A construction method, system, device and storage medium fusing BIM and AI

Abstract

The application discloses a kind of fusion BIM and AI's construction method, system, equipment and storage medium, it is related to intelligent construction and building digitization technical field, including importing building data into BIM modeling software, obtain BIM building data;Call AI design model, based on BIM building data and the design requirement of user, generate building layout scheme and space arrangement result, and automatically update to BIM model;AI drawing optimization is carried out to the BIM model after updating, load building specification library and user-defined rule, identify cross-professional conflict point in combination with collision detection algorithm;In construction stage, call the BIM model after drawing optimization, carry out risk prediction and decision deduction to real-time construction and operation data by AI.The method described in the application realizes automatic design and model dynamic updating, can improve compliance and cross-professional consistency in design stage, can realize dynamic monitoring and intelligent decision support in whole construction process.

Description

Technical Field

[0001] This invention relates to the field of intelligent construction and digital building technology, specifically to a construction method, system, equipment, and storage medium that integrates BIM and AI. Background Technology

[0002] With the rapid development of digitalization and intelligentization in the construction industry, BIM technology has gradually become the core data carrier in the engineering design, construction, and operation and maintenance phases. By digitally representing the geometric features, material properties, and construction progress of building components, it achieves data connectivity throughout the entire building lifecycle. Artificial intelligence (AI) technology is also widely used in areas such as architectural design optimization, engineering simulation analysis, and intelligent decision support, enabling building projects to have intelligent assistance and dynamic response capabilities in the stages of visual design, construction management, and operation and maintenance. The combination of BIM and AI has become an important direction for the development of building information technology, enabling data-driven design optimization and intelligent prediction in complex engineering projects.

[0003] However, existing BIM and AI integration methods still have certain limitations. Traditional BIM modeling relies on manual input, resulting in lagging model updates and a lack of deep collaboration with AI algorithms, making it difficult to meet the needs of cross-disciplinary design review, construction phase risk prediction, and dynamic optimization throughout the entire lifecycle. The lack of a unified detection mechanism for model data across different disciplines leads to the failure to identify and correct design-phase conflicts in a timely manner. During construction and operation phases, BIM models are disconnected from real-time monitoring data, hindering intelligent risk warning and decision-making simulations. Therefore, there is an urgent need for a construction method that integrates BIM and AI to achieve data linkage and intelligent analysis throughout the entire process from design and drawing review to construction and operation, thereby improving the precision and safety management of engineering construction. Summary of the Invention

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

[0005] Therefore, the technical problem solved by this invention is that existing BIM modeling and AI design optimization methods have defects such as lagging model updates, insufficient cross-disciplinary data interaction, untimely identification of design conflicts, and data disconnect between construction and operation and maintenance stages, as well as the problem of how to achieve collaborative linkage between BIM data and AI intelligent analysis throughout the entire process of architectural design, drawing review and optimization, and construction and operation and maintenance.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a construction method integrating BIM and AI, comprising importing building data into BIM modeling software to obtain BIM building data.

[0007] By calling upon the AI ​​design model, based on BIM building data and user design requirements, the system generates building layout schemes and spatial arrangement results, and automatically updates them to the BIM model.

[0008] The updated BIM model is optimized using AI-based drawing review, and building code libraries and user-defined rules are loaded. Collision detection algorithms are used to identify cross-disciplinary conflict points.

[0009] During the construction phase, the optimized BIM model is used, and AI is employed to perform risk prediction and decision simulation based on real-time construction and operation data.

[0010] The process of identifying cross-disciplinary conflict points by combining collision detection algorithms includes spatial matching and comparison of architectural, structural and mechanical and electrical professional data based on the updated BIM model, establishing a three-dimensional detection area, calling collision detection algorithms to identify the intersection or overlap between components of different disciplines, determining the conflicting components and their corresponding locations, and classifying and recording the detection results.

[0011] Using AI to perform risk prediction and decision simulation on real-time construction and operation and maintenance data includes linking construction phase data with BIM model component information to establish a construction and operation and maintenance dataset; the AI ​​model trains a risk identification model based on historical parameters and real-time monitoring data, performs risk prediction, executes multi-scenario decision simulation, outputs risk level and response plan, and updates it to the BIM model simultaneously.

