Semantic reasoning method for constraint conditions of bim component design of fabricated building based on construction data

Abstract

The application discloses a prefabricated building BIM component design constraint condition semantic reasoning method based on construction data, belongs to the prefabricated building design field of electric digital data processing, and realizes the structured processing and semantic analysis of multi-source heterogeneous construction data by constructing a construction knowledge graph and semantic reasoning rules. The method can automatically identify key constraint elements in the construction data and convert them into machine-readable "entity-relation-attribute" triples to form an inferable design constraint condition library. This enables architects to adjust parameters based on real construction logic in the early design stage, significantly improving the accuracy of BIM component design and construction feasibility, and overcoming the problem of disconnection with site conditions caused by relying on experience in traditional design.

Description

Technical Field

[0001] This invention relates to the field of prefabricated building design using electronic digital data processing, and specifically to a semantic reasoning method for design constraints of prefabricated building BIM components based on construction data. Background Technology

[0002] The design of BIM components for prefabricated buildings mainly includes five aspects: prefabricated buildings, building information models, construction data, architectural design constraints, and semantic reasoning of design constraints.

[0003] Prefabricated buildings refer to buildings assembled on-site from prefabricated components. Specifically, it is a new type of building production method that uses prefabricated component production in factories and on-site assembly installation as the construction mode. It is characterized by standardized design, factory production, assembly construction, integrated decoration, and information management (five aspects). It integrates various stages such as design, construction, and operation and maintenance, and pursues the maximization of the target value of the building throughout its entire life cycle.

[0004] The Model Context Protocol (MCP) is an open protocol standard whose core function is to allow large language models to securely and systematically connect to and use external tools, data sources, and services. By defining a unified specification, it enables large language models to dynamically discover and invoke external resources (such as databases, APIs, or software functions), thereby extending the capabilities of large language models. This allows them not only to generate answers based on training data but also to acquire information, perform operations, and process contextual information unknown or private during training in real time.

[0005] Building Information Modeling (BIM) is a digital technology that integrates and stores building-related information in a virtual model. Typical information can be categorized into design, construction, and operation / maintenance data. This data can be shared with professionals at all stages of the building's lifecycle, guiding them to make efficient, accurate, and correct decisions. Feature-based modeling (FBM) is a 3D modeling method that uses a series of engineering-meaning "features" (such as holes, chamfers, extrusions, and bodies of revolution; it should be noted that the definition of "feature" here is adapted to the BIM context) as the basic units for building the model. Design intent is linked to these features in the form of parameters and constraints, making the model creation process more intuitive and efficient, and facilitating subsequent modifications, updates, and the integration of manufacturing information. The applications of Building Information Modeling (BIM) technology mainly include, but are not limited to, the following aspects: (1) Visual design converts traditional two-dimensional drawings into three-dimensional building models, which can intuitively examine design details based on construction needs and reduce communication barriers between designers and builders; (2) Collision detection: Integrate the models of structural, mechanical and electrical and decoration disciplines with the architectural design model into a BIM model. Use BIM software to automatically detect conflicts and contradictions between the architectural design model and other professional models, optimize architectural design, and reduce construction problems. (3) Virtual construction (construction process simulation): By linking the BIM model with the construction schedule, the construction process can be simulated intuitively and accurately, reducing architectural design changes and increasing the feasibility of the construction plan; (4) Coordinate and collaborate with structural, mechanical and electrical and decoration disciplines, and conduct information exchange with relevant disciplines through integrated BIM models to increase the participation of each discipline in architectural design and improve the authenticity and constructability of the design; (5) Component splitting and detailed design: the building is split and detailed into building components that meet the requirements of factory prefabrication, or suitable building components are selected from the BIM component library for detailed improvement and combination design. The detailed design information and results are stored and exported through BIM, and accurate processing and assembly drawings are produced to promote the integration of design and construction.

[0006] Construction data refers to the multi-source information data generated and collected throughout the entire process of prefabricated building construction, from design, production, transportation to assembly. This data includes component dimensional accuracy, structural connection parameters, assembly tolerances, construction progress, and environmental conditions. These data record the specific execution of each construction stage, linking design decisions with actual construction results. This provides an empirical basis for verifying the rationality and feasibility of design decisions and serves as an important basis for analyzing potential design constraints and construction logic.

