Method and system for rapid dismantling of BIM models

Abstract

This invention relates to the technical field of building information modeling, and in particular to a method and system for rapid dismantling of BIM models. By using a pre-integrated decision tree classification algorithm and K-means clustering algorithm, the present invention can improve the processing efficiency of large, complex BIM models and effectively deal with the challenge of large amounts of data. High-efficiency data hierarchy and classification optimizes the overall model processing procedure, enabling deeper data correlation and analysis in multi-disciplinary areas such as architecture, construction, electrical engineering, and plumbing, thereby improving the convenience and efficiency of interdisciplinary collaboration. Furthermore, the additional data optimization algorithm and logical decomposition speed up the identification and classification of various components in the model. [Solution] The method includes steps of hierarchizing the BIM model using an integrated decision tree classification algorithm and K-means clustering algorithm based on the type and functional requirements of architectural elements, and further classifying the type, function and planning stage to generate a hierarchical data structure.

Description

[Technical Field]

[0001] The present invention relates to the technical field of building information modeling, and in particular to a method and system for rapid dismantling of BIM models. [Background technology]

[0002] Building Information Modeling (BIM) is a digital technology-based solution for the construction industry that uses 3D modeling software and other related technologies to create and manage digital representations of the physical and functional characteristics of building projects. BIM technology provides decision support, improves design efficiency, optimizes construction and maintenance, and enhances collaboration and communication among project stakeholders throughout the entire process of designing, constructing, and maintaining building projects, including multi-disciplinary aspects such as architecture, construction, electrical, and plumbing.

[0003] The rapid dismantling method for BIM models aims to improve the efficiency and accuracy of building design and construction processes by quickly breaking down large, complex BIM models into smaller, more manageable, and more analyzable components. This approach reduces the need for manual intervention, reduces errors, and improves overall project management efficiency. In many cases, this rapid dismantling method is achieved through the application of advanced computer vision technology and deep learning algorithms. For example, analyzing and processing BIM models using convolutional neural networks can accurately identify and classify the various components within the model. Furthermore, this method combines object detection and semantic segmentation technology to accurately distinguish between architectural elements such as walls, floors, and piping, thereby not only speeding up model processing but also improving the accuracy and efficiency of the demolition process.

[0004] Traditional BIM model processing methods struggle to efficiently process large, complex BIM models, especially when data volumes increase dramatically. Their relatively limited data processing capabilities reduce the efficiency of collaboration and communication throughout the entire process, from design to construction and maintenance. In areas involving many disciplines, such as the integrated processing of architecture, structural engineering, electrical engineering, and plumbing, data correlation and analysis cannot be fully realized, making collaboration between disciplines somewhat inconvenient and making it difficult to quickly identify and classify the various components within the model. While existing technologies provide a solid foundation for BIM model processing, there is still room for improvement in terms of processing efficiency, depth of data integration, and degree of automation. Summary of the Invention [Problem to be solved by the invention]

[0005] The present invention aims to propose a method and system for quickly dismantling a BIM model to solve the shortcomings of the prior art. [Means for solving the problem]

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for rapid dismantling of a BIM model includes the following steps:

[0007] S1: Based on the type and function requirements of architectural elements, the BIM model is hierarchically organized using the integrated decision tree classification algorithm and K-means clustering algorithm, and then the type, function, and planning stage are classified to generate a hierarchical data structure.

[0008] S2: Based on the hierarchical data structure, a data optimization algorithm based on heuristic rules is used to perform differentiation processing on the main domain and the subordinate domain, and in addition, the decomposition procedure is optimized to generate an optimized decomposition plan.

[0009] S3: Based on the optimized decomposition plan, a decomposition engine based on predicate logic is used to perform logical reasoning, and in addition, a decomposition path is selected to generate a logical decomposition path.

[0010] S4: Based on the logical decomposition path, an incremental change detection algorithm is used to update the changed parts in the model to generate updated partitioned data.

[0011] S5: Based on the updated divided data, the data is converted into streaming data using a data compression and optimization algorithm to generate streaming divided data.

[0012] S6: Based on the streaming divided data, data quality is optimized using data integrity verification to generate quality verified data.

[0013] S7: Based on the quality verified data, the interdependencies between components are identified by means of network topology analysis, and an inter-component relationship analysis report is generated.

[0014] S8: Based on the inter-component relationship analysis report, a data fusion and integration algorithm is used to integrate the results of the hierarchical data structure, the optimized disassembly plan, the logical disassembly path, the updated division data, the streaming division data, and the quality verified data, and the BIM model is completely disassembled to generate a disassembled BIM model.

[0015] The hierarchical data structure includes a data hierarchy of multiple functions such as structural, electrical, and piping. The logical decomposition path specifically refers to a decomposition method and sequence path derived based on a logical formula. The updated partition data specifically refers to information for changing and updating the changed parts of the model. The streaming partition data specifically refers to an optimized dataset. The quality verified data specifically refers to a verified dataset. The inter-component relationship analysis report specifically refers to a report showing the dependency relationships and network connection strength between components.

[0016] As a further proposal of the present invention, the step of hierarchizing the BIM model using the integrated decision tree classification algorithm and K-means clustering algorithm based on the type and function requirements of building elements, and further classifying the type, function and planning stage to generate a hierarchical data structure includes the following sub-steps:

[0017] S101: Based on the type and functional requirements of architectural elements, the BIM model data is initially classified using a decision tree classification algorithm to generate initial classification data.

[0018] S102: Based on the primary classified data, a K-means clustering algorithm is used to perform feature division on the data to generate feature classified data.