[0012] As a preferred embodiment of the construction method integrating BIM and AI described in this invention, the process of obtaining BIM building data includes: importing the geometric information, material properties, and specification parameters of the building into BIM modeling software and generating a three-dimensional model; performing data parsing and correlation mapping on the building data through the BIM model to obtain BIM building data; and transmitting the BIM data to a central platform through an API interface for data cleaning and standardization.

[0013] As a preferred embodiment of the construction method integrating BIM and AI described in this invention, the central platform includes processing steps for storing, parsing, and retrieving BIM building data. The central platform performs the reception, classification, and standardization of BIM building data, and retrieves and compares BIM building data during AI design and AI drawing review optimization processes.

[0014] As a preferred embodiment of the construction method integrating BIM and AI described in this invention, the generation of building layout scheme and spatial arrangement results includes: based on the building land area, plot ratio and functional zoning information in the BIM building data, calling the AI ​​design model for automatic arrangement and generating a layout scheme; after the layout scheme is generated, optimizing the flow and dividing the internal space of the building according to the personnel flow path, equipment layout and entrance and exit locations, forming the building layout scheme and spatial arrangement results, and automatically synchronizing them to the central platform to update the BIM model data.

[0015] As a preferred embodiment of the construction method integrating BIM and AI described in this invention, the loading of the building code library and user-defined rules includes: establishing a unified code data indexing system for parameterized building code data; calling user-defined rule sets; and setting personalized review parameters for specific building types, project functions, and construction conditions.

[0016] As a preferred embodiment of the construction method integrating BIM and AI described in this invention, the method of identifying cross-disciplinary conflict points by combining collision detection algorithms includes: performing spatial matching and geometric comparison of architectural, structural and MEP data in the updated BIM model; establishing a three-dimensional detection area based on the coordinates, dimensions and layout information of components in the BIM model; calling the collision detection algorithm to calculate the geometric intersection and spatial overlap between components of different disciplines; identifying conflicting components and their corresponding positions; and classifying and recording the detected conflict information after the identification is completed.

[0017] As a preferred embodiment of the construction method integrating BIM and AI described in this invention, the step of using AI to perform risk prediction and decision-making simulation of real-time construction and operation data includes: associating and matching the progress, quality, safety, and equipment operation data collected during the construction phase with the component information in the BIM model to establish a time-series construction and operation dataset; training a risk identification model based on historical construction parameters and real-time monitoring data to make probabilistic predictions of construction deviations, equipment failures, and structural anomalies; after the prediction results are generated, calling the inference engine to perform multi-scenario decision-making simulations, outputting risk levels and response plans, and synchronizing the prediction and simulation results to the BIM model to update the project's full lifecycle data.

[0018] Another objective of this invention is to provide a construction system that integrates BIM and AI. Through the collaborative work of the data import module, scheme generation module, drawing review and optimization module, and risk prediction module, it solves the problems of lagging model updates, untimely identification of cross-disciplinary drawing review conflicts, and insufficient accuracy of risk prediction during the construction phase in current BIM and AI integration technologies.

[0019] As a preferred embodiment of the construction system integrating BIM and AI described in this invention, it includes: a data import module, a scheme generation module, a drawing review and optimization module, and a risk prediction module; the data import module is used to import building data into BIM modeling software, perform structured analysis and transformation of geometric information, material properties, and specification parameters, and generate BIM building data; the scheme generation module is used to call the AI ​​design model, and based on the input BIM building data and user design requirements, perform automatic generation of building layout and spatial arrangement optimization to form a building layout scheme, and automatically update the generated results to the BIM model; the drawing review and optimization module is used to load the building specification library and user-defined rules on the updated BIM model, perform AI drawing review and optimization, and automatically check and verify the logical consistency of the model according to the loaded specification parameters and user-defined review standards; the risk prediction module is used to call the BIM model optimized by drawing review during the construction phase, combine AI to process real-time construction and operation and maintenance data, perform risk prediction and decision inference, and generate prediction results of construction deviations, structural risks, and equipment anomalies, providing support for engineering decision-making.

[0020] Another object of the present invention is to provide a construction device that integrates BIM and AI, including a memory and a processor, the memory storing a computer program, the processor executing the computer program as a step to realize a construction method that integrates BIM and AI.