[0007] Architectural design constraints refer to a series of restrictions or rules imposed on a project during the architectural design, construction, and management processes to ensure the project's safety, functionality, economy, and compliance. These constraints typically involve multiple dimensions such as technical specifications, laws and regulations, environmental conditions, resource limitations, and functional requirements. In the context of the construction data in this patent, they specifically refer to the constraints (parameter values ​​or attribute values) extracted from actual construction needs and used to determine whether the BIM model's design parameter variables meet those actual construction requirements.

[0008] Semantic reasoning of design constraints refers to a technical method in which computer systems, based on natural language processing technology and combined with knowledge from fields such as logic and cognitive science, analyze and infer the semantic information in design constraints to understand their deeper meaning and logical relationships. It aims to analyze the associations and dependencies between constraints through relation identification and rule-based reasoning. Applications of semantic reasoning technology mainly include, but are not limited to, the following: (1) Compliance check: By analyzing the logical rules of building codes, standard clauses and other texts, the design scheme is verified to meet mandatory requirements such as safety, thereby improving review efficiency and reducing human error. (2) Design constraint conflict detection: identify logical contradictions and conflicts between design constraints of different disciplines and different stages, such as space requirement conflicts and inconsistent performance goals, so as to achieve early warning and coordination optimization; (3) Spatial relationship reasoning, based on formal semantic model analysis of spatial topological relationships, such as connectivity and visibility, to help optimize the rationality of functional layout; (4) Design intent derivation: extract key semantics from unstructured user demand descriptions, generate preliminary concepts that conform to intent through logical reasoning, and improve the accuracy of creative generation; (5) Design decision support: Based on domain knowledge and historical cases, reasoning rules are constructed to provide designers with solutions and modification guidance that meet the constraints, thereby enhancing the rationality of the design.

[0009] Currently, semantic reasoning technology and Building Information Modeling (BIM) have been gradually applied to the design, construction, and operation and maintenance phases of prefabricated buildings, playing a crucial role in design constraint management and decision support. The integrated application of semantic reasoning technology and Building Information Modeling (BIM) technology mainly includes, but is not limited to, the following aspects: (1) Ontology-based design constraint semantic modeling: By constructing a domain ontology, the attribute information and spatial topological relationships of components in the BIM model are transformed into a knowledge graph with semantic association, thereby establishing a design constraint network that can support logical reasoning and realizing machine-readable design conditions; (2) Integrated management of multi-source constraints: Using semantic reasoning technology, constraints from multiple sources and types, such as design specifications, owner requirements, and construction conditions, are semantically integrated to build a unified design constraint system and solve the problems of information dispersion and conflict. (3) Intelligent inspection and optimization of design schemes: The reasoning engine is used to automatically inspect the BIM model, identify design conflicts and defects that do not meet the constraints, and provide correction suggestions based on the reasoning results to assist in design optimization; (4) Parametric design and constraint-driven model generation combine semantic reasoning with parametric BIM modeling to achieve model adjustment and optimization based on constraints. However, the application of semantic reasoning technology in the intelligent generation and dynamic evolution of BIM models is still not deep enough, especially in dealing with complex design constraints, where there are still technical challenges. In general, BIM technology provides a structured expression and visualization platform for design information, while semantic reasoning technology undertakes the formal representation and logical reasoning functions of design constraints. The integrated application of the two can realize the "intelligent management" of design constraints, transforming scattered and implicit design requirements into executable and verifiable semantic rules.

[0010] In summary, semantic reasoning and Building Information Modeling (BIM) technologies, as important means to improve the intelligence level of building design, have shown promising application prospects in the management of design constraints for prefabricated buildings. BIM technology, through information integration, 3D visualization, and collaborative management, provides a rich data foundation for the expression and verification of design constraints; semantic reasoning technology, with its powerful logical reasoning capabilities, injects intelligent processing capabilities for design constraints into the BIM model. The integrated application of these two technologies has initially realized automated checking and intelligent decision support for design constraints.

[0011] However, in practical application, it has been found that the current combination of BIM and semantic reasoning technology in the prefabricated building design stage still has the following shortcomings and limitations: 1. Architects lack methods to extract and transform real construction data into BIM component design constraints.