[0019] S103: Based on the function-classified data, an association rule learning algorithm is used to analyze the relationship between the data and the planning stage, thereby generating planning stage-related data.

[0020] S104: Based on the planning stage related data, data integration is performed as a means to generate a hierarchical data structure.

[0021] In a further proposal of the present invention, the step of performing differentiation processing on the primary and subordinate regions based on the hierarchical data structure using a data optimization algorithm based on heuristic rules, and optimizing the decomposition procedure to generate an optimized decomposition plan, includes the following substeps:

[0022] S201: Based on the hierarchical data structure, a heuristic rule-based analysis algorithm is used to analyze data in key areas to generate a key area processing plan.

[0023] S202: Analyze the dependent regions using a simplified processing algorithm based on the main region processing plan to generate a dependent region processing plan.

[0024] S203: Based on the main area processing plan and the dependent area processing plan, a data fusion technique and a balancing processing strategy are employed to generate a comprehensive data processing plan.

[0025] S204: Based on the comprehensive data processing plan, a multi-objective optimization technique is employed to perform differentiated data processing to generate an optimized decomposition plan.

[0026] As a further proposal of the present invention, the step of performing logical inference using a decomposition engine based on predicate logic based on the optimized decomposition plan, and additionally selecting decomposition paths to generate logical decomposition paths, includes the following sub-steps:

[0027] S301: Based on the optimized decomposition plan, a predicate logic algorithm is used to perform elementary logical reasoning to generate an elementary logical reasoning result.

[0028] S302: Based on the result of the preliminary logical reasoning, a logical optimization technique is employed to further optimize the logical structure of the decomposition path, thereby generating a logically optimized path result.

[0029] S303: Based on the logical optimization route result, a weighted sorting algorithm is used to rank the route options, generating a weighted ranked decomposition result.

[0030] S304: Based on the weighted ranking decomposition result, a decision path analysis method is adopted to select an appropriate decomposition path, and in addition, logical reasoning efficiency is optimized to generate a logical decomposition path.

[0031] As a further proposal of the present invention, the step of updating the changed parts in the model based on the logical decomposition path using an incremental change detection algorithm to generate updated partitioned data comprises the following sub-steps:

[0032] S401: Based on the logical decomposition path, incremental change detection techniques are employed to identify and update the changed parts in the model to generate preliminary update data.

[0033] S402: Based on the preliminary update data, synchronized update data is generated by maintaining data consistency using a data synchronization protocol.

[0034] S403: Based on the synchronous update data, process the data differences by way of a difference query in the database to generate differentiated update data.

[0035] S404: Based on the differentiated update data, a data merging algorithm is used to align the data to generate updated divided data.

[0036] As a further proposal of the present invention, the step of streaming data using a data compression and optimization algorithm to generate streaming divided data based on the updated divided data includes the following sub-steps:

[0037] S501: Based on the updated divided data, a data compression / encoding technique is employed to reduce the amount of data and generate elementary compressed data.

[0038] S502: The data structure is optimized by data reorganization based on the preliminary compressed data, and optimized data is generated.

[0039] S503: The data format is reconstructed based on the optimized data as a means for data reconstruction, and reconstructed data is generated.

[0040] S504: Based on the reconstructed data, the data is streamed using a data streaming framework to generate streaming divided data.

[0041] As a further proposal of the present invention, the step of optimizing data quality by means of data integrity verification based on the streaming divided data to generate quality verified data comprises the following sub-steps:

[0042] S601: Based on the streaming divided data, a preliminary quality verification is performed using data integrity verification to generate preliminary quality verification data.

[0043] S602: Based on the preliminary quality verification data, a data cleansing algorithm is used to remove inconsistent and erroneous data to generate cleansed data.

[0044] S603: Based on the cleansed data, an integrity enhancement technique is employed to optimize the accuracy of the data to generate integrity assessment data.

[0045] S604: Optimizing data quality using a comprehensive data quality assessment framework based on the integrity assessment data to generate quality verified data.

[0046] As a further proposal of the present invention, the step of identifying interdependencies between components by means of network topology analysis based on the quality verified data and generating an inter-component correlation analysis report comprises the following sub-steps:

[0047] S701: Based on the quality verified data, a graph theory algorithm is used to analyze the connectivity between components to generate a preliminary component relationship diagram.

[0048] S702: Based on the preliminary inter-component relationship diagram, components and dependency paths are identified by network flow analysis, and a component dependency diagram is generated.

[0049] S703: Based on the component dependency diagram, a cluster analysis algorithm is used to distinguish groups of multiple components, and a component group analysis result is generated.

[0050] S704: Based on the component group analysis result, further analyze the dependency and interaction relationships between the components by considering multiple network parameters as a means of network topology optimization, and generate an inter-component relationship analysis report.

[0051] As a further proposal of the present invention, the step of integrating the results of the hierarchical data structure, the optimized disassembly plan, the logical disassembly path, the updated division data, the streaming division data, and the quality verified data using a data fusion and integration algorithm based on the inter-component relationship analysis report, and completing the disassembly of the BIM model to generate a disassembled BIM model includes the following substeps:

[0052] S801: Based on the inter-component correlation analysis report, the multi-stage data results are preliminarily matched by means of related data fusion to generate preliminarily matched data.

[0053] S802: Based on the preliminary matched data, a dimensional analysis method is adopted to match multifaceted data to create a comprehensive multidimensional data view, and a multidimensional fusion data set is generated.

