean distance between their respective weight vectors.

[0164] In MOEA / D, each generation of the population consists of the current optimal solutions to each subproblem. When optimizing each subproblem, the algorithm only uses information from neighboring subproblems. When updating the population, for the For each subproblem, solutions to two neighboring subproblems are randomly selected and cross-mutated to generate a new solution. The objective function value of the new solution is calculated, and the solutions to each subproblem within the neighborhood are compared and replaced. This design reduces redundancy in information exchange and computational complexity, while allowing adjacent subproblems to share high-quality solutions, achieving co-evolution and driving the entire population towards the Pareto optimal front.

[0165] This embodiment addresses the problems of uneven solution set distribution and easy trapping in local optima in traditional MOEA / D algorithms for discrete optimization problems. It introduces a dynamic neighborhood strategy mechanism to construct an improved MOEA / D algorithm, MOEA / DI, to balance algorithm convergence and solution set diversity. The core of this mechanism is that the neighborhood size decreases linearly with the iteration progress, starting from the initial maximum neighborhood... Gradually reduce to the final minimum neighborhood The core calculation formula for replacing the traditional MOEA / D design with a fixed neighborhood size is:

[0166] (8),

[0167] In equation (8): Let be the size of the current neighborhood in generation gen, and gen be the current iteration generation. Let be the total number of iterations in the algorithm, `max()` be the maximum value function, and `round()` be the rounding function. This mechanism is designed to align with the phased optimization requirements of evolutionary algorithms: "early exploration, later utilization." In the early iterations, a large neighborhood is used to enhance global exploration, broadly covering the multidimensional discrete solution space to avoid local optima. In the later iterations, a small neighborhood is used to enhance local utilization, finely adjusting discrete parameters to improve solution accuracy. The implementation logic involves pre-calculating the maximum neighborhood of each individual based on the Euclidean distance of the weight vectors. During each iteration, the neighborhood size is dynamically adjusted according to the current progress, and a corresponding number of individuals are selected from the pre-calculated neighborhood as the current neighborhood. Parent selection and offspring updates are both based on this dynamic neighborhood, achieving a dynamic balance between exploration and utilization.

[0168] In engineering optimization design, while finite element analysis (FEM) methods offer high accuracy, they are computationally expensive and time-consuming, especially in high-dimensional parameter spaces or complex nonlinear problems. In surrogate model-based optimization design, constructing a high-accuracy surrogate model typically requires a large number of sample points. Furthermore, the surrogate model approximates the real physical model, and its accuracy is limited by the quality and quantity of training data. Relying solely on surrogate models for structural optimization may lead to a "precision paradox"—even with 95% accuracy, the remaining 5% error can have catastrophic consequences. Therefore, FEM-based optimization methods are "accurate but not fast," while surrogate model-based optimization methods are "fast but not accurate." This embodiment employs a surrogate model-assisted multi-disciplinary collaborative optimization strategy to develop the MOEA / DI-GNN algorithm. In the initial optimization phase, Latin hypercube sampling is used to generate... An initial batch of samples was used to perform energy consumption and structural simulation calculations. This batch of samples was then used to train an initial surrogate model, ensuring the model possessed basic fitting capabilities. Based on this, a model with a scale of [missing information] was generated. The initial population is used to initiate multi-disciplinary collaborative optimization iterations. The algorithm iterates to a specified number of generations. Then, based on historical samples, the initial surrogate model is incrementally updated to continuously improve its prediction accuracy. Next, based on the real population in the MOEA / D algorithm, multi-generation rapid optimization is performed using the updated surrogate model. The optimization results output by the surrogate model are then alternately replaced with those from the real population to generate offspring, and iterative optimization continues based on these offspring. This "real optimization—model update—surrogate optimization—population update" cycle is repeated until the algorithm reaches the preset maximum number of iterations, at which point it terminates, completing the global optimization process. The algorithm flow is as follows: Figure 11 As shown.

[0169] (6) Step E6: Determine the optimal solution

[0170] In multidisciplinary collaborative optimization, the Pareto Front identifies a set of cost-effective design solutions. Solutions on the Pareto Front are all non-dominated, meaning no single solution is superior to another across all objective functions. Additional decision preferences or criteria are needed to select the final optimal solution from the Pareto Front. This embodiment employs the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) to select the optimal solution. Its core principle is to use geometric distance to measure the closeness of each solution to the optimal solution.