[0012] Currently, in prefabricated building design, the core challenge facing architects is how to systematically extract key constraints from real construction data (such as component dimensional accuracy, assembly tolerances, and construction schedule) that can guide the design. Although BIM technology provides a platform for information integration and knowledge graphs have the ability to store and express semantic constraints, architects still lack clear methodologies and tools to effectively transform "construction data" into "design constraints."

[0013] Specifically, this manifests in several ways: Insufficient understanding of construction data and conversion of design rules: Architects struggle to systematically identify which construction data (such as hoisting path restrictions and factory production error ranges) should be extracted as key constraints affecting the design. They also lack the methodology to convert this data into executable parameter rules in the BIM model, leading to design decisions relying on experience rather than data-driven approaches. Lack of a mapping mechanism between design parameters and constraints: During the detailed design of BIM components, design parameters such as component dimensions and node connection methods fail to establish a dynamic mapping relationship with design constraints (such as transportation height limits and assembly process requirements), preventing architects from adjusting component attributes in real time based on actual construction conditions. Disconnection in the transmission of design-construction information: Because construction data is not effectively converted into semantic constraints prior to the design phase, design concepts such as component standardization and modularization are difficult to implement in the detailed design phase. Often, problems such as component mismatch and assembly conflicts only surface during the construction phase, requiring "secondary redesign" and resulting in a double loss of efficiency and quality. As a result, component designs often fail to fully reflect the requirements of factory production and on-site assembly. The detailed design results frequently encounter compatibility issues in actual construction, requiring multiple design iterations based on construction feedback. This severely restricts the improvement of design quality and the efficiency of design-build integration.

[0014] 2. The semantic reasoning method for deriving design constraints from real construction data is still unclear, and there is a lack of reasoning mechanisms for architectural design.

[0015] Semantic reasoning applications in the construction field are mostly concentrated in specific scenarios such as late-stage design review, code compliance verification, and collision detection. Its reasoning rules and mechanisms mainly revolve around "model review" and "conflict identification," while few studies focus on how to establish a complete semantic reasoning method to support the automatic conversion from real construction data to architectural design constraints.

[0016] Specific shortcomings include: Semantic reasoning does not support causal analysis and constraint generation of construction data: Existing semantic reasoning technologies are mostly used for compliance review and risk prediction, lacking the reasoning logic to identify causal relationships from multi-source construction data such as component size errors and construction environment, and to automatically generate component-level design constraints, making it difficult to drive the automated generation and dynamic evolution of BIM models. A proactive reasoning mechanism from design constraints to parameter modification is lacking: Although knowledge graphs can store semantic relationships between components, materials, and specifications in the form of triples, their reasoning capabilities have not been used to build a constraint-based proactive parameter modification mechanism. Adjustments to design parameters still rely mainly on manual intervention and post-review, failing to respond in real-time to changes in construction conditions during the design process. A semantic reasoning system for architectural design has not yet been established: The current application of semantic reasoning in prefabricated buildings is characterized by "reverse" and "fragmentation," lacking reasoning rules and algorithmic support oriented towards architectural design generation. In other words, the current semantic reasoning technology has not yet achieved the transformation from "model review" to "design generation and adjustment," and has failed to form a semantic reasoning application system for prefabricated component design. This has resulted in the failure to form a closed-loop intelligent interaction between BIM models and knowledge graphs, which restricts the integrated process from design to construction and the enhancement of value throughout the entire life cycle. Summary of the Invention

[0017] In view of at least one of the above-mentioned technical problems, the present invention provides a semantic reasoning method for design constraints of prefabricated building BIM components based on construction data, to map standardized building design constraint rules, thereby providing support for constraint modeling, and intelligently adjusting BIM component parameter information for forward-looking reverse optimization to ultimately improve ease of construction. The specific technical solution is as follows: A semantic reasoning method for design constraints of prefabricated building BIM components based on construction data mainly includes the following steps: S1, collect construction data generated during the actual construction process of prefabricated buildings and establish a construction database; S2, analyze construction data and extract the actual construction requirements of prefabricated buildings as key construction features. The requirements extraction directions include: construction process, material composition and technical implementation; construct a BIM component feature framework that maps key construction features; S3, design constraints are derived based on the BIM component feature framework. The design constraints include geometric constraints and textual constraints. S4 transforms the design constraints according to machine language rules, builds a knowledge graph according to semantic relationships, and completes semantic reasoning and constraint storage.