[0021] Another object of the present invention is to provide a construction storage medium that integrates BIM and AI, wherein a computer program is stored thereon, and when the computer program is executed by a processor, the steps of the construction method integrating BIM and AI are implemented.

[0022] The beneficial effects of this invention are as follows: The construction method integrating BIM and AI provided by this invention enables centralized management and digital modeling of building geometry, material properties, and specification parameters by importing building data into BIM modeling software; by calling AI design models, building layout schemes and spatial arrangement results are generated based on BIM building data and user design requirements, enabling automated design and dynamic model updates; by performing AI review and optimization on the updated BIM model and loading building specification libraries and user-defined rules, compliance and cross-disciplinary consistency in the design phase can be improved; by calling the BIM model optimized by review and combining it with AI to perform risk prediction and decision simulation on real-time construction and operation and maintenance data during the construction phase, dynamic monitoring and intelligent decision support throughout the construction process can be achieved. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is an overall flowchart of a construction method integrating BIM and AI provided in Embodiment 1 of the present invention.

[0025] Figure 2 This is a general framework diagram of an AI intelligent agent for a construction system that integrates BIM and AI, provided in Embodiment 2 of the present invention.

[0026] Figure 3 This is an AI-powered full-process planning and control diagram for a construction system that integrates BIM and AI, provided in Embodiment 2 of the present invention.

[0027] Figure 4 This is a diagram illustrating an AI-assisted drawing review system for a construction system integrating BIM and AI, as provided in Embodiment 2 of the present invention. Detailed Implementation

[0028] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0029] Example 1, referring to Figure 1 As one embodiment of the present invention, a construction method integrating BIM and AI is provided, comprising:

[0030] S1: Import the building data into the BIM modeling software to obtain BIM building data.

[0031] Furthermore, obtaining BIM building data includes importing the building's geometric information, material properties, and specification parameters into BIM modeling software to generate a 3D model. The building data is then parsed and mapped using the BIM model 200 to obtain BIM building data. The BIM data is then transmitted to the central platform via an API interface for data cleaning and standardization.

[0032] It should be noted that the geometric information, component attributes, material characteristics, and specification parameters of the building project are imported into the BIM modeling software. The geometric information includes the building's shape, floor height, component dimensions, and spatial coordinates; the material characteristics include strength, density, and fire resistance rating; and the specification parameters include structural safety factors, energy-saving standards, and fire safety distances. Using the parametric modeling function of the BIM modeling software, the imported building information is logically modeled and geometrically reconstructed to generate a complete 3D BIM building model.

[0033] Furthermore, after generating the BIM model 200, the building data within the model is parsed and mapped. By establishing spatial topological relationships and logical dependencies between components, a multi-dimensional information network of building components is constructed, achieving data unification of building components at the geometric, attribute, and specification levels.

[0034] Through the data interface of BIM model 200, the component attribute table, spatial location information and structural connection relationship in the model are output as standardized data format; through the API interface, the BIM building data is transmitted to the central platform. After receiving the data, the central platform performs data cleaning and standardization processing, including removing duplicate data, supplementing missing parameters, unifying data units and naming conventions, etc., to form a structured and unified BIM building dataset.

[0035] It should also be noted that parametric modeling is implemented during the import phase of geometric information, component attributes, material characteristics, and specification parameters to ensure the integrity and consistency of building component information. During the model generation and data parsing phases, spatial topology and semantic mapping mechanisms enable automatic association of multi-dimensional information between components. In the data transmission and cleaning phases, API interfaces and standardized processing rules achieve cross-platform data consistency and structured storage, thereby constructing a unified BIM building dataset. This method effectively improves the accuracy, reusability, and lifecycle data collaboration efficiency of building information.

[0036] Furthermore, after data cleaning, the central platform classifies and hierarchically manages the BIM building data, storing building component data, spatial information data, material performance data, and specification parameter data in different data domains to support rapid retrieval and dynamic comparison in subsequent AI design model 100 calls, AI drawing review optimization, and AI risk prediction processes, thereby providing a unified data support environment for the digital and intelligent construction of the entire building lifecycle.

[0037] Furthermore, the central platform includes processing steps for storing, parsing, and retrieving BIM building data. It performs the reception, classification, and standardization of BIM building data through the central platform, and calls upon BIM building data for updating and comparison during the AI ​​design and AI drawing review optimization process.

