BIM model structure design examination method and system based on large model

Abstract

The invention discloses a BIM model structure design review method and system based on a large model, and relates to the technical field of building information modeling, and the method comprises the steps: carrying out BIM model data preprocessing based on an AI large model; adopting an AI large model built-in rule engine to perform structural design specification review on the BIM model, and automatically generating a review report based on a natural language processing technology; bIM model collision detection is executed based on a deep learning algorithm, potential structure conflicts are identified, error points are displayed in combination with three-dimensional visualization, and adjustment suggestions are provided according to historical data and an optimization algorithm. According to the method, the data quality and consistency are improved, key information in the BIM model is clearer and more concise, the accuracy of the review process is improved, potential problems in the structural design process can be found and processed in time, the overall intelligent level of the BIM model is improved, and the building design process is more accurate, intelligent and efficient.

Description

Technical Field

[0001] The present invention relates to the technical field of building information modeling, and specifically to a BIM model structure design review method and system based on a large model. Background Art

[0002] In recent years, with the wide application of Building Information Modeling (BIM) technology, the engineering construction industry has gradually developed towards intelligence and digitization. Through three-dimensional digital means, BIM technology has realized collaborative design among multiple disciplines such as building structure, mechanical and electrical, and heating, ventilation, and air conditioning, improving the design efficiency and construction quality of building engineering projects. However, with the increasing complexity of BIM models, the accuracy, standardization, and optimization ability of structural design have become important guarantees for project quality. To ensure that BIM models comply with relevant building design standards and specifications, a series of BIM model review methods have been gradually developed in the industry, including automatic review based on rule matching, collision detection based on geometric analysis, and optimization analysis based on numerical simulation. These methods have improved the compliance and constructability of BIM models to a certain extent, but still rely on manually set rules or thresholds and are difficult to adapt to the ever-changing building design requirements. In recent years, the development of artificial intelligence technology, especially deep learning and natural language processing technology, has provided new ideas for BIM model review. Through AI algorithms, design specifications can be automatically learned and extracted, improving the automation and accuracy of review.

[0003] There are still many limitations in the application of existing BIM model review methods. Existing methods mainly rely on rule-based review engines, whose rule libraries are usually manually defined and require continuous maintenance and update by professionals, resulting in limited coverage of review rules and difficulty in adapting to the dynamic changes of building design standards. Due to the certain ambiguity of building design specifications, rule systems based on rigid matching are difficult to handle complex design constraints, resulting in limited accuracy of review results. The collision detection and structural optimization capabilities are weak. Traditional methods usually perform structural collision detection based on geometric calculations, which can only identify explicit geometric conflicts and cannot analyze potential structural bearing capacity problems. At the same time, for the detected conflict points, existing technologies mainly rely on manual adjustment and lack intelligent optimization suggestions based on historical data, affecting the optimization efficiency of BIM models. The generation method of review reports is relatively single. Existing technologies usually output review results in a templated manner, lacking in-depth analysis of design problems and unable to provide readable reports in natural language, affecting designers' understanding and correction of problems. Summary of the Invention

[0004] In view of the above problems, the present invention is proposed.

[0005] Therefore, the technical problem to be solved by the present invention is that the existing BIM model structure design review method based on large models has a rule base relying on manual maintenance, making it difficult to dynamically adapt to changes in design specifications, limited collision detection capabilities, unable to perform in-depth structural optimization, a single way of generating review reports, lacking intelligent analysis, and the problem of how to combine AI large models to achieve efficient data preprocessing, intelligent rule review, and automated BIM design review with deep learning optimization of BIM models.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: A BIM model structure design review method based on large models, including performing BIM model data preprocessing based on an AI large model; using the built-in rule engine of the AI large model to conduct structural design specification review on the BIM model, and automatically generating a review report based on natural language processing technology; performing BIM model collision detection based on deep learning algorithms, identifying potential structural conflicts, combining three-dimensional visualization to display error points, and providing adjustment suggestions according to historical data and optimization algorithms; the rule engine includes constructing a BIM structure design knowledge graph based on Ontology semantic modeling, extracting rules from building design standards, and converting them into logical inference rules; generating a review report includes automatically generating a BIM review report based on natural language processing technology, using the BERT pre-trained language model to extract technical parameters, generating a review report summary through a Seq2Seq structure, and screening high-risk review points by the TextRank algorithm; providing adjustment suggestions includes combining GNN and LSTM prediction models to enable the BIM model to identify conflicts in the design, making dynamic optimization suggestions based on historical building data, and generating a 3D interactive optimization plan.