[0054] S803: Based on the multidimensional fusion dataset, a data processing technique is employed to optimize and simplify the data to generate a subdivided dataset.

[0055] S804: Based on the disaggregated data set, a comprehensive data management strategy is adopted to perform final fusion and reconciliation of data, and in addition, proofreading of information consistency and completeness is performed to generate a disassembled BIM model.

[0056] The BIM model rapid dismantling system is used to implement the above-mentioned BIM model rapid dismantling method, and includes an elementary classification module, a function division module, a planning stage related module, a data hierarchy construction module, a main domain processing module, a subordinate domain processing module, a comprehensive data processing module, a multi-objective optimization module, a logical dismantling path module, and a data fusion and integration module.

[0057] The preliminary classification module uses a decision tree classification algorithm to preliminary classify BIM model data based on the type and functional requirements of building elements, thereby generating preliminary classified data.

[0058] The function division module performs function division on the data using a K-means clustering algorithm based on the preliminary classified data to generate function classified data.

[0059] The planning stage association module analyzes the relationship between the data and the planning stage based on the function classification data using an association rule learning algorithm to generate planning stage association data.

[0060] The data hierarchy construction module performs data integration by means of data merging based on the planning stage related data to generate a hierarchical data structure.

[0061] The main area processing module analyzes the main area data based on the hierarchical data structure using a heuristic rule-based analysis algorithm to generate a main area processing plan.

[0062] The dependent region processing module analyzes the dependent regions using a simplified processing algorithm based on the main region processing scheme to generate a dependent region processing scheme.

[0063] The comprehensive data processing module generates a comprehensive data processing plan by adopting a data fusion technique and a balancing processing strategy based on the main area processing plan and the dependent area processing plan.

[0064] The multi-objective optimization module employs a multi-objective optimization technique to perform differentiated data processing based on the comprehensive data processing plan, thereby generating an optimized decomposition plan.

[0065] The logical decomposition path module generates a logical decomposition path by performing logical reasoning using a decomposition engine based on predicate logic based on the optimized decomposition plan.

[0066] The data fusion and alignment module uses a data fusion and alignment algorithm based on the logical decomposition path to align the results of the primary classification module, the function division module, the planning stage related module, the data hierarchy construction module, the main area processing module, the dependent area processing module, the comprehensive data processing module, the multi-objective optimization module, and the logical decomposition path module, to complete the decomposition of the BIM model and generate a disassembled BIM model.

[0067] The present invention has the following advantages and beneficial effects over the prior art.

[0068] This invention uses a pre-integrated decision tree classification algorithm and K-means clustering algorithm to improve the processing efficiency of large, complex BIM models and effectively address the challenge of large amounts of data. Highly efficient data hierarchy and classification optimizes the overall model processing procedure, enabling deeper data correlation and analysis across multiple disciplines, such as architecture, structural engineering, electrical engineering, and plumbing, thereby improving the convenience and efficiency of interdisciplinary collaboration. Furthermore, additional data optimization algorithms and logical decomposition not only speed up the identification and classification of various components in the model, but also improve the degree of automation. This invention significantly improves processing efficiency, the depth of data fusion, and the degree of automation, effectively overcoming the limitations of existing technologies and providing a more advanced and comprehensive solution for BIM model processing. [Brief explanation of the drawings]

[0069] [Figure 1] FIG. 1 is a schematic diagram of the working procedure of the present invention. [Figure 2] FIG. 2 is a flow diagram of the S1 subdivision of the present invention. [Figure 3] FIG. 3 is a flow diagram of the S2 subdivision of the present invention. [Figure 4] FIG. 4 is a flow diagram of the S3 subdivision of the present invention. [Figure 5] FIG. 5 is a flow diagram of the S4 subdivision of the present invention. [Figure 6]FIG. 6 is a flow diagram of the S5 subdivision of the present invention. [Figure 7] FIG. 7 is a flow diagram of the S6 subdivision of the present invention. [Figure 8] FIG. 8 is a flow diagram of the S7 subdivision of the present invention. [Figure 9] FIG. 9 is a flow diagram of the S8 subdivision of the present invention. [Figure 10] FIG. 10 is a system flow diagram of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0070] We will now describe the present invention in more detail in conjunction with the drawings and examples, so as to make the objectives, technical solutions and advantages of the present invention clearer. It should be understood that the specific examples described herein are only used to illustrate the present invention, without limiting the present invention.

[0071] In describing the present invention, directions or positional relationships indicated by terms such as "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," and "outside" are merely for the convenience of describing and simplifying the present invention based on the directions or positional relationships shown in the drawings, and should not be understood to imply or explicitly state that the described devices or elements must have a particular orientation or be positioned or operated in a particular direction. Therefore, this should not be construed as imposing limitations on the present invention. Furthermore, in describing the present invention, "plurality" means two or more than two, unless expressly and specifically limited otherwise. [Example]

[0072] Example 1 As shown in the drawing, the method for rapid demolition of a BIM model includes the following steps:

[0073] S1: Based on the type and function requirements of architectural elements, the BIM model is hierarchically organized using the integrated decision tree classification algorithm and K-means clustering algorithm, and then the type, function, and planning stage are classified to generate a hierarchical data structure.

[0074] S2: Based on the hierarchical data structure, a data optimization algorithm based on heuristic rules is used to perform differentiation processing on the main domain and the subordinate domain, and in addition, the decomposition procedure is optimized to generate an optimized decomposition plan.

[0075] S3: Based on the optimized decomposition plan, a decomposition engine based on predicate logic is used to perform logical reasoning, and in addition, a decomposition path is selected to generate a logical decomposition path.