[0038] It should be noted that after the central platform completes the BIM building data cleaning, the standardized building information is analyzed and classified for storage in layers. The central platform identifies and distinguishes the building component data, spatial location information data, material performance data, and specification parameter data item by item according to the preset building information classification rules. The data classification module is called to allocate the above data to the corresponding data fields according to the data attribute tags, and component information field, spatial information field, material performance field, and specification parameter field are established respectively.

[0039] In detail, during the data entry process, the central platform adopts a distributed database structure for each data domain and establishes cross-domain association tables based on index fields to enable rapid matching of multi-source data during the AI ​​design model 100 call phase. At the same time, the source, update time and associated component number of various types of data are recorded through metadata description files to ensure efficient retrieval and dynamic comparison of heterogeneous data from different BIM modeling sources under a unified index.

[0040] Furthermore, in supporting AI design modeling and AI drawing review optimization, the central platform calls target fields of various data domains through model interfaces to provide AI algorithms with multi-dimensional support data on structural safety, energy efficiency standards, and component performance. In the AI ​​risk prediction stage, the system quickly indexes the historical parameters and spatial relationships of target components through feature vectorization, realizing dynamic risk identification and automatic comparison of building models, thereby forming a digital and intelligent support system that runs through the entire life cycle of buildings.

[0041] It should also be noted that the central platform has completed the hierarchical and structured management of BIM building data, providing a unified data support environment for subsequent AI-driven building optimization design, energy consumption simulation and risk prediction.

[0042] S2: Call AI design model 100, generate building layout scheme and spatial arrangement results based on BIM building data and user design requirements, and automatically update to BIM model 200.

[0043] Furthermore, the generation of building layout schemes and spatial arrangement results includes: based on the building land area, plot ratio and functional zoning information in the BIM building data, calling the AI ​​design model 100 for automatic arrangement and generating a layout scheme; after the layout scheme is generated, optimizing the flow of people, equipment layout and entrance and exit locations, and dividing the internal space of the building into circulation and areas, forming the building layout scheme and spatial arrangement results, and automatically synchronizing them to the central platform to update the data of the BIM model 200.

[0044] It should be noted that, based on the cleaned BIM building database in the central platform, the land area, plot ratio, and functional zoning information of the target building are read and used as the input data source for AI design model 100. During the model training phase, AI design model 100 performs deep learning through large sample building layout data to master the reasonable division rules of building functional areas and the logic of circulation layout. Thus, during the automatic layout phase, it can generate an initial layout scheme that conforms to the specifications based on the building type, plot conditions, and functional requirements.

[0045] Furthermore, after the layout plan is generated, the AI ​​computing module in the central platform is invoked to perform flow analysis on information such as personnel flow paths, equipment layout, and entrance and exit locations.

[0046] Specifically, a path planning algorithm based on the internal traffic rules of a building is used to simulate pedestrian routes and identify flow conflicts and congestion points.

[0047] The path planning algorithm is expressed as follows:

[0048] ,

[0049] in, The total cost of the entire path, The Euclidean distance between two adjacent points. The obstacle or congestion coefficient between nodes These are weighting coefficients used to balance the effects of path length and congestion factors. This represents the total number of nodes in the path.

[0050] A preferred scheme for the weighting coefficients is as follows: In public buildings (such as exhibition halls and shopping malls), where spaces are open and pedestrian density is high, it is advisable to... To enhance congestion avoidance, in industrial plants or equipment areas where obstacles are fixed and paths are relatively regular, it is advisable to... To ensure a compact path, in office buildings or residential settings, a regular layout and ample passageways are preferable. This will improve traffic efficiency.

[0051] Based on the equipment layout requirements and the accessibility of maintenance channels, the functional areas of the building's interior space are divided and optimized to form a spatial layout with a clear structure and reasonable circulation. After the building layout and spatial arrangement optimization are completed, the final results are synchronized back to the central platform, and the BIM data update interface is called to replace the component information, spatial coordinates and attribute parameters in the model, so as to achieve data consistency between the layout scheme and the BIM model 200.