[0018] In some embodiments of this disclosure, the sources of construction data in step S1 include at least one of the following: processing records of prefabricated components, assembly deviations, node connection methods, hoisting and transportation restrictions.

[0019] In some embodiments of this disclosure, the prefabricated component is an existing finished component, and the information source is existing technical data related to the prefabricated component. The technical data includes at least one of the component's technical drawings, technical specifications, or construction instructions.

[0020] In some embodiments of this disclosure, the prefabricated components are existing finished components that have been applied in prefabricated building engineering projects. The information sources also include real construction information collected from existing prefabricated building engineering projects, including information on the prefabricated components during factory manufacturing, transportation, and on-site construction stages collected by related devices such as radio frequency identification, code identification, laser scanning, and image capture, as well as IoT technology.

[0021] In some embodiments of this disclosure, in step S2, the source of the actual construction requirements includes at least one of engineering documents, information collected from interviews and / or surveys of the execution team, construction progress information, and on-site survey graphic information.

[0022] In some embodiments of this disclosure, the method for extracting real needs includes data parsing and classification, problem identification and classification, and semantic transformation of needs.

[0023] In some embodiments of this disclosure, in step S3, the geometric constraints include BIM parametric formulas and BIM parameter values, and the textual constraints include the characterization names of different components.

[0024] In some embodiments of this disclosure, the BIM component feature framework includes identifying the location of building sub-components, defining component features abstracted from FBM, specifying different types of sub-component names, and at least one of the component's dimensional attributes.

[0025] In some embodiments of this disclosure, the method for converting design constraints according to machine language rules includes manual identification and / or automatic identification.

[0026] In some embodiments of this disclosure, the manual identification is based on the experience of operators to summarize and classify design constraints, and then extract and translate common rules. The automatic identification is directly translated through a large artificial intelligence model.

[0027] In some embodiments of this disclosure, MCP technology can be used to establish a connection between a large artificial intelligence model and knowledge graphs and BIM-related software. Then, by inputting natural language commands to the large artificial intelligence model, it can use MCP technology to call the corresponding software functions, and automatically complete the entire process from reasoning analysis to translation and generation of knowledge graphs.

[0028] Compared with existing technologies, the above-mentioned semantic reasoning method for prefabricated building BIM component design constraints based on construction data has the following advantages: This invention achieves structured processing and semantic parsing of multi-source heterogeneous construction data (such as component dimensional accuracy, assembly tolerance, construction progress, etc.) by constructing a construction knowledge graph and semantic reasoning rules. This method can automatically identify key constraint elements in construction data and transform them into machine-readable "entity-relationship-attribute" triples, forming a reasonable design constraint library. This allows architects to adjust parameters based on real construction logic in the early design stages, significantly improving the accuracy and feasibility of BIM component design and overcoming the problem of disconnection from site conditions caused by reliance on experience-based judgment in traditional design. This invention overcomes the limitations of current semantic reasoning technology in design generation and adjustment, proposing a semantic reasoning mechanism for the detailed design of prefabricated building components. By deeply integrating the construction knowledge graph with the BIM model, it achieves automatic verification and intelligent correction of component parameters based on constraints. This method can detect the compliance of BIM component parameters and construction constraints in real time during the design process, and drive the automatic adjustment of parameters based on semantic reasoning results. It effectively solves the lag problem of the traditional "modeling-detection-manual modification" process, significantly improving design efficiency and quality. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the process of the present invention; Figure 2 This is a schematic diagram of the actual construction data extracted by BC based on the project construction drawings in an embodiment of the present invention; Figure 3 This is a schematic diagram of the BIM component feature frame in an embodiment of the present invention; Figure 4 This is a schematic diagram of five types of design constraints related to the connection design of prefabricated components in an embodiment of the present invention; Figure 5 This is a schematic diagram of the manual recognition method for semantic reasoning in an embodiment of the present invention; Figure 6 This is a schematic diagram of the user interface containing constraints extracted from the Deepseek large language model by inputting natural language text into the model in this embodiment of the invention. Figure 7 This is a schematic diagram of the user interface for automatically translating constraints in the Deepseek large language model in an embodiment of the present invention; Figure 8 This is a user interface diagram illustrating how the Deepseek large language model, in accordance with the syntax logic of the Cypher language, completes and expands constraints into Cypher statements according to the Cypher language in an embodiment of the present invention. Figure 9This is a schematic diagram of the construction of a knowledge graph based on Neo4j in an embodiment of the present invention; Figure 10 In this embodiment of the invention, the large-scale artificial intelligence model and MCP technology are used in synergy to automatically complete the process from reasoning and analysis to translation and generation of knowledge graphs. Detailed Implementation

[0030] To better understand the purpose, structure, and function of this invention, the technical solutions in the embodiments of this invention will be clearly and completely described below. It should be noted that, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing specific embodiments only and is not intended to limit this application. The terms "comprising" and "having," and any variations thereof, are open-ended and intended to cover non-exclusive inclusion.