[0007] As a preferred solution of the BIM model structure design review method based on large models according to the present invention, wherein: the data preprocessing includes using the outlier detection algorithm of the AI large model to train an RNN based on historical BIM design data, automatically identifying outliers in BIM structure parameters, and automatically filling them; performing unified format conversion of BIM structure data through a Transformer model, parsing BIM data generated by different software into IFC standard format; using a semantic parsing model to clean unstructured text information, removing redundant design annotations and duplicate BIM component labels.

[0008] As a preferred solution of the BIM model structural design review method based on a large model according to the present invention, wherein: the rule engine includes constructing a BIM structural design knowledge graph based on Ontology semantic modeling, extracting the rules of building design standards, and converting the rules of building design standards into logical inference rules; the knowledge structure of the knowledge graph includes a design specification layer, a rule conversion layer, and a BIM data layer; the design specification layer includes regulatory clauses; the rule conversion layer includes logical inference rules; the BIM data layer includes actual design parameters extracted from BIM data.

[0009] As a preferred solution of the BIM model structural design review method based on a large model according to the present invention, wherein: the structural design specification review includes that the rule mapping adopts a decision tree algorithm to perform specification matching according to the attribute values of the BIM structural model; based on FuzzyLogic optimization, for BIM structural parameters that do not meet the fixed threshold, a fuzzy logic method is used to provide optimization suggestions within the tolerance range and output the review results.

[0010] As a preferred solution of the BIM model structural design review method based on a large model according to the present invention, wherein: the automatically generating a review report based on natural language processing technology includes using a BERT pre-trained language model to parse the BIM design specification, extracting core technical parameters, including load standards and structural constraint conditions; automatically generating a review report summary through a Seq2Seq structure to summarize BIM structural design problems; using the TextRank algorithm to automatically screen high-risk review points and provide key correction suggestions.

[0011] As a preferred solution of the BIM model structural design review method based on a large model according to the present invention, wherein: the identifying potential structural conflicts includes using a PointNet++ deep learning framework to analyze the 3D point cloud data of the BIM structural model to detect beam-column collisions and floor penetration problems.

[0012] As a preferred solution of the BIM model structural design review method based on a large model according to the present invention, wherein: the providing adjustment suggestions includes using a GNN to perform BIM component relationship reasoning to identify the mechanical relationships between floors, beams, and walls and provide adjustment solutions; using an LSTM prediction model combined with historical building project data to provide BIM structural optimization suggestions based on big data analysis and generate a 3D interactive optimization solution.

[0013] Another object of the present invention is to provide a BIM model structural design review system based on a large model, which can perform structural design specification review on a BIM model by using a rule engine built into an AI large model, and solves the problem of lack of intelligent analysis in the current BIM model structural design review technology based on a large model.

[0014] As a preferred solution of the BIM model structural design review system based on large models according to the present invention, it includes a data processing module, a review module, and an optimization module; the data processing module is used for preprocessing BIM model data based on an AI large model; the review module includes a rule engine module and a report generation module, the rule engine module is used for conducting structural design specification review on the BIM model by using the built-in rule engine of the AI large model, and the report generation module is used for automatically generating a review report based on natural language processing technology; the optimization module includes a conflict identification module and an adjustment module, the conflict identification module is used for performing BIM model collision detection based on a deep learning algorithm to identify potential structural conflicts, and the adjustment module is used for combining three-dimensional visualization to display error points and providing adjustment suggestions according to historical data and optimization algorithms.

[0015] A computer device includes a memory and a processor, the memory stores a computer program, and the processor executes the computer program to implement the steps of the BIM model structural design review method based on large models.

[0016] A computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, it implements the steps of the BIM model structural design review method based on large models.