[0076] S4: Based on the logical decomposition path, an incremental change detection algorithm is used to update the changed parts in the model to generate updated partitioned data.

[0077] S5: Based on the updated divided data, the data is converted into streaming data using a data compression and optimization algorithm to generate streaming divided data.

[0078] S6: Based on the streaming divided data, data quality is optimized using data integrity verification to generate quality verified data.

[0079] S7: Based on the quality verified data, the interdependencies between components are identified by means of network topology analysis, and an inter-component relationship analysis report is generated.

[0080] S8: Based on the inter-component relationship analysis report, a data fusion and integration algorithm is used to integrate the results of the hierarchical data structure, the optimized disassembly plan, the logical disassembly path, the updated division data, the streaming division data, and the quality verified data, and the BIM model is completely disassembled to generate a disassembled BIM model.

[0081] The hierarchical data structure includes a data hierarchy of multiple functions such as structural, electrical, and piping. The logical decomposition path specifically refers to a decomposition method and sequence path derived based on a logical formula. The updated partition data specifically refers to information for changing and updating the changed parts of the model. The streaming partition data specifically refers to an optimized dataset. The quality verified data specifically refers to a verified dataset. The inter-component relationship analysis report specifically refers to a report showing the dependency relationships and network connection strength between components.

[0082] This method uses a pre-built decision tree classification algorithm, a K-means clustering algorithm, and a heuristic rule-based data optimization algorithm to quickly and accurately classify and optimize BIM model data, thereby achieving significant time-saving results. Furthermore, its flexibility and customizability make it suitable for various projects and situations, allowing for customization of decomposition rules according to specific needs, thereby enhancing applicability. Furthermore, this method uses logical reasoning and data integrity verification to ensure the consistency and accuracy of decomposition results, thereby improving the quality of model data. It also uses data compression and optimization algorithms and data streaming processing to reduce the amount of decomposed data, improving the efficiency of subsequent analysis and application. It also uses quality-verified data and inter-component correlation analysis reports to identify potential design and construction issues, improving the project team's understanding and familiarity with the model. The automation of procedures helps reduce the costs of plan decomposition and maintenance.

[0083] As shown in Figure 2, the step of hierarchizing the BIM model using the integrated decision tree classification algorithm and K-means clustering algorithm based on the type and function requirements of building elements, and further classifying the type, function and planning stage to generate a hierarchical data structure includes the following substeps:

[0084] S101: Based on the type and functional requirements of architectural elements, the BIM model data is initially classified using a decision tree classification algorithm to generate initial classification data.

[0085] S102: Based on the primary classified data, a K-means clustering algorithm is used to perform feature division on the data to generate feature classified data.

[0086] S103: Based on the function-classified data, an association rule learning algorithm is used to analyze the relationship between the data and the planning stage, thereby generating planning stage-related data.

[0087] S104: Based on the planning stage related data, data integration is performed as a means to generate a hierarchical data structure.

[0088] Step S101 involves using a decision tree classification algorithm to initially classify BIM model data based on the building element type and functional requirements to generate initial classification data, which helps separate model elements into different categories according to their type. Step S102 involves performing functional decomposition on the data based on the initial classification data using a K-means clustering algorithm to generate functional classification data, which helps group elements with similar functions to better understand the functional layout in the model. Step S103 involves using an association rule learning algorithm to analyze the relationship between the data and planning stages based on the functional classification data to generate planning stage-related data, which helps determine the evolution and correlation of different functional elements in various planning stages. In step S104, the initial classification data, functional classification data, and planning stage-related data are integrated using data merging to generate a complete hierarchical data structure. This complete hierarchical data structure contains detailed information about the building element type, functional classification, and planning stage, providing a basis for subsequent decomposition and analysis.

[0089] As shown in FIG. 3, the step of performing differentiation processing on the primary and secondary regions based on the hierarchical data structure using a data optimization algorithm based on heuristic rules, and optimizing the decomposition procedure to generate an optimized decomposition plan, includes the following substeps:

[0090] S201: Based on the hierarchical data structure, a heuristic rule-based analysis algorithm is used to analyze data in key areas to generate a key area processing plan.

[0091] S202: Analyze the dependent regions using a simplified processing algorithm based on the main region processing plan to generate a dependent region processing plan.

[0092] S203: Based on the main area processing plan and the dependent area processing plan, a data fusion technique and a balancing processing strategy are employed to generate a comprehensive data processing plan.

[0093] S204: Based on the comprehensive data processing plan, a multi-objective optimization technique is employed to perform differentiated data processing to generate an optimized decomposition plan.

[0094] Step S201 analyzes data from a key domain based on a hierarchical data structure using a heuristic rule-based analysis algorithm to determine the characteristics and requirements of the key domain and generate a key domain processing plan to ensure that the data in that domain meets specific optimization requirements. Step S202 analyzes dependent domains using a simplified processing algorithm based on the key domain processing plan. Since dependent domains may require different processing methods, a dependent domain processing plan is generated to ensure that the data in the dependent domains are also appropriately optimized. Step S203 employs data fusion techniques and balancing strategies to generate a comprehensive data processing plan based on the key domain processing plan and the dependent domain processing plan. This processing plan takes into account the requirements of the key domain processing and dependent domains, thereby ensuring that the data in the entire model is appropriately optimized and processed during the decomposition process. Step S204 employs multi-objective optimization techniques to consider differentiated data processing and generate a final optimized decomposition plan. This optimized decomposition plan optimizes the comprehensive data processing plan in various aspects, thereby achieving multiple optimization goals, such as reducing data redundancy and improving data quality.