[0031] As shown in the attached diagram. Figures 1 to 10 As shown, this disclosure provides a semantic reasoning method for design constraints of prefabricated building BIM components based on construction data. This method maps standardized building design constraint rules, thereby providing support for constraint modeling. It intelligently adjusts BIM component parameter information for forward-looking reverse optimization, ultimately improving ease of construction. The main steps include:

[0032] S1 uses a systematic approach to collect and organize multi-source heterogeneous data from the real construction process to establish a construction database. The systematic approach includes data parsing and classification, problem identification and classification, and semantic transformation of requirements.

[0033] This embodiment summarizes and organizes data from a prefabricated timber structure project, including design drawings (such as architectural design drawings, component construction drawings, and manufacturing drawings), textual materials (such as engineering records, interview records, project reports, collision detection reports, and related papers), image materials (such as model images, analysis diagrams, on-site photos of the production, transportation, and assembly stages, and as-built photos), as well as PPTs, images, and videos used for promotion on the project's website portal, as shown in Table 1 below: Table 1. Quantitative and Qualitative Data Collection for Prefabricated Timber Structure Projects

[0034] The sources of construction data include at least one of the following: processing records of prefabricated components, assembly deviations, node connection methods, hoisting and transportation restrictions. These data not only cover mathematical constraints such as dimensional tolerances, component combination rules, and connector types, but also macro-constraints such as construction sequence, spatial accessibility, and node processing priority.

[0035] The prefabricated component is an existing finished component, and the information source is the existing technical data related to the prefabricated component. The technical data includes at least one of the following: technical drawings (schematic drawings, detailed drawings, and construction drawings, etc.), technical specifications, and construction specifications related to the component from design drawings to the final installation process.

[0036] The prefabricated components are existing finished components that have been applied in prefabricated building projects. The information sources also include real construction information collected from existing prefabricated building projects, including information on the prefabricated components during factory manufacturing, transportation, and on-site construction stages collected by related devices such as radio frequency identification, code recognition, laser scanning, and image capture, as well as IoT technology.

[0037] S2 involves sorting and classifying the aforementioned multi-source data files, comparing the before and after versions of the component drawings, extracting "key issues" from the original drawings, and analyzing "construction requirements" from the final drawings. This accurately identifies key issues and construction requirements as key construction features, laying the foundation for constructing the BIM component feature framework. Figure 2 As shown, the analysis of data from each stage of prefabricated building construction identifies clear construction needs and key issues. This analysis is conducted across multiple dimensions, including longitudinal (time, construction process), horizontal (space, material composition), and vertical (implementation, technical realization), to construct a BIM component feature framework that maps key construction characteristics. This framework clearly defines various characteristics of the components and specifically sets "attributes" to describe non-geometric information, as well as configuring "parameters" to define geometric shapes, such as... Figure 3 As shown.

[0038] The BIM component feature framework uses a standardized alphabetical coding system to express four key information categories: the location of a building sub-component, component features, sub-component name, and component size, through specific letters and their arrangement.

[0039] The key feature information in the BIM component feature framework needs to correspond one-to-one with the prefabricated building design information and actual construction requirements. This framework clearly defines the various features of the component, specifically sets the "attributes" used to describe non-geometric information, and configures the "parameters" used to define the geometric shape.

[0040] The specific content of the BIM component feature framework is as follows: (1) Identify the location of building components (wall panels, columns, and floors); (2) Defined the component features (joints, connectors, penetrations, lifting points and tolerances) abstracted from FBM; (3) The component names of different types of sub-components are specified; (4) Includes the size attributes of the components (including nine types such as length and width).