[0171] (7) Experimental Analysis

[0172] Taking a four-story school dormitory building as an example, this building adopts a steel modular construction structure. The building comprises two standard floors and is located in a certain city. According to meteorological data, the coldest month in this city is January, with an average temperature of 7.8℃. There are approximately 10 days with an average daily temperature ≤5℃, classifying it as a temperate region. The main structure of this steel modular building is designed for a service life of 50 years. The building's safety level is Level II, its seismic fortification category is Class C, the site category is Class II, the seismic fortification intensity is 8 degrees, and the ground roughness is Class B. This steel modular building structure has two standard floors: one floor and floors 2-4.

[0173] Information was extracted and preprocessed from the architectural drawings of each standard floor. After equivalent feature matching and classification of the obtained module information, a total of 22 module categories were obtained. Information on additional components such as columns, beams, roofs, and floor slabs was defined through manual interaction to fill in the information of the integrated digital foundation. The final result is a 4314KB JSON file, representing the IFC model of the module within each standard floor.

[0174] Compared to traditional manual BIM modeling, structural modeling, and energy consumption modeling based on architectural drawings, this integrated digital infrastructure integrates and expresses data from various disciplines, and categorizes and integrates information to suit the standardized production characteristics of modular buildings, resulting in a significant advantage in file size. As shown in Table 4 below, the total size of the architectural IFC model, the HDF5 model file of the structure in OpenSees, and the IDF file of energy consumption is 32707KB, while the size of the integrated digital infrastructure proposed in this embodiment is only 4314KB, saving 86.81% of storage space.

[0175] Table 4 Manual Modeling and Digital Foundation File Sizes

[0176]

[0177] The integrated digital foundation is a digital representation of steel modular building structures. Regarding the accuracy verification of the digital foundation, this embodiment systematically compares the building material costs, finite element model analysis results, and energy consumption model analysis results obtained based on the integrated digital foundation with the calculation results of traditional manual modeling to verify the accuracy and reliability of the proposed method. The root mean square error of variation (CV(RMSE)) is used as the core indicator in this accuracy evaluation.

[0178] The material cost, energy consumption, and maximum inter-story drift angle of the structural analysis are compared, as shown in Table 5 below.

[0179] Table 5. Results of manual and automatic calculations

[0180]

[0181] As shown in Table 5, the CV (RMSE) values ​​for material cost, finite element model analysis results, and energy consumption model analysis results are all 0. This result indicates that the proposed method for calculating indicators based on an integrated digital foundation is consistent with the results of manual modeling calculations. Therefore, it can be concluded that the integrated digital foundation framework proposed in this embodiment can meet the professional accuracy requirements achievable through manual model building and can serve as an effective alternative to manual modeling calculations.

[0182] Regarding the overall efficiency of the integrated digital infrastructure, the total time to complete the entire process is only 623 seconds. In contrast, the traditional manual method requires engineers to separately build a building BIM model, a structural OpenSees model, and write an energy consumption IDF file, all of which involve manual data interaction and setup, accumulating approximately 10 hours of time. By comparison, the integrated digital infrastructure improves the overall efficiency of professional modeling and performance analysis by approximately 60 times, achieving a reduction from hours to minutes.

[0183] To improve the prediction accuracy and generalization ability of the graph neural network surrogate model, this embodiment employs Bayesian optimization to optimize key parameters (learning rate, hidden layer dimension, dropout rate, etc.) of the structural surrogate model and the energy-intensive surrogate model. Bayesian optimization constructs a probabilistic surrogate model of parameters and model performance, iteratively selecting the parameter combination most likely to improve performance. Compared to traditional grid search and random search, it significantly reduces the computational cost of parameter tuning while maintaining global optimization capabilities. The final optimized GNN surrogate model hyperparameter settings are shown in Table 6.