[0095] As shown in FIG. 4, the step of performing logical reasoning using a decomposition engine based on predicate logic based on the optimized decomposition plan, and then selecting decomposition paths to generate logical decomposition paths, includes the following substeps:

[0096] S301: Based on the optimized decomposition plan, a predicate logic algorithm is used to perform elementary logical reasoning to generate an elementary logical reasoning result.

[0097] S302: Based on the result of the preliminary logical reasoning, a logical optimization technique is employed to further optimize the logical structure of the decomposition path, thereby generating a logically optimized path result.

[0098] S303: Based on the logical optimization route result, a weighted sorting algorithm is used to rank the route options, generating a weighted ranked decomposition result.

[0099] S304: Based on the weighted ranking decomposition result, a decision path analysis method is adopted to select an appropriate decomposition path, and in addition, logical reasoning efficiency is optimized to generate a logical decomposition path.

[0100] Step S301, which uses a predicate logic algorithm to perform elementary logical reasoning based on the optimized decomposition plan, aims to extract relationships and conditions from the optimized decomposition plan and generate elementary logical reasoning results. For example, it determines which elements need to be split in what order. Step S302, which employs logical optimization techniques based on the results of the elementary logical reasoning, further optimizes the logical structure of the decomposition path. This logical structure includes logical relationships of rearrangement and organization, thereby ensuring the rationality and efficiency of the decomposition path and generating an optimized logical path result. Step S303, which uses a weighted sorting algorithm to rank path options based on the optimized logical path result, generates a weighted ranked decomposition result by considering various factors such as the complexity of the decomposition path and data relevance. Step S304, which employs a decision path analysis method to select an appropriate decomposition path based on the weighted ranked decomposition result and further optimizes the efficiency of the logical reasoning, ensures that the selected decomposition path is optimal and meets specific decomposition goals and requirements.

[0101] As shown in FIG. 5, the step of updating the changed parts in the model based on the logical decomposition path using an incremental change detection algorithm to generate updated partitioned data includes the following substeps:

[0102] S401: Based on the logical decomposition path, incremental change detection techniques are employed to identify and update the changed parts in the model to generate preliminary update data.

[0103] S402: Based on the preliminary update data, synchronized update data is generated by maintaining data consistency using a data synchronization protocol.

[0104] S403: Based on the synchronous update data, process the data differences by way of a difference query in the database to generate differentiated update data.

[0105] S404: Based on the differentiated update data, a data merging algorithm is used to align the data to generate updated divided data.

[0106] Step S401, employing incremental change detection techniques to identify and update changed parts of the model based on the logical decomposition path, aims to determine the necessary model changes guided by the decomposition path and generate preliminary update data. Step S402, employing a data synchronization protocol to maintain data consistency based on the preliminary update data, enables the updated data to be synchronized with the original model data so that the updated data is consistent with other parts of the model. Step S403, employing a database difference query to process data differences based on the synchronized update data, serves to determine the changes in the data in the model and also to generate differentiated update data that reflects the changes due to the decomposition path. Step S404, employing a data merging algorithm to align the data based on the differentiated update data, aligns the new update data with the original model data to generate final updated partition data.

[0107] As shown in FIG. 6, the step of streaming data to generate streaming divided data using a data compression and optimization algorithm based on the updated divided data includes the following substeps:

[0108] S501: Based on the updated divided data, a data compression / encoding technique is employed to reduce the amount of data and generate elementary compressed data.

[0109] S502: The data structure is optimized by data reorganization based on the preliminary compressed data, and optimized data is generated.

[0110] S503: The data format is reconstructed based on the optimized data as a means for data reconstruction, and reconstructed data is generated.

[0111] S504: Based on the reconstructed data, the data is streamed using a data streaming framework to generate streaming divided data.

[0112] Step S501 of employing data compression and encoding techniques to reduce the data volume based on the updated partitioned data helps reduce data transfer and storage costs by compressing the data to a smaller size using a compression algorithm to generate elementary compressed data. Step S502 of optimizing the data structure based on the elementary compressed data using data reorganization enables data reorganization, thereby improving data access efficiency and readability, and generating optimized data. Step S503 of reconstructing the data format based on the optimized data using data reconstruction restores the data to its original format to ensure data integrity and accuracy, and generates reconstructed data. Step S504 of streaming the data using a data streaming framework based on the reconstructed data enables data to be processed in a streaming manner, thereby ensuring data continuity and real-timeness during the process, and finally generating streaming partitioned data.

[0113] As shown in FIG. 7, the step of optimizing data quality by means of data integrity verification based on the streaming divided data to generate quality verified data includes the following substeps:

[0114] S601: Based on the streaming divided data, a preliminary quality verification is performed using data integrity verification to generate preliminary quality verification data.

[0115] S602: Based on the preliminary quality verification data, a data cleansing algorithm is used to remove inconsistent and erroneous data to generate cleansed data.

[0116] S603: Based on the cleansed data, an integrity enhancement technique is employed to optimize the accuracy of the data to generate integrity assessment data.

[0117] S604: Optimizing data quality using a comprehensive data quality assessment framework based on the integrity assessment data to generate quality verified data.