[0017] Advantages of the present invention: The BIM model structural design review method based on large models provided by the present invention preprocesses BIM model data based on an AI large model, improves data quality and consistency, reduces the risk of manual intervention, conducts structural design specification review on the BIM model by using the built-in rule engine of the AI large model, and automatically generates a review report based on natural language processing technology, avoiding human omissions and biases in subjective judgments, improving the accuracy and effectiveness of decision-making, performing BIM model collision detection based on a deep learning algorithm to identify potential structural conflicts, combining three-dimensional visualization to display error points, and providing adjustment suggestions according to historical data and optimization algorithms. Designers can view error points in real time, quickly make decisions and correct problems, improving the scientificity and accuracy of design optimization. The present invention achieves better results in terms of BIM model data processing efficiency, structural design specification review accuracy, and structural conflict identification and optimization suggestion generation. Description of the Drawings

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for description in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0019] Figure 1 The overall flowchart of a BIM model structural design review method based on a large model provided for the first embodiment of the present invention. Detailed implementation manners

[0020] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will describe in detail the specific implementation manners of the present invention with reference to the accompanying drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0021] Embodiment 1, referring to Figure 1 , which is an embodiment of the present invention, provides a BIM model structural design review method based on a large model, including: S1: Perform BIM model data preprocessing based on an AI large model.

[0022] Furthermore, the data preprocessing includes using the outlier detection algorithm of the AI large model to train an RNN based on historical BIM design data, automatically identifying outliers in BIM structural parameters, and automatically filling them; performing unified format conversion of BIM structural data through a Transformer model, parsing BIM data generated by different software into the IFC standard format; using a semantic parsing model to clean unstructured text information, removing redundant design annotations and duplicate BIM component labels.

[0023] It should be noted that through the BIM model data preprocessing steps based on the AI large model, first, through the outlier detection algorithm of the AI large model, the RNN (Recurrent Neural Network) is trained using historical BIM design data to automatically identify outliers in the BIM structure parameters. This process can quickly and accurately detect the parts of the data that do not conform to the conventional rules and fill them automatically, avoiding the cumbersome process of manual inspection and correction in the traditional method, improving the accuracy and reliability of the BIM model data, and thus laying a solid foundation for subsequent design review, collision detection and other steps; through the Transformer model for unified format conversion of BIM structure data, the BIM data generated by different software is parsed into a unified IFC standard format, which can ensure the seamless connection of the BIM data generated by different software, realize the data compatibility between different software platforms, improve the efficiency of cross-platform collaborative work, not only reduce the cumbersome work of manually converting data, but also ensure that the subsequent steps can be processed in a unified format, avoiding errors caused by format differences; by adopting a semantic parsing model to clean unstructured text information, redundant design annotations and duplicate BIM component labels are removed, further optimizing the data quality, making the key information in the BIM model clearer and more concise, which is conducive to improving the efficiency of subsequent analysis and review.

[0024] S2: Use the built-in rule engine of the AI large model to conduct a structural design specification review on the BIM model and automatically generate a review report based on natural language processing technology.

[0025] Furthermore, the rule engine includes building a BIM structural design knowledge graph based on Ontology semantic modeling, extracting the rules of building design standards, and converting the rules of building design standards into logical inference rules; the knowledge structure of the knowledge graph includes a design specification layer, a rule conversion layer, and a BIM data layer; the design specification layer includes regulatory clauses; the rule conversion layer includes logical inference rules; the BIM data layer includes actual design parameters extracted from BIM data.

[0026] It should also be noted that a preferred solution for constructing a BIM structural design knowledge graph specifically includes extracting regulatory clauses from architectural design specifications and converting each clause into a standardized node. The regulatory clauses include the design requirements for structural components such as beams, columns, and walls; using natural language processing techniques to extract the logical relationships between the clauses and mapping these relationships into the ontology structure to form a knowledge graph between the design specification layer and the BIM data layer. The logical relationships include the dependency relationships between design requirements, such as "if the span of a beam exceeds 6 meters, additional steel bars are required"; mapping the structural data in the BIM model to the nodes of the knowledge graph. The BIM model data includes design parameters such as the dimensions, materials, and loads of beams, columns, and walls; associating BIM data with design specification requirements through the knowledge graph to provide a basis for structural design review; a preferred solution for extracting the rules of architectural design standards specifically includes extracting explicit rules from architectural design specifications. Each rule includes the structural parameters of the design requirements. The structural parameters include the maximum span of a beam, the minimum thickness of a wall, etc.; using data mining techniques to mine potential rules from BIM design data. The potential rules include the common range of structural parameters in historical design data. The data mining technique uses an association rule mining algorithm such as the Apriori algorithm; combining the explicit rules extracted from the design specifications with the potential rules mined from BIM data to form a rule library; matching the rule library with the design data of the BIM model to provide a basis for structural review and generate optimization suggestions.