[0052] It should also be noted that by deeply integrating the AI ​​design model 100 with BIM building data, the building layout and spatial arrangement scheme is automatically generated based on the cleaned building land area, plot ratio and functional zoning information in the central platform. The path planning algorithm is called to dynamically balance the path length and spatial congestion within the optimal range of weight coefficients, thereby achieving adaptive optimization of the flow of people and equipment inside the building.

[0053] S3: Perform AI-based drawing review optimization on the updated BIM model 200, load the building code library and user-defined rules, and combine collision detection algorithms to identify cross-disciplinary conflict points.

[0054] Furthermore, loading the building code library and user-defined rules includes establishing a unified code data indexing system for parameterized building code data, calling user-defined rule sets, and setting personalized review parameters for specific building types, project functions, and construction conditions.

[0055] It should be noted that the building code entries and their parameterized data are imported into the central platform's code database. The building code entries are then subjected to structured parsing and parameter extraction to form a mapping relationship between code entries and corresponding parameters. Through semantic word segmentation and rule matching algorithms, the code data is classified and coded according to building type, building function, construction stage, and safety level to establish a unified code data index system, enabling the code data to be quickly retrieved and called by the AI ​​drawing review module.

[0056] Specifically, to meet the unique design requirements of different types of buildings, users can define personalized rule sets in the interface. Users can set review parameters, including floor height limits, evacuation distances, sunlight spacing, energy consumption indicators, and structural load ranges. These custom parameters are then compared with standard parameters in the specification database to form a dynamically updated rule set.

[0057] Furthermore, when performing code review, the AI-powered drawing review model automatically matches the corresponding rule set based on the building type, function, and construction conditions, and loads user-defined parameters. It performs consistency analysis and deviation detection on the component information, spatial layout, and construction constraints of the building model, thereby identifying structural items that do not comply with the code or exceed the design tolerance, thus achieving automation and personalization of building code review.

[0058] It should also be noted that by establishing a mapping relationship between standard items and parameters through structured parsing and semantic word segmentation algorithms, the AI ​​drawing review module can quickly call standard data, which significantly improves the review response speed and matching accuracy. It allows users to flexibly define review parameters for different building types and construction conditions, such as floor height limits, evacuation distances, illuminance ranges, structural loads, etc., and dynamically integrate these personalized parameters into the standard database to achieve collaborative review of standard parameters and custom rules.

[0059] Furthermore, the identification of cross-disciplinary conflict points by combining collision detection algorithms includes spatial matching and geometric comparison of the data of architecture, structure and MEP in the updated BIM model 200, establishing a three-dimensional detection area based on the coordinates, dimensions and layout information of the components in the BIM model 200, calling the collision detection algorithm to calculate the geometric intersection and spatial overlap between components of different disciplines, identifying conflicting components and their corresponding positions, and classifying and recording the detected conflict information after the identification is completed.

[0060] It should be noted that after loading the structural, architectural, and MEP (Mechanical, Electrical, and Plumbing) data into the updated BIM model 200, a unified processing method was applied based on the coordinate systems of each professional model. This process converted the component coordinates, dimensions, and installation elevations from different disciplines into corresponding spatial data point sets in the common spatial reference coordinate system. Specifically, the architectural data mainly includes the geometric boundaries of walls, doors, windows, beams, columns, and floor slabs; the structural data includes 3D solid models of reinforcing bars, beams, columns, and foundation components; and the MEP data includes the paths and cross-sectional information of pipeline components such as pipes, cable trays, cables, and ducts. An automatic detection mesh is generated within the 3D detection space, and a spatial index structure is established using spatial partitioning and octree hierarchical indexing technology, significantly improving collision detection efficiency.

[0061] Specifically, the collision detection algorithm is expressed as follows:

[0062] ,

[0063] in, For components With components The collision determination result is 1, indicating a collision occurred, and 0, indicating no collision. This represents the minimum spatial distance between points on the component surface. The volume overlap ratio between components. Distance threshold This is the volume overlap threshold.

[0064] It should be noted that a preferred scheme for the distance threshold is... , The average linear dimension of the components involved in the detection can make tiny gaps smaller than 5% of the component size insensitive, thus avoiding misjudgments caused by modeling errors or numerical noise; at the same time, when the distance between two components is less than this proportion, they can be considered to be close to the level of potential collision risk, which has good stability and applicability.