[0041] The specific definitions of the features of the BIM component feature frame are as follows: (1) Connectors: refers to connectors between building components, such as hollow steel (HSS) pipes that connect upper and lower columns; (2) Penetration: refers to the opening created when a component passes through another component, such as the hole left in the floor by a steel pipe; (3) Joint components: refers to gaskets or filling materials used to close gaps between components, such as placing a 1.58mm thick alignment plate between columns to correct the structural height; (4) Tolerance: refers to the allowable deviation reserved for the manufacturing and installation process. For example, the gap between steel pipes A and B is reserved at 5mm, which can fluctuate up or down by 2mm. (5) Lifting point: refers to the lifting-related components designed for lifting operations, such as the lifting point dimensions used for wall panel lifting.

[0042] S3, based on the BIM component feature framework, derives and defines the design constraints that affect the ease of construction of three types of prefabricated components: wall panels, floor slabs, and columns. As a prerequisite for attribute ontology modeling, BIM component features are semantic expressions through BIM parameter entries as "carriers". These "features" can exist objectively without relying on BIM.

[0043] Sources of actual construction needs include at least one of the following: engineering documents, information gathered from interviews and / or surveys of the implementation team, construction progress information, and on-site survey textual and graphic information.

[0044] Methods for extracting real needs include data parsing and classification, problem identification and classification, and semantic transformation of needs.

[0045] Based on the operational characteristics of BIM software, design constraints can be categorized as follows: (1) Geometric constraints, including BIM parametric formulas and BIM parameter values; (2) Textual constraints, representing the names of different components.

[0046] By employing two types of design constraints, real-world construction knowledge is transformed into executable and verifiable rules within the BIM model, such as... Figure 4 As shown.

[0047] The BIM component feature framework includes identifying the location of building sub-components, defining component features abstracted from FBM, specifying component names for different types of sub-components, and at least one of the component's dimensional attributes.

[0048] Constraints are described in natural language, such as component connection rules, tolerance requirements, or assembly logic. The reasoning behind design constraints transforms implicit experience into explicit parameterized rules, including geometric dimensions, spatial relationships, and functional requirements, laying the foundation for subsequent computer processing.

[0049] The BIM parameter items mainly include time-related BIM parameters, spatial BIM parameters, and implementation-related BIM parameters.

[0050] Time-based BIM parameters, including the construction process.

[0051] The architectural design constraints corresponding to the construction process of prefabricated buildings and their components include planned and actual completion time (minutes / hours / days), completion status (completed / under construction / not completed), assembly procedures (steps), and other BIM parameters related to the construction process. The "planned" part corresponds to architectural design information, and the "actual" part corresponds to actual construction requirements.

[0052] Spatial BIM parameters, which include material composition; The architectural design constraints corresponding to prefabricated buildings and their component materials include the geometric dimensions of the designed and actual components, material physical parameters, detailed structural components, and other BIM parameters related to the material composition; the "design" part corresponds to architectural design information, and the "actual" part corresponds to actual construction requirements.

[0053] Implementation-type BIM parameters, which include technical implementations.

[0054] The architectural design constraints corresponding to prefabricated buildings and their component technologies include the connection methods between the design and the actual components, the connection locations of the components, and other BIM parameters related to the technology implementation. The "Design" section corresponds to architectural design information, while the "Actual" section corresponds to actual construction requirements.

[0055] S4 converts the design constraints according to machine language rules, establishes a knowledge graph according to semantic relationships, and completes semantic reasoning and constraint storage. It should be noted that the design constraints here are readable in the BIM environment.

[0056] Among them, such as Figure 5 As shown, the design constraints described in natural language are "extracted" by translating them into Cypher code, and the descriptions specific to the constraints are filtered out. The specific methods include: Manual identification involves summarizing and classifying constraints based on the experience of operators, then extracting and translating common rules. Classification can be based on node labels, node names, relationships, values, and units. The classified elements are then reorganized into "entity-relationship-attribute" triples. The Cypher language is then used to convert the triples into nodes and relationships that can be stored in a graph database. The converted constraint rules are then integrated into a graph database (such as Neo4j) to build a knowledge graph that supports semantic queries and real-time verification.