[0027] It should be noted that the design specification review includes that the rule mapping uses a decision tree algorithm to perform specification matching according to the attribute values of the BIM structural model; based on FuzzyLogic optimization, for BIM structural parameters that do not meet the fixed threshold, a fuzzy logic method is used to provide optimization suggestions within the tolerance range and output the review results.

[0028] It should also be noted that the input features of the decision tree algorithm include parameters such as the dimensions, material types, and load requirements of each structural element in the BIM model. The parameters reflect whether the structural design meets the requirements of the design specifications. The structural elements include beams, columns, walls, etc.; the processing flow of the decision tree algorithm includes: using the labeled BIM design data for training to generate a decision tree model; based on the rules in the design specifications, the splitting nodes of the decision tree correspond to different parameters of the structural elements. The parameters include the span of a beam, the thickness of a wall, etc.; according to the classification results generated by the decision tree, outputting structural review opinions and providing necessary design optimization suggestions; continuously optimizing the decision tree model to improve the review accuracy of the structural design.

[0029] It should also be noted that the automatic generation of the review report based on natural language processing technology includes parsing the BIM design specification using the BERT pre-trained language model to extract core technical parameters, including load standards and structural constraint conditions; automatically generating an abstract of the review report through a Seq2Seq structure to summarize BIM structural design problems; using the TextRank algorithm to automatically screen high-risk review points and provide key correction suggestions.

[0030] It should also be noted that a preferred solution for the source and format of the training data of the Seq2Seq model specifically includes collecting a training data set containing BIM model design data and corresponding review reports. The BIM design data includes parameters such as the dimensions, loads, and materials of each structural element, and the review report includes design compliance assessment, error point annotation, modification suggestions, etc.; converting the BIM design data into a standardized format, which is JSON or XML, and each data item includes the attributes and design parameters of the structural element; converting the review report into a text format as the target output of the Seq2Seq model, and performing word segmentation and stop word removal on the output text; performing standardized processing on the BIM design data, normalizing numerical data to ensure the consistency of the scales of different features; using the Seq2Seq model for training, and using the encoder to encode the input data during the training process and generating the review report text through the decoder.

[0031] It should also be noted that a preferred solution for automatically screening high-risk review points using the TextRank algorithm specifically includes preprocessing the BIM design specification using the TextRank algorithm to remove redundant characters and irrelevant content, and performing sentence segmentation and standardization processing; regarding each sentence in the BIM design specification as a node in the graph, calculating the similarity between sentences, and calculating the semantic similarity between sentences through a word vector model; constructing a sentence graph based on the sentence similarity, where the edges between nodes represent the sentence similarity; using the PageRank algorithm to perform iterative calculations on the sentence graph to obtain the score of each sentence, thereby determining the most critical sentences in the text; screening out the sentences with high scores as high-risk review points according to the calculated scores and providing key correction suggestions.

[0032] It should also be noted that the rule engine built into the AI large model realizes the automatic extraction of building design standards and specification review through the BIM structural design knowledge graph constructed based on Ontology semantic modeling; by converting building design standards into logical inference rules, the system can perform precise structural design specification matching according to the specific data and attribute values of the BIM model, quickly identify the parts that do not meet the design requirements. This process not only reduces the workload of manual inspection but also automatically reviews a large amount of design data through efficient rule mapping (based on the decision tree algorithm), greatly improving efficiency and accuracy; at the same time, based on the method optimized by Fuzzy Logic, it can provide optimization suggestions within the tolerance range for BIM structural parameters that do not meet the design specifications through fuzzy logic methods, avoiding overcorrection caused by strict standards and enhancing the feasibility and practicality of the design scheme; the generation of an automatic review report based on natural language processing technology uses the BERT pre-trained language model to parse the BIM design specification, extract core technical parameters, and automatically summarize the possible problems in the design; generate a summary of the review report through the Seq2Seq structure and use the TextRank algorithm to screen out high-risk review points and automatically propose correction suggestions, greatly simplifying the report generation process and improving the quality and pertinence of the report; this process of automatic review report not only speeds up the project process but also improves the accuracy of the review process, enabling potential problems in the structural design process to be discovered and processed in a timely manner.

[0033] S3: Perform BIM model collision detection based on deep learning algorithms, identify potential structural conflicts, combine three-dimensional visualization to display error points, and provide adjustment suggestions based on historical data and optimization algorithms.