[0118] Step S601 involves performing a preliminary quality verification using data integrity verification on the streaming split data to generate preliminary quality verification data. This step aims to identify potential problems or inconsistencies in the data and generate preliminary quality verification data containing possible problems or errors. Step S602 involves using a data cleansing algorithm to remove inconsistent and erroneous data based on the preliminary quality verification data, which helps ensure data consistency and accuracy and generate cleansed data from which the detected problems have been eliminated. Step S603 involves employing integrity enhancement techniques to further optimize data accuracy based on the cleansed data, which allows missing data to be filled and erroneous data to be corrected, and generates integrity assessment data to ensure data completeness and accuracy. Step S604 involves further optimizing data quality by considering individual data quality factors collectively based on the integrity assessment data, which helps ensure that the data meets quality standards in various aspects and generates quality-verified data to verify the quality and reliability of the data.

[0119] As shown in FIG. 8, the step of identifying interdependencies between components by means of network topology analysis based on the quality verified data and generating an inter-component relationship analysis report includes the following substeps:

[0120] S701: Based on the quality verified data, a graph theory algorithm is used to analyze the connectivity between components to generate a preliminary component relationship diagram.

[0121] S702: Based on the preliminary inter-component relationship diagram, components and dependency paths are identified by network flow analysis, and a component dependency diagram is generated.

[0122] S703: Based on the component dependency diagram, a cluster analysis algorithm is used to distinguish groups of multiple components, and a component group analysis result is generated.

[0123] S704: Based on the component group analysis result, further analyze the dependency and interaction relationships between the components by considering multiple network parameters as a means of network topology optimization, and generate an inter-component relationship analysis report.

[0124] Step S701, analyzing the connectivity between components using a graph theory algorithm based on the quality-verified data, aims to identify potential connections between components by generating an elementary component relationship diagram including basic connections between components. Step S702, identifying components and dependency paths using network flow analysis based on the elementary component relationship diagram, helps to determine dependencies between components and generate a component dependency diagram including the dependency paths. Step S703, distinguishing multiple component groups using a cluster analysis algorithm based on the component dependency diagram, helps to classify the components into various categories or groups and generate component group analysis results, helping to further understand the relationships between the components. Step S704, further analyzing the dependency and interaction relationships between components by considering multiple network parameters using network topology optimization based on the component group analysis results, helps to deeply understand the relationships between various components and generate an inter-component relationship analysis report, helping to provide detailed information about the dependencies and interactions.

[0125] As shown in FIG. 9, the step of integrating the results of the hierarchical data structure, the optimized disassembly plan, the logical disassembly path, the updated division data, the streaming division data, and the quality verified data using a data fusion and integration algorithm based on the inter-component relationship analysis report, completing the disassembly of the BIM model, and generating a disassembled BIM model includes the following substeps:

[0126] S801: Based on the inter-component correlation analysis report, the multi-stage data results are preliminarily matched by means of related data fusion to generate preliminarily matched data.

[0127] S802: Based on the preliminary matched data, a dimensional analysis method is adopted to match multifaceted data to create a comprehensive multidimensional data view, and a multidimensional fusion data set is generated.

[0128] S803: Based on the multidimensional fusion dataset, a data processing technique is employed to optimize and simplify the data to generate a subdivided dataset.

[0129] S804: Based on the disaggregated data set, a comprehensive data management strategy is adopted to perform final fusion and reconciliation of data, and in addition, proofreading of information consistency and completeness is performed to generate a disassembled BIM model.

[0130] Step S801, which involves initially integrating multi-stage data results using related data fusion based on the inter-component relationship analysis report, aims to integrate multi-stage data into a single integrated set of data containing information from multiple data sources. Step S802, which involves employing dimensional analysis techniques to integrate multifaceted data based on the initial integrated data to create a comprehensive multidimensional data view, integrates multidimensional and multi-angle information into a single multidimensional fusion dataset, helping to provide a more comprehensive dataset. Step S803, which involves employing data processing techniques to optimize and simplify the multidimensional fusion dataset, may include operations such as data cleansing, deduplication, and filtering to generate a segmented dataset, ensuring data quality and accuracy. Step S804, which involves employing comprehensive data management strategies to perform final data integration and integration based on the segmented dataset, enables the integration of various data and the verification of information consistency and completeness, resulting in a disassembled BIM model containing data results from all stages.

[0131] As shown in Figure 10, the BIM model rapid dismantling system is used to implement the above-mentioned BIM model rapid dismantling method, and includes an elementary classification module, a function division module, a planning stage related module, a data hierarchy construction module, a main domain processing module, a subordinate domain processing module, a comprehensive data processing module, a multi-objective optimization module, a logical dismantling path module, and a data fusion and integration module.

[0132] The preliminary classification module uses a decision tree classification algorithm to preliminary classify BIM model data based on the type and functional requirements of the building elements, thereby generating preliminary classified data.

[0133] The function division module performs function division on the data using a K-means clustering algorithm based on the preliminary classified data to generate function classified data.

[0134] The planning stage association module analyzes the relationship between the data and the planning stage based on the function classification data using an association rule learning algorithm to generate planning stage association data.

[0135] The data hierarchy construction module performs data integration by means of data merging based on the planning stage related data to generate a hierarchical data structure.

[0136] The main area processing module analyzes the main area data based on the hierarchical data structure using a heuristic rule-based analysis algorithm to generate a main area processing plan.

[0137] The dependent region processing module analyzes the dependent regions using a simplified processing algorithm based on the main region processing scheme to generate a dependent region processing scheme.

[0138] The comprehensive data processing module generates a comprehensive data processing plan by adopting data fusion techniques and balancing strategies based on the main area processing plan and the dependent area processing plan.

[0139] The multi-objective optimization module employs a multi-objective optimization technique to perform differentiated data processing based on the comprehensive data processing plan, thereby generating an optimized decomposition plan.