[0057] The specific content is as follows: Sentence segmentation involves inputting natural language text containing constraints into the translator and extracting phrases containing constraints, including component names, component functional perspectives, and component dimensions (dimensional values, units, dimensional relationships, etc.). Semantic annotation, based on the core "entity-relationship-attribute" triple structure of knowledge graphs, classifies phrases into five labels ("node label", "node name", "relationship", "value" and "unit"), thereby extracting effective components from natural language text; The parsing process reorganizes the labeled phrases, words, numbers, or units into structured triples. Code generation follows the syntax and logic of the Cypher language to convert triples into nodes and relationships in a knowledge graph to support querying and application.

[0058] Automatic recognition: Direct translation through large-scale artificial intelligence models, such as the Deepseek large language model, or by sending natural language commands to large-scale artificial intelligence models, the entire process from reasoning and analysis to translation and knowledge graph generation is completed by directly applying the collaborative application of large-scale artificial intelligence models and MCP technology.

[0059] The specific content is as follows: Sentence segmentation, such as Figure 6 As shown, input natural language text containing constraints into the Deepseek large language model, input the established BIM component feature framework, and command it to extract phrases containing constraints, including sub-component names, sub-component locations, component feature perspectives, and component dimensions (dimensional values, dimensional units, dimensional relationships, etc.). Semantic annotation, such as Figure 7 As shown, based on the "entity-relationship-attribute" triple structure of the knowledge graph core, phrases are classified into five labels ("node label", "node name", "relationship", "value" and "unit"). The manually translated triple and label structure is input into the Deepseek large language model, which is then instructed to match the obtained phrases with the structure of this process, thereby extracting effective components from natural language text and translating them into Cypher code fragments. Manual verification and review: The Cypher code snippet obtained from Deepseek is checked against the constraints derived in step 4 to obtain an accurate Cypher code snippet that can express the constraints. Code completion involves inputting a correct Cypher code snippet into the Deepseek large language model. The model then completes and expands this code into Cypher statements according to the syntax and logic of the Cypher language. This transforms constraints into nodes and relationships in a knowledge graph to support querying and applications. For example... Figure 8 As shown, in this embodiment, the constraints related to the three main components of a prefabricated wood structure project—wall panels, floor slabs, and columns—as well as the 19 sub-components related to the connection, are obtained through semantic reasoning to obtain complete Cypher code. This code is then input into Neo4j to construct a knowledge graph that includes 68 nodes, 75 relationships, and multiple properties, thus completely and accurately expressing the component connection-related information.

[0060] The technical solution proposed in this invention provides a semantic reasoning method for design constraints of prefabricated building BIM components based on construction data. This method facilitates the automated transformation and intelligent application of construction data into design constraints, promoting the shift of prefabricated building design from experience-driven to data-driven approaches. Compared with existing technologies, it has the following main advantages: Establish a semantic transformation mechanism from construction data to design constraints to improve the accuracy of design decisions.

[0061] To address the current challenge faced by architects in effectively transforming real-world construction data into design constraints, this invention achieves structured processing and semantic parsing of multi-source, heterogeneous construction data (such as component dimensional accuracy, assembly tolerances, and construction schedules) by constructing a construction knowledge graph and semantic reasoning rules. This method automatically identifies key constraint elements in construction data and transforms them into machine-readable "entity-relationship-attribute" triples, forming a reasonable design constraint library. This allows architects to adjust parameters based on real-world construction logic in the early design phase, significantly improving the accuracy and feasibility of BIM component design and overcoming the disconnect between traditional design methods that rely on experience and on-site conditions.

[0062] 2. Implement semantic reasoning capabilities for component-level in-depth design, supporting real-time verification and dynamic optimization during the design process.

[0063] This invention overcomes the limitations of current semantic reasoning technology in design generation and adjustment, proposing a semantic reasoning mechanism for the detailed design of prefabricated building components. By deeply integrating the construction knowledge graph with the BIM model, it achieves automatic verification and intelligent correction of component parameters based on constraints. This method can detect the compliance of BIM component parameters and construction constraints in real time during the design process, and drive the automatic adjustment of parameters based on semantic reasoning results. It effectively solves the lag problem of the traditional "modeling-detection-manual modification" process, significantly improving design efficiency and quality.