[0034] Furthermore, identifying potential structural conflicts includes using the PointNet++ deep learning framework for 3D point cloud data analysis of the BIM structural model to detect beam-column collisions and floor penetration problems.

[0035] It should also be noted that a preferred solution for 3D point cloud data analysis specifically includes extracting three-dimensional geometric data from the BIM model, including the geometric shapes and spatial coordinates of structural elements; converting the extracted geometric data into point cloud data, where the contours and surface information of each structural element are converted into a set of discrete points containing XYZ coordinates, and the point cloud data is generated through three-dimensional geometric transformation; performing denoising processing and downsampling on the generated point cloud data to remove redundant points and reduce data complexity; normalizing the processed point cloud data into a data format suitable for deep learning algorithms, such as an XYZ coordinate matrix; using a spatial collision detection algorithm to detect whether a collision occurs between beams and columns, and the collision detection determines whether there is a spatial overlap by calculating the relative positions of beam and column nodes in the point cloud data; using a geometric detection algorithm to detect whether a floor slab penetrates other structures, and the penetration detection determines whether there is a penetration phenomenon by calculating the distances between the floor slab and other structural elements (such as columns, walls).

[0036] It should be noted that providing adjustment suggestions includes reasoning about the mechanical relationships between floor slabs, beams, and walls through GNN to identify them and providing adjustment solutions; using an LSTM prediction model combined with historical building project data to provide BIM structure optimization suggestions based on big data analysis and generating 3D interactive optimization solutions.

[0037] It should also be noted that a preferred solution for constructing the GNN graph structure specifically includes defining each structural element in the BIM model as a node in the graph, where the node includes the attribute information of the structural element, and the attribute information includes the dimensions, materials, load requirements, etc. of the structural element; defining the edges between nodes, where the edges represent the mechanical relationships or geometric connection relationships between nodes, and the attributes of the edges include the transmitted stress, the distances between adjacent nodes, etc.; constructing the adjacency matrix of the graph based on the relationships of structural elements in the BIM model, where the connections between each node and other nodes represent the corresponding mechanical or geometric relationships; converting the constructed graph data into an input data format suitable for graph neural networks, including the adjacency matrix, node feature matrix, and edge feature matrix; a preferred solution for the training process of LSTM specifically includes collecting structural design data from historical building projects, including the structural parameters of each project and their optimization measures; converting the historical data into a time series data format suitable for long short-term memory networks (LSTM), where each data point includes a time series representing the various parameters of the structural design; constructing an LSTM model, with the input of the model being the historical data of the structural design and the output being the predicted structural optimization solution; training the LSTM model using historical building project data and optimizing the model parameters through the backpropagation algorithm; using the trained LSTM model to predict new BIM design data and generating a structural design optimization solution.

[0038] It should also be noted that for BIM model collision detection based on deep learning algorithms, the PointNet++ deep learning framework is used to analyze the 3D point cloud data of the BIM structural model, so as to identify potential structural conflicts. Especially for common structural problems such as beam-column collisions and floor penetrations, the deep learning model can automatically identify these problems and output the precise error locations through training on a large amount of structural data, improving the efficiency and accuracy of collision detection; by combining three-dimensional visualization technology, the system can directly display the error points in the three-dimensional space, enabling engineers to intuitively see the problem locations and avoiding conflict problems that are difficult to detect in traditional two-dimensional drawings; by combining historical building project data with the LSTM prediction model, the optimization algorithm based on big data analysis provides targeted adjustment suggestions for the BIM structure, predicts the best solution for structural adjustment according to the laws and trends in historical data, and provides the optimized adjustment solution in the three-dimensional interactive interface, improving the efficiency and precision of design adjustment; this adjustment suggestion not only provides solutions based on existing problems, but also takes into account historical data and big data analysis, providing predictive solutions for possible future structural problems, improving the adaptability and flexibility of the project; through the combination of this deep learning and big data analysis, the overall intelligent level of the BIM model is improved, making the building design process more precise, intelligent and efficient.