[0184] Table 6 Proxy Model Parameter Design Table

[0185]

[0186] In the improved MOEAD algorithm, to ensure a balance between computational efficiency and optimization performance, after multiple preliminary experiments, the population size N was set to 100 individuals. This value guarantees population diversity while controlling the reasonable consumption of computational resources. In the initial stage of the algorithm, 1000 Latin hypercube samples were used to train the graph neural network (GNN) surrogate model. This order of magnitude of samples allows the surrogate model to initially achieve relatively good predictive ability. The number of optimization iterations was set to 50, with a surrogate model update every 10 generations and iterative optimization based on the surrogate model. The fine-tuning learning rate of the surrogate model was set to 0.00001, allowing the model's predictive ability to steadily improve. The number of iterations for the surrogate model was set to 20 generations, allowing for sufficient exploration of the solution space. The polynomial crossover and mutation probabilities were configured to 0.8 and 0.1, respectively. This relatively high crossover rate maintains stable population evolution; the relatively low mutation rate helps the algorithm maintain a certain level of exploration ability without excessively destroying the obtained excellent gene combinations.

[0187] For structural compliance classification prediction, after initial training based on LHS (Latin Hypercube Sampling) samples and iterative updates using the improved MOEA / D algorithm, the confusion matrix of the GCN model was obtained. This performance validation used 2000 labeled structural samples as the test set, of which 1754 were actually compliant and 246 were actually non-compliant. In the confusion matrix, the number of samples that were actually compliant and correctly predicted as compliant by the model (True Negative TN) was 1727, the number of samples that were actually compliant but misclassified as non-compliant by the model (False Positive FP) was 27, the number of samples that were actually non-compliant but missed being classified as compliant by the model (False Negative FN) was 28, and the number of samples that were actually non-compliant but correctly predicted as non-compliant by the model (True Positive TP) was 218. Based on the above confusion matrix statistics, the core performance indicators of the model were calculated as follows.

[0188] Recall: Measured by the proportion of samples correctly identified by the model within a given true category. The recall rate of non-compliant samples directly reflects the model's ability to control the underreporting of structural safety hazards. The formula for calculating the recall rate of the compliant class is: Compliant Class Recall = Number of True Negative Samples ÷ (Number of True Negative Samples + Number of False Positive Samples). Substituting the data, we get: 1727 ÷ (1727 + 27) ≈ 98.5%. The formula for calculating the recall rate of the non-compliant class is: Non-compliant Class Recall = Number of True Positive Samples ÷ (Number of True Positive Samples + Number of False Negative Samples). Substituting the data, we get: 218 ÷ (218 + 28) ≈ 88.6%.

[0189] Precision: This measures the proportion of samples that the model predicts as belonging to a certain category, reflecting the model's false positive rate. The formula for calculating the precision of the qualified class is: Qualified Class Precision = Number of True Negative Samples ÷ (Number of True Negative Samples + Number of False Negative Samples). Substituting the data, we get: 1727 ÷ (1727 + 28) ≈ 98.4%. The formula for calculating the precision of the unqualified class is: Unqualified Class Precision = Number of True Positive Samples ÷ (Number of True Positive Samples + Number of False Positive Samples). Substituting the data, we get: 218 ÷ (218 + 27) ≈ 88.9%.

[0190] Overall accuracy: This measures the proportion of correct predictions among all model predictions. The formula is: Overall accuracy = (Number of true positive samples + Number of true negative samples) ÷ Total number of test samples. Substituting the data, we get: (218 + 1727) ÷ 2000 = 1945 ÷ 2000 ≈ 97.3%.

[0191] The calculation results above show that the GCN model has an overall accuracy of up to 97.3%.

[0192] For building heating and cooling load regression prediction, the fitting curves between the predicted and actual values ​​of the HGCN surrogate model on the test set are as follows: Figure 12 As shown. From Figure 12 It can be seen that the GCN model also has high prediction accuracy.

[0193] To compare the optimization performance of different multidisciplinary collaborative optimization methods, under the same design constraints, the Pareto fronts obtained by MOEA / DI-GNN, MOEA / DI (without surrogate model updates), MOEA / D (without dynamic neighborhood improvement), and the traditional multidisciplinary collaborative optimization algorithms NSGA2 and MOPSO were compared. In all different multidisciplinary collaborative optimization methods, the population was set to 100, and the number of iterations was 50. The final Pareto solution sets of each algorithm are shown below. Figure 13 As shown. By Figure 13 It can be seen that the Pareto front obtained by the surrogate model-assisted optimization method is more advanced and more uniformly distributed in the target space than other algorithms, which means that the optimization algorithm can effectively explore the target space.