[0064] Compared with existing BIM-based prefabricated building design methods, the core innovation of this invention lies in moving semantic reasoning technology from the later model review stage to the design generation and optimization stage, making construction knowledge the intelligent core driving the dynamic evolution of the BIM model. Specifically, it achieves the following technical advantages: (1) It constructs a semantic bridge between construction data and design constraints, transforming scattered and implicit construction requirements into computable and inferable design rules, providing architects with a systematic method to integrate construction knowledge into the design process; (2) A closed-loop reasoning mechanism of “data-constraint-parameter” has been formed, which supports real-time self-verification and autonomous optimization of BIM components in the design process, and significantly reduces secondary modifications caused by mismatch between design and construction conditions; (3) A semantic constraint system for multi-disciplinary collaboration has been established, enabling structural, mechanical and electrical, decoration and other disciplines to coordinate parameters and optimize models under a unified construction knowledge graph framework, thus promoting the realization of truly integrated design of prefabricated buildings.

[0065] Through the above innovations, this invention not only improves the information depth and design accuracy of BIM models, but more importantly, it provides a digitally integrated solution for prefabricated buildings from design to construction, laying a technical foundation for the industry to maximize value throughout the entire life cycle.

[0066] It is understood that the above description is only for illustrating specific embodiments of this application, but the protection scope of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be covered within the scope of disclosure of this application.

Claims

1. A semantic reasoning method for design constraints of prefabricated building BIM components based on construction data, characterized in that, The main steps include: S1, collect construction data generated during the actual construction process of prefabricated buildings and establish a construction database; S2, analyze construction data and extract the actual construction requirements of prefabricated buildings as key construction features. The requirements extraction directions include: construction process, material composition and technical implementation; construct a BIM component feature framework that maps key construction features; S3, design constraints are derived based on the BIM component feature framework. The design constraints include geometric constraints and textual constraints. S4 transforms the design constraints according to machine language rules, establishes a knowledge graph based on semantic relationships, and completes semantic reasoning and constraint storage.

2. The semantic reasoning method for design constraints of prefabricated building BIM components based on construction data according to claim 1, characterized in that, In step S1, the sources of construction data include at least one of the following: processing records of prefabricated components, assembly deviations, node connection methods, hoisting and transportation restrictions.

3. The semantic reasoning method for design constraints of prefabricated building BIM components based on construction data according to claim 2, characterized in that, The prefabricated component is an existing finished component, and the information source is existing technical data related to the prefabricated component. The technical data includes at least one of the component's technical drawings, technical specifications, or construction instructions.

4. The semantic reasoning method for design constraints of prefabricated building BIM components based on construction data according to claim 3, characterized in that, The prefabricated components are existing finished components that have been applied in prefabricated building projects. The information sources also include real construction information collected from existing prefabricated building projects, including information on the prefabricated components during factory manufacturing, transportation, and on-site construction stages collected by related devices such as radio frequency identification, code recognition, laser scanning, and image capture, as well as IoT technology.

5. The semantic reasoning method for design constraints of prefabricated building BIM components based on construction data according to claim 1, characterized in that, In step S2, the sources of actual construction needs include at least one of the following: engineering documents, information collected from interviews and / or surveys of the execution team, construction progress information, and on-site survey graphic information.

6. The semantic reasoning method for design constraints of prefabricated building BIM components based on construction data according to claim 5, characterized in that, Methods for extracting real needs include data parsing and classification, problem identification and classification, and semantic transformation of needs.

7. The semantic reasoning method for design constraints of prefabricated building BIM components based on construction data according to claim 1, characterized in that, In step S3, geometric constraints include BIM parametric formulas and BIM parameter values, and textual constraints include the characterization names of different components.

8. The semantic reasoning method for design constraints of prefabricated building BIM components based on construction data according to claim 1, characterized in that, In step S3, the BIM component feature framework includes identifying the location of building sub-components, defining component features abstracted from FBM, specifying the names of different types of sub-components, and at least one of the component's size attributes.

9. The semantic reasoning method for design constraints of prefabricated building BIM components based on construction data according to claim 1, characterized in that, In step S4, the method for converting design constraints according to machine language rules includes manual identification and / or automatic identification.

10. The semantic reasoning method for design constraints of prefabricated building BIM components based on construction data according to claim 9, characterized in that, The manual identification is based on the experience of the operators to summarize and classify the design constraints, and then extract and translate common rules. The automatic identification is achieved by directly translating through a large artificial intelligence model, or by directly applying a large artificial intelligence model and MCP technology to complete the entire process from reasoning analysis to translation and knowledge graph generation.

Images