[0039] Example 2 is an embodiment of the present invention, which provides a BIM model structure design review method based on a large model. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0040] First, BIM models were imported from multiple architectural design software, and the data had various formats. To process this data, preprocessing was carried out based on large AI models. The specific operations included: using the RNN algorithm to detect outliers in the structural parameters (such as the span of beams, the thickness of walls, etc.) in the BIM model; through training on historical design data, automatically identifying and filling in outliers, reducing manual intervention and ensuring data accuracy; converting the BIM model data generated by different software into a unified IFC format and using the Transformer model to standardize the data format, improving data compatibility during cross-platform collaborative work; text information cleaning: using a semantic parsing model to clean unstructured text information, such as redundant design annotations, duplicate BIM component labels, etc., making subsequent analysis more efficient; structural design review After data preprocessing, the BIM model was sent to the review module. Using a knowledge graph based on Ontology, architectural design standards were converted into inference rules. Combining with the actual data (such as structural parameters) of the BIM model, structural design specification matching was carried out through a decision tree algorithm; using the BERT pre-trained model to extract the core technical parameters in the design specification, using a Seq2Seq structure to generate a summary of the review report, and screening high-risk review points through the TextRank algorithm to automatically provide correction suggestions. This process greatly improved the accuracy and efficiency of design review; using the PointNet++ deep learning framework to analyze the 3D point cloud data of the BIM model, automatically identifying potential problems such as beam-column collisions and floor penetrations; based on historical project data and the LSTM prediction model, combined with a graph neural network (GNN) to provide targeted adjustment solutions, displaying the error points to designers through 3D visualization and providing real-time optimization suggestions. Refer to Table 1 for recording and analyzing the experimental data.

[0041] Table 1 Experimental Data Record Sheet

[0042] Experiments show that Method B (the system of the present invention) by adopting an outlier detection algorithm based on a large AI model and a format conversion technology of the Transformer model, successfully reduces the data preprocessing time from 120 seconds of Method A (the existing system) to 75 seconds, a reduction of 37.5%. This improvement stems from the introduction of automated processing by the AI model, avoiding the multi-step operations that need to be carried out manually in traditional methods; the automated outlier detection not only improves the accuracy of the data, but also reduces the time of manual intervention and possible errors. In addition, the application of the Transformer model in BIM data format conversion eliminates the format compatibility problems in the manual conversion process, providing a unified standard format for subsequent analysis and further enhancing the overall efficiency; in terms of the accuracy of error detection, the accuracy of Method B (the system of the present invention) has been increased from 85% of the prior art to 95%, with an increase of 11.76%. This significant improvement is mainly due to the application of the large AI model, especially the RNN algorithm. By training on historical BIM data, Method B (the system of the present invention) can more accurately identify abnormal situations in structural parameters. Compared with traditional methods, Method B can effectively reduce the incidence of false detection and missed detection, thus providing more reliable data support for subsequent structural review and optimization; for the format conversion time, Method B (the system of the present invention) successfully shortens the conversion time from 300 seconds of the prior art to 120 seconds, a reduction of 60%. This improvement shows that through automated IFC format conversion, not only the time cost brought by manual conversion is eliminated, but also the data consistency and compatibility are enhanced. In the scenario where BIM models involve data generated by multiple software, automated conversion can avoid errors caused by format differences, thus ensuring the effectiveness and unity of data in cross-platform collaboration; the collision detection and optimization suggestion generation links also show obvious advantages. Method B (the system of the present invention) through the deep learning frameworks PointNet++ and LSTM model, successfully reduces the collision detection time from 250 seconds to 150 seconds, and the optimization suggestion generation time from 180 seconds to 120 seconds, with an increase of 40% and 33.33% respectively; the PointNet++ framework can automatically identify potential structural conflicts such as beam-column collisions and floor penetrations by analyzing the 3D point cloud data of the BIM model, while the LSTM model combined with historical building project data provides optimization suggestions based on big data analysis; compared with traditional methods, Method B (the system of the present invention) not only improves the accuracy of collision detection, but also provides a more accurate structural optimization plan, enhancing the efficiency of design adjustment; in terms of the total processing time, the comprehensive processing time of Method B (the system of the present invention) is reduced from 850 seconds of the existing system to 565 seconds, a reduction of 33.53%. This overall optimization is the result of the interaction of technological advancements in multiple aspects, demonstrating the powerful advantages of the integration of AI, big data, and deep learning technologies. Through the coordinated efforts of data preprocessing, format conversion, error detection, collision detection, and optimization, Method B (the system of the present invention) not only improves the processing efficiency but also ensures higher processing accuracy and the level of design intelligence.