[0140] The logical decomposition path module generates a logical decomposition path by performing logical inference using a decomposition engine based on predicate logic based on the optimized decomposition plan.

[0141] The data fusion and alignment module uses a data fusion and alignment algorithm based on the logical decomposition path to align the results of the primary classification module, the function division module, the planning stage related module, the data hierarchy construction module, the main area processing module, the dependent area processing module, the comprehensive data processing module, the multi-objective optimization module, and the logical decomposition path module, to complete the decomposition of the BIM model and generate a disassembled BIM model.

[0142] The system uses algorithms such as quality verification and data purification to improve data quality, reducing errors and problems during planning and subsequent repair costs. It can process data from multiple sources and in various formats, providing users with a comprehensive data view and enabling better understanding and analysis of BIM model data. The multi-objective optimization module and logical decomposition path module provide intelligent decision-making support, helping users generate optimal decomposition plans and paths, improving the effectiveness of construction planning and management. The data fusion and integration module ensures consistency and completeness among the data results from each module, avoiding data fragmentation and inconsistency problems and improving the reliability of BIM model data. In summary, this BIM model rapid demolition system provides comprehensive support for building and construction planning, improving work efficiency, data quality, and decision-making levels, thereby providing a solid foundation for successful planning implementation and management, and is of great significance in improving productivity and quality in the building and construction fields.

[0143] The above content is merely a better embodiment of the present invention and does not limit other embodiments of the present invention. For those skilled in the art, equivalent embodiments obtained by changing or modifying the technical content set forth above may be applicable to other fields. However, any simple modifications and equivalent changes or modifications made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. Based on the type and function requirements of building elements, use the integrated decision tree classification algorithm and K-means clustering algorithm to classify the BIM model, and further classify the type, function and planning stage to generate a hierarchical data structure; According to the hierarchical data structure, a data optimization algorithm based on heuristic rules is used to perform differentiation processing on the primary and secondary regions, and then the decomposition procedure is optimized to generate an optimized decomposition plan; performing logical inference using a predicate logic-based decomposition engine to generate a logical decomposition path based on the optimized decomposition plan; updating the changes in the model based on the logical decomposition path using an incremental change detection algorithm to generate updated partitioned data; a step of streaming the data using a data compression and optimization algorithm based on the updated divided data to generate streaming divided data; Optimizing data quality by means of data integrity verification based on the streaming divided data to generate quality verified data; Based on the quality verified data, identifying interdependencies between components by means of network topology analysis, and generating an inter-component relationship analysis report; and a step of integrating the results of the hierarchical data structure, the optimized decomposition plan, the logical decomposition path, the updated division data, the streaming division data, and the quality verified data using a data fusion and integration algorithm based on the inter-component relationship analysis report, thereby completing the decomposition of the BIM model and generating a dismantled BIM model; The hierarchical data structure includes a data hierarchy of multiple functions such as structure, electricity, and piping, the logical decomposition path specifically refers to a decomposition method and order path derived based on a logical formula, the updated division data specifically refers to information on changing and updating the changed part of the model, the streaming division data specifically refers to an optimized dataset, the quality verified data specifically refers to a verified dataset, and the inter-component relationship analysis report specifically refers to a report showing the dependency relationships and network connection strength between the components. A method for rapid dismantling of BIM models.

2. The step of classifying the BIM model based on the type and function requirements of the building elements using the integrated decision tree classification algorithm and K-means clustering algorithm, and further classifying the type, function and planning stage to generate a hierarchical data structure; a sub-step of using a decision tree classification algorithm to initially classify the BIM model data based on the typology and functional requirements of the building elements to generate initial classified data; a sub-step of performing feature decomposition on the data using a K-means clustering algorithm based on the preliminary classified data to generate feature classified data; a sub-step of analyzing the relationship between the data and the planning stage based on the function classification data using an association rule learning algorithm to generate planning stage related data; and performing data integration by means of data merging based on the planning stage related data to generate a hierarchical data structure. The method for quickly dismantling a BIM model according to claim 1.

3. The step of performing differentiation processing for the primary and secondary regions based on the hierarchical data structure using a data optimization algorithm based on heuristic rules, and optimizing the decomposition procedure to generate an optimized decomposition plan, a sub-step of analyzing the data of the main area based on the hierarchical data structure using a heuristic rule-based analysis algorithm to generate a main area processing plan; a sub-step of analyzing the dependent regions using a simplified processing algorithm based on the main region processing plan to generate a dependent region processing plan; a sub-step of employing a data fusion technique and a balancing processing strategy to generate a comprehensive data processing plan based on the main area processing plan and the subordinate area processing plan; and performing a differentiated data processing based on the comprehensive data processing plan by adopting a multi-objective optimization technique to generate an optimized decomposition plan. The method for quickly dismantling a BIM model according to claim 2.

4. the step of performing logical inference using a decomposition engine based on predicate logic based on the optimized decomposition plan, and further selecting decomposition paths to generate logical decomposition paths; a sub-step of performing elementary logical reasoning using a predicate logic algorithm based on the optimized decomposition plan to generate an elementary logical reasoning result; a sub-step of employing a logic optimization technique based on the result of the preliminary logic reasoning to further optimize the logic structure of the decomposition path to generate a logic optimization path result; a sub-step of ranking the route options using a weighted sorting algorithm based on the logically optimized route results to generate a weighted ranked decomposition result; a sub-step of selecting an appropriate decomposition path by employing a decision path analysis method based on the weighted ranking decomposition result, and further optimizing logical reasoning efficiency to generate a logical decomposition path; The method for quickly dismantling a BIM model according to claim 3.