[0194] Based on the above method, this embodiment also provides a multi-disciplinary collaborative optimization system for steel modular building structures based on the improved MOEA / D algorithm. The key features are: an integrated digital foundation construction module, a multi-disciplinary index automatic calculation module, a graph neural network surrogate model construction module, a multi-disciplinary collaborative optimization problem construction module, an optimization problem iterative solution module, and an optimal solution selection module, which are respectively used to execute steps E1, E2, E3, E4, E5, and E6 in the multi-disciplinary collaborative optimization method for steel modular building structures based on the improved MOEA / D algorithm.

[0195] Furthermore, this embodiment also provides a computer-readable storage medium storing a computer program, the key point of which is that when the program is executed by a processor, it implements the multi-disciplinary collaborative optimization method for steel modular building structures based on the improved MOEA / D algorithm or the multi-disciplinary collaborative optimization system for steel integrated modules based on the improved MOEA / D algorithm.

[0196] It should be noted that, through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to implement the methods or systems described in the various examples or some parts of the examples.

[0197] In summary, this invention constructs a complete technical closed loop from data integration, index calculation, surrogate model construction, intelligent optimization to optimal decision-making through steps E1 to E6. Experimental results show that the integrated digital foundation JSON data constructed by this invention can save 86.81% of storage space compared with traditional manual modeling. The automatic calculation results of material cost, energy consumption index, and structural performance are almost completely consistent with manual modeling, and the overall efficiency of multi-disciplinary modeling and performance analysis is improved by about 60 times. The accuracy of the trained GCN structural surrogate model reaches 97.3%, and the predicted values ​​of the HGCN energy consumption surrogate model are highly fitted with the actual values, which can reliably replace the traditional time-consuming finite element analysis and energy consumption simulation. The MOEA / D algorithm improved by the dynamic neighborhood strategy combined with the surrogate model-assisted optimization generates a better Pareto front distribution with better convergence. Finally, the optimal solution is objectively selected through the TOPSIS ideal solution method, overcoming the shortcomings of traditional methods that rely on subjective experience. This invention systematically solves the problems of fragmented data from multiple disciplines, reliance on manual calculation of indicators, excessive time consumption in numerical simulation, and strong subjectivity in the selection of optimal solutions in existing technologies, providing complete technical support for the efficient, accurate, and integrated design of steel modular building structures.

[0198] The above are merely preferred examples of this application and are not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application shall be included within the scope of protection of this application.

Claims

1. A multi-disciplinary collaborative optimization method for steel modular building structures based on an improved MOEA / D algorithm, characterized in that, Including the following steps: E1. Extract geometric data from DXF drawings, fill in the code and information fusion of the three disciplines of architecture, structure and energy consumption, and construct integrated digital basic JSON data; E2. Based on the integrated digital foundation JSON data, automatically calculate the cost of building materials, automatically build a structural finite element analysis model and extract structural performance indicators, and automatically build an energy consumption analysis model and calculate the building's heating and cooling loads. E3. Based on the integrated digital foundation JSON data, construct homogeneous graph data and heterogeneous graph data, and use the structural performance index and building heating and cooling load calculated in step E2 as supervision signals to construct and train a structural proxy model based on graph convolutional network and an energy consumption proxy model based on heterogeneous graph convolutional network, respectively. The structural proxy model is used to predict whether the structural performance exceeds the limit, and the energy consumption proxy model is used to predict the building heating and cooling load. E4. Using the minimization of building material costs and building heating and cooling loads calculated in step E2 as optimization objectives, and the structural performance indicators extracted in step E2 as constraints, structural components and enclosure structures are selected as core design variables to construct the optimization problem. E5. The improved MOEA / D algorithm is used to perform optimization iteration to solve the optimization problem. The improved MOEA / D algorithm introduces a dynamic neighborhood strategy and calls the structural proxy model and energy consumption proxy model constructed in step E3 to replace finite element analysis and energy consumption simulation for rapid prediction during the iteration process, and outputs the Pareto optimal solution set. Step E5 specifically includes: The Latin hypercube sampling method was used to generate initial samples. Structural finite element analysis and energy consumption simulation were performed on the initial samples to obtain structural performance indicators and building heating and cooling loads, and an initial sample set was constructed. The initial proxy model is obtained by training the structural proxy model and the energy consumption proxy model using the initial sample set; An initial population is generated, and multi-disciplinary collaborative optimization iterations are initiated. A dynamic neighborhood strategy is introduced into the MOEA / D algorithm, where the neighborhood size decreases linearly with the iteration progress. The current neighborhood size in generation gen is: ， Where gen is the current iteration number. Let be the total number of iterations in the algorithm, max() be the function to find the maximum value, and round() be the rounding function. For the initial maximum neighborhood, It is the smallest neighborhood; Each iteration to a preset number of generations, the initial surrogate model is incrementally updated based on historical samples, and multi-generation fast optimization is performed based on the updated surrogate model. The optimization results output by the surrogate model are then replaced with the real population at intervals to generate offspring. This process is repeated until the preset maximum number of iterations is reached, and the Pareto optimal solution set is output. E6. Filter the Pareto optimal solution set to obtain the optimal design variables.