[0043] Compared with the prior art, the present invention not only improves the work efficiency but also enhances the intelligent processing ability of the system, providing a more efficient and accurate BIM model processing solution for the architectural design industry, and having great practical application value and market potential.

[0044] Embodiment 3, an embodiment of the present invention, provides a BIM model structure design review system based on a large model, including a data processing module, a review module, and an optimization module.

[0045] Among them, the data processing module is used for preprocessing BIM model data based on an AI large model.

[0046] It should be noted that the operation of the system starts from the data processing module. Through the technology based on the AI large model for preprocessing BIM model data, first, the data processing module performs outlier detection on the input BIM model data, automatically identifying and filling in the missing or abnormal data in the model to ensure the integrity and accuracy of the data; then, the data processing module performs data format conversion generated by different software, unifying the data of the BIM model into the IFC standard format to ensure data compatibility between different platforms; at the same time, unstructured text information is cleaned and optimized through a semantic parsing model, removing redundant design annotations and duplicate component labels, improving the efficiency and data quality of subsequent processing. These preprocessing steps provide a clean, standardized, and accurate basic data for the subsequent operations of the review module and the optimization module.

[0047] The review module includes a rule engine module and a report generation module. The rule engine module is used for reviewing the structural design specifications of the BIM model by using the rule engine built in the AI large model, and the report generation module is used for automatically generating a review report based on natural language processing technology.

[0048] It should be noted that after preprocessing by the data processing module, the processed data enters the review module. The rule engine module first conducts a review of the structural design specifications of the BIM model based on the rule engine built into the AI large model and in combination with the constructed BIM structural design knowledge graph; by matching with the architectural design standards, the system can identify the parts that do not meet the specifications and provide optimization suggestions within the tolerance range for the parameters that do not meet the standards according to the decision tree algorithm and fuzzy logic. The rule engine module ensures a comprehensive inspection of the model data and generates a structural review result; based on the review process, the report generation module automatically generates a review report. By adopting natural language processing technology, the BERT pre-trained language model parses the design specification, automatically extracts the core technical parameters and key information, generates a summary of the review report using the Seq2Seq structure, and screens out high-risk points through the TextRank algorithm to provide key correction suggestions. After these information are aggregated, a detailed review report is formed, providing clear and accurate feedback to the designers. The optimization module receives the information provided by the review module and conducts subsequent structural conflict detection and optimization adjustment. The optimization module receives the information provided by the review module and conducts subsequent structural conflict detection and optimization adjustment.

[0049] The optimization module includes a conflict identification module and an adjustment module. The conflict identification module is used to perform BIM model collision detection based on deep learning algorithms to identify potential structural conflicts. The adjustment module is used to combine three-dimensional visualization to display the error points and provide adjustment suggestions according to historical data and optimization algorithms.

[0050] It should be noted that after the review module generates a report, the system enters the optimization stage. The conflict identification module uses deep learning algorithms, such as the PointNet++ framework, to perform collision detection on the three-dimensional point cloud data of the BIM model. This module can automatically identify potential structural conflict problems, such as beam-column collisions and floor penetrations, and identify and locate the error points in the model. Through the training of the deep learning model, the conflict identification module improves the accuracy and efficiency of collision detection; after identifying the structural conflicts, the adjustment module combines three-dimensional visualization technology to present the error points to the designers in an intuitive way. The designers can clearly see the problem points in the model through the interactive interface. At the same time, the adjustment module combines historical building data with the LSTM prediction model and provides targeted adjustment suggestions based on optimization algorithms; these suggestions not only propose solutions based on the existing conflict problems, but also predict possible structural problems through the analysis of historical data and generate adjustment plans. The optimization suggestions at this stage improve the accuracy and adaptability of the BIM model design.

[0051] The data processing module provides a unified data input for the optimization module by converting the original BIM data into a standardized format. The conflict identification and adjustment suggestions of the optimization module rely on the accuracy and consistency of the data. If the data processing module fails to correctly handle outliers, format conversion, or text cleaning, the results of the optimization module may be affected, leading to incorrect conflict identification or inaccurate adjustment suggestions. During the operation of the system, the suggestions and conflict detection results output by the optimization module can be fed back to the data processing module for further data optimization. For example, during the conflict detection process, some incorrect or abnormal data may be identified and marked as problems by the optimization module, and the system can feed this data back to the data processing module for further processing, thus continuously improving data optimization and accuracy.