5. the step of updating the changed parts in the model based on the logical decomposition path using an incremental change detection algorithm to generate updated partitioned data; a sub-step of employing incremental change detection techniques to identify and update changes in the model based on the logical decomposition path to generate preliminary update data; a sub-step of generating synchronized update data based on the preliminary update data using a data synchronization protocol while maintaining data consistency; a sub-step of processing data differences by way of a difference query in a database based on the synchronized update data to generate differentiated update data; and a sub-step of aligning data using a data merging algorithm based on the differentiated update data to generate updated partitioned data. The method for quickly dismantling a BIM model according to claim 4.

6. the step of streaming data using a data compression and optimization algorithm based on the updated divided data to generate streaming divided data, a sub-step of employing a data compression and encoding technique based on the updated divided data to reduce the amount of data and generate elementary compressed data; a sub-step of optimizing a data structure by means of data reorganization based on the initially compressed data to generate optimized data; a sub-step of reconstructing a data format based on the optimized data by data reconstruction to generate reconstructed data; and streaming data based on the reconstructed data using a data streaming framework to generate streaming partitioned data. The method for quickly dismantling a BIM model according to claim 5.

7. The step of optimizing data quality by means of data integrity verification based on the streaming divided data to generate quality verified data, a sub-step of performing a preliminary quality verification based on the streaming divided data by means of data integrity verification to generate preliminary quality verification data; a sub-step of using a data cleansing algorithm based on the preliminary quality verification data to remove inconsistent and erroneous data to generate cleansed data; a sub-step of employing an integrity enhancement technique based on the cleansed data to optimize data accuracy and generate integrity assessment data; and optimizing data quality using a comprehensive data quality assessment framework based on the integrity assessment data to generate quality verified data. The method for quickly dismantling a BIM model according to claim 6.

8. the step of identifying interdependencies between components by means of network topology analysis based on the quality verified data and generating an inter-component relationship analysis report; a sub-step of analyzing the connectivity between components based on the quality verified data using a graph theory algorithm to generate a preliminary relationship diagram between the components; a sub-step of identifying components and dependency paths based on the preliminary component relationship diagram by means of network flow analysis, and generating a component dependency diagram; a sub-step of using a cluster analysis algorithm to distinguish multiple component groups based on the component dependency graph and generate a component group analysis result; and further analyzing the dependency and interaction relationships between the components by considering multiple network parameters as a means of network topology optimization based on the component group analysis result, thereby generating an inter-component relationship analysis report. The method for rapid dismantling of a BIM model according to claim 7.

9. The step of integrating the results of the hierarchical data structure, the optimized decomposition plan, the logical decomposition path, the updated division data, the streaming division data, and the quality verified data using a data fusion and integration algorithm based on the inter-component relationship analysis report, and completing the decomposition of the BIM model to generate a disassembled BIM model; a sub-step of initially integrating the multi-stage data results by means of related data fusion based on the inter-component relationship analysis report to generate initially integrated data; a sub-step of adopting a dimensional analysis method based on the preliminary aligned data to integrate multifaceted data to create a comprehensive multidimensional data view and generate a multidimensional fusion dataset; a sub-step of employing a data processing technique to optimize and simplify the data based on the multidimensional fusion dataset to generate a refined dataset; and employing a comprehensive data management strategy based on the disaggregated data set to perform final fusion and reconciliation of the data, as well as performing calibration for consistency and completeness of the information to generate a disassembled BIM model. The method for rapid dismantling of a BIM model according to claim 7.

10. Used to carry out the method for rapid dismantling of a BIM model according to any one of claims 1 to 9; The system includes an elementary classification module, a function decomposition module, a planning stage related module, a data hierarchy construction module, a main domain processing module, a dependent domain processing module, a comprehensive data processing module, a multi-objective optimization module, a logical decomposition path module, and a data fusion and integration module; The preliminary classification module uses a decision tree classification algorithm to preliminary classify the BIM model data based on the type and function requirements of the building elements to generate preliminary classification data; The function division module performs function division on the data using a K-means clustering algorithm based on the primary classification data to generate function classification data; The planning stage association module analyzes the relationship between the data and the planning stage based on the function classification data using an association rule learning algorithm to generate planning stage association data; The data hierarchy construction module performs data integration by means of data merging based on the planning stage related data to generate a hierarchical data structure; The main area processing module analyzes the main area data based on the hierarchical data structure using a heuristic rule-based analysis algorithm to generate a main area processing plan; The dependent region processing module analyzes the dependent region using a simplified processing algorithm based on the main region processing plan to generate a dependent region processing plan; The comprehensive data processing module generates a comprehensive data processing plan by adopting a data fusion technique and a balancing processing strategy based on the main area processing plan and the dependent area processing plan; The multi-objective optimization module employs a multi-objective optimization technique to perform differentiated data processing based on the comprehensive data processing plan, thereby generating an optimized decomposition plan; The logical decomposition path module performs logical reasoning using a decomposition engine based on predicate logic to generate a logical decomposition path based on the optimized decomposition plan; the data fusion and alignment module uses a data fusion and alignment algorithm based on the logical decomposition path to align the results of the primary classification module, the function division module, the planning stage related module, the data hierarchy construction module, the main area processing module, the dependent area processing module, the comprehensive data processing module, the multi-objective optimization module and the logical decomposition path module, to complete the decomposition of the BIM model and generate a disassembled BIM model; A rapid demolition system for BIM models.

Images