2. The multi-disciplinary collaborative optimization method for steel modular building structures based on the improved MOEA / D algorithm according to claim 1, characterized in that, The method for constructing isomorphic graph data in step E3 includes: Each individual component in a building structure is considered a node, and the node characteristics include the component's basic properties, spatial orientation, and cross-sectional dimensions. The Axis Aligned Bounding Box (AABB) algorithm is used to determine the connection relationship between components. If the projections of the bounding boxes of two components on the X, Y, and Z coordinate axes overlap, a connection relationship is determined to exist, and a bidirectional edge is established. The building structure is abstracted as an isomorphic graph. The prediction task of the graph convolutional network (GCN) structural surrogate model is a binary classification prediction of structural performance. It outputs a 10-dimensional binary label vector, which corresponds to whether the inter-story drift angle in the X direction of the building exceeds the limit, whether the inter-story drift angle in the Y direction of the building exceeds the limit, whether the bending strength of the beam exceeds the limit, whether the shear strength of the beam exceeds the limit, whether the deflection of the beam exceeds the limit, whether the overall stability of the beam exceeds the limit, whether the strength of the column exceeds the limit, whether the slenderness ratio of the column exceeds the limit, whether the in-plane stability of the column exceeds the limit, and whether the out-of-plane stability of the column exceeds the limit.

3. The multi-disciplinary collaborative optimization method for steel modular building structures based on the improved MOEA / D algorithm according to claim 1, characterized in that, The method for constructing heterogeneous graph data in step E3 includes: The nodes are classified into five categories: hot zone nodes, spatial nodes, building envelope nodes, ground nodes, and outdoor nodes. Twelve edge relationships are constructed, including hot zone containing space relationship, space containing enclosure structure relationship, enclosure structure connecting to the outside relationship, enclosure structure connecting to the ground relationship, enclosure structure connecting to enclosure structure relationship, enclosure structure connecting to hot zone relationship and its reverse relationship. The building energy consumption system is abstracted as a heterogeneous graph, and the prediction task of the Heterogeneous Graph Convolutional Network (HGCN) energy consumption proxy model is to regress and predict the total building cooling and heating load.

4. The multi-disciplinary collaborative optimization method for steel modular building structures based on the improved MOEA / D algorithm according to claim 1, characterized in that, In step E4, the optimization problem is represented as: ， in, For n design variables Composition Dimensional design variables; For the construction cost function, It is an energy consumption function; Here, m represents the i-th performance constraint specified in the design specifications, and m represents the total number of performance constraints. and The first Design variables The upper and lower boundaries are defined by find, min, and st, which represent finding, minimizing, and satisfying the condition, respectively. This includes the dimensions of various modular columns, the dimensions of various modular beams, the dimensions of various modular longitudinal beams, wall thickness, door thickness, floor slab thickness, roof thickness, and window heat transfer coefficient; Construction cost function Designed for , For the material cost of steel modular building structures; This is the penalty value when a single constraint is not satisfied; The number of individuals that do not meet the structural performance constraints, including the inter-story drift angle limit, component strength, stiffness, and overall stability requirements; Energy consumption function Designed for , This is the sum of the building's heating and cooling loads for steel modular building structures.