Claims

1. A BIM model structure design review method based on a large model, characterized in that, Including: Preprocessing BIM model data based on large AI models; Using the built-in rule engine of large AI models to conduct structural design specification reviews on BIM models, and automatically generating review reports based on natural language processing technology; Performing BIM model collision detection based on deep learning algorithms, identifying potential structural conflicts, combining three-dimensional visualization to display error points, and providing adjustment suggestions according to historical data and optimization algorithms; The rule engine includes constructing a BIM structural design knowledge graph based on Ontology semantic modeling, extracting rules from architectural design standards, and converting them into logical inference rules; Generating a review report includes automatically generating a BIM review report based on natural language processing technology, using the BERT pre-trained language model to extract technical parameters, generating a review report summary through the Seq2Seq structure, and screening high-risk review points using the TextRank algorithm; Providing adjustment suggestions includes combining GNN and LSTM prediction models to enable the BIM model to identify conflicts in the design, making dynamic optimization suggestions based on historical building data, and generating a 3D interactive optimization plan.

2. The BIM model structure design review method based on a large model according to claim 1, characterized in that: The data preprocessing includes, Using the outlier detection algorithm of large AI models to train an RNN based on historical BIM design data, automatically identifying outliers in BIM structural parameters, and automatically filling them; Converting the unified format of BIM structural data through the Transformer model, and parsing BIM data generated by different software into the IFC standard format; Using a semantic parsing model to clean unstructured text information, removing redundant design annotations and duplicate BIM component labels.

3. The BIM model structure design review method based on a large model according to claim 1 or 2, characterized in that: The rule engine includes, Based on Ontology semantic modeling, constructing a BIM structural design knowledge graph, extracting the rules of architectural design standards, and converting the rules of architectural design standards into logical inference rules; The knowledge structure of the knowledge graph includes a design specification layer, a rule conversion layer, and a BIM data layer; The design specification layer includes regulatory provisions; The rule conversion layer includes logical inference rules; The BIM data layer includes actual design parameters extracted from BIM data.

4. The method for reviewing the BIM model structure design based on the large model according to claim 3, wherein: The structural design specification review includes, Rule mapping uses a decision tree algorithm to perform specification matching according to the attribute values of the BIM structural model; Based on FuzzyLogic optimization, for BIM structural parameters that do not meet the fixed threshold, a fuzzy logic method is used to provide optimization suggestions within the tolerance range and output the review results.

5. The BIM model structure design review method based on a large model according to claim 1 or 4, characterized in that: The automatically generating a review report based on natural language processing technology includes, Using the BERT pre-trained language model to parse the BIM design specification, and extracting core technical parameters, including load standards and structural constraint conditions; Automatically generating a review report summary through the Seq2Seq structure to summarize BIM structural design problems; Using the TextRank algorithm to automatically screen high-risk review points and provide key correction suggestions.

6. The method for BIM model structure design review based on large models as claimed in claim 1, 2 or 4, characterized in that: The identifying potential structural conflicts includes, Using the PointNet++ deep learning framework to perform 3D point cloud data analysis of the BIM structural model, and detecting beam-column collisions and floor penetration problems.

7. The method for reviewing the BIM model structure design based on the large model according to claim 6, characterized in that: The providing adjustment suggestions includes, Conduct BIM component relationship reasoning through GNN, identify the mechanical relationships among floor slabs, beams, and walls, and provide adjustment plans; Adopt an LSTM prediction model combined with historical building project data, provide BIM structure optimization suggestions based on big data analysis, and generate 3D interactive optimization plans.

8. A BIM model structure design review system based on a large model, characterized in that: Include a data processing module, a review module, and an optimization module; The data processing module is used for preprocessing BIM model data based on a large AI model; The review module includes a rule engine module and a report generation module. The rule engine module is used to conduct structural design specification reviews on the BIM model using the built-in rule engine of the large AI model, and the report generation module is used to automatically generate review reports based on natural language processing technology; The optimization module includes a conflict identification module and an adjustment module. The conflict identification module is used to perform BIM model collision detection based on deep learning algorithms to identify potential structural conflicts. The adjustment module is used to combine three-dimensional visualization to display error points and provide adjustment suggestions based on historical data and optimization algorithms.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the large model-based BIM model structural design review method described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the large model-based BIM model structural design review method described in any one of claims 1 to 7.

Images