5. The multi-disciplinary collaborative optimization method for steel modular building structures based on the improved MOEA / D algorithm according to claim 1, characterized in that, Step E6 specifically involves: constructing a decision matrix for the objective function values ​​of each design scheme in the Pareto optimal solution set; using vector normalization to eliminate the influence of dimensions; determining the positive and negative ideal solutions; calculating the Euclidean distance and relative proximity of each scheme to the positive and negative ideal solutions; and determining the scheme with the largest relative proximity as the design scheme with the best overall performance and outputting it.

6. The multi-disciplinary collaborative optimization method for steel modular building structures based on the improved MOEA / D algorithm according to any one of claims 1 to 5, characterized in that, Step E1 specifically includes the following steps: E11. Input DXF drawing, extract geometric data of grid, wall, door and window, module frame and hot zone frame according to line type layer based on Dxfgrabber library, calculate center line coordinates and endpoints, and output initial component data in JSON format; E12: The architectural profession performs 19-bit encoding, parameter filling, collision correction, and component classification; the structural profession calculates room loads and fills in the mechanical materials of components; the energy consumption profession performs 23-bit encoding, space identification, and enclosure structure filling, and outputs JSON data for each profession. E13. Based on the five principles of cross-disciplinary integration, multi-level, classification, master-slave dependency, and coding collaboration, a tree-structure integration method is adopted to integrate the JSON data of the three disciplines according to the hierarchy and output integrated digital basic JSON data.

7. The multi-disciplinary collaborative optimization method for steel modular building structures based on the improved MOEA / D algorithm according to claim 6, characterized in that, In step E12, the 19-bit code of the architectural profession adopts a hierarchical segmented interpretation rule, defining the building type identifier, standard floor number, floor number, professional type, module type, module number, component type, geometric type, spatial orientation and component number bit by bit; The 23-bit code for energy consumption uses a hierarchical segmented interpretation rule, defining the building type identifier, standard floor number, floor number, professional type, thermal zone type, thermal zone number, space type, space number, building envelope type, building envelope geometry type, building envelope orientation, and building envelope number bit by bit.

8. The multi-disciplinary collaborative optimization method for steel modular building structures based on the improved MOEA / D algorithm according to claim 6, characterized in that, Step E2 specifically includes: E21. Based on the module classification characteristics of integrated digital foundation JSON data, a step-by-step accumulation method from module level, standard level to building level is adopted to multiply the volume and unit price of structural components, enclosure structural components and other fixed components and sum them to calculate the total cost of building materials. E22. Extract the spatial geometry information of the thermal zone and the thermal parameters of the building envelope from the integrated digital foundation JSON data, generate IDF files through Eppy automatic mapping, call EnergyPlus to perform building heat balance simulation, and calculate the sum of the annual cumulative heating load and cooling load. E23. Extract the geometric information and mechanical parameters of structural components from the integrated digital foundation JSON data, automatically generate node elements, apply boundary conditions and seismic loads in OpenSees, solve the node displacements and component internal forces through finite element analysis, and extract the inter-story drift angles and component strength, stiffness and stability indices.

9. The multi-disciplinary collaborative optimization method for steel modular building structures based on the improved MOEA / D algorithm according to claim 8, characterized in that, Step E22 specifically includes: Extract the spatial geometry information of the thermal zone and the thermal parameters of the building envelope from the integrated digital foundation JSON data. The geometric information includes the vertex coordinates of the thermal zone, space, and building envelope, and the thermal parameters include the material name and thermal performance of each building envelope. Based on three-dimensional spatial topology analysis, the external boundary conditions of each enclosure structure are automatically calculated to determine the attributes of cross-thermal zone connection, outdoor connection or ground connection. The thermal zone spatial geometry information is mapped to the thermal zone, spatial and enclosure structure geometry models in the IDF file using the Eppy library, and the thermal parameters of the enclosure structure are mapped to the corresponding material properties. EnergyPlus was used to perform building thermal balance simulation on the IDF file, and the hourly cooling load and hourly heating load of each thermal zone were calculated using the conduction transfer function method. The hourly cooling load is added up over the entire year to obtain the total annual cooling load, and the hourly heating load is added up over the entire year to obtain the total annual heating load. The two are then summed to obtain the total annual cooling and heating load.

Images