[0065] A preferred scheme for volume overlap threshold is , The average volume of the components ensures that a substantial collision is only determined when the overlapping volume exceeds 2% of the component volume, thus balancing detection accuracy and computational efficiency and reducing misjudgments of slight contact or boundary overlap.

[0066] The volume overlap ratio between components is expressed as:

[0067] ,

[0068] in, For components The three-dimensional volume, For components The three-dimensional volume.

[0069] When the collision detection algorithm is invoked, the minimum distance matrix is ​​calculated based on the spatial geometric relationship between components. When the minimum distance between two components is lower than the distance threshold or the volume overlap rate exceeds the preset ratio, it is determined to be a potential conflict.

[0070] After identification, the detected conflict information is output in the form of tables and 3D annotations, and the conflict location is highlighted in the BIM model 200. Users can click on the conflict point through the interactive interface to view the corresponding component information and spatial attributes. It supports classification, retrieval and filtering based on professional type, conflict category or severity level.

[0071] It should also be noted that by introducing distance thresholds and volume overlap thresholds, it is possible to effectively distinguish between minor contact and substantial collision, avoiding misjudgments caused by modeling errors or geometric noise. It can still operate stably even when the modeling accuracy of different disciplines is inconsistent. In the updated BIM model 200 structure, the coordinate systems of architecture, structure and MEP are unified and mapped to the common space benchmark, realizing geometric alignment and data fusion between heterogeneous models. This ensures that the collision detection algorithm can directly call the coordinates, dimensions and installation elevation information of components from different disciplines, improving the universality and scalability of the algorithm.

[0072] S4: During the construction phase, call the optimized BIM model 200 after review and use AI to perform risk prediction and decision simulation based on real-time construction and operation data.

[0073] Furthermore, using AI to perform risk prediction and decision-making simulations on real-time construction and operation and maintenance data includes associating and matching progress, quality, safety, and equipment operation data collected during the construction phase with component information in BIM model 200 to establish a time-series construction and operation and maintenance dataset. The AI ​​model trains a risk identification model based on historical construction parameters and real-time monitoring data to make probabilistic predictions of construction deviations, equipment failures, and structural anomalies. After the prediction results are generated, the inference engine is called to perform multi-scenario decision-making simulations, output risk levels and response plans, and synchronize the prediction and simulation results to BIM model 200 to update the project's full lifecycle data.

[0074] It should be noted that multi-source data is acquired during the construction phase, including progress data, quality inspection data, safety monitoring data, and equipment operating status data. After real-time acquisition via on-site sensors, ranging devices, and video monitoring devices, the data is automatically correlated with the component's unique identifier and the component information in the BIM model 200 to establish a data association at the component level.

[0075] Multi-source data collected during construction and operation are organized chronologically to form a time-series dataset. Based on historical construction parameters and real-time monitoring data, an AI model is trained to construct a risk identification model. This model uses construction deviations, equipment failure characteristics, structural deformation parameters, and environmental disturbances as input variables. During training, the model parameters are optimized through gradient updates to obtain an identification function that reflects the dynamic characteristics of risks. Real-time monitoring data is then input into the trained AI model to perform risk identification and probability prediction. When abnormal equipment operating parameters or excessive structural fluctuations are detected, the model outputs a risk level and corresponding risk label. Based on the output, the inference engine is invoked to perform multi-scenario decision-making simulations, generating simulation results for potential construction risks, equipment failures, and structural instability. The simulation content includes risk propagation paths, countermeasures, and resource scheduling schemes.

[0076] After the simulation is completed, the generated risk level, risk location, and simulation solution are synchronously recorded into the BIM model 200 dataset, and the risk location is highlighted in the model using 3D annotations. Users can view the time evolution of various risk events and corresponding response plans through an interactive interface.

[0077] It should also be noted that by processing construction progress, quality, safety, and equipment operation data in a time-series manner, the AI ​​model can capture risk evolution trends during continuous learning, realizing the transformation of risk from static identification to dynamic prediction. This effectively avoids risk omissions and delayed responses caused by data lag. The prediction results and deduction schemes are recorded synchronously in the BIM model 200, and risky components and risky areas are highlighted in the model in the form of 3D annotations. This allows risk information to be visualized in a spatial semantic way, making it easier for construction managers to intuitively analyze the risk location, impact range, and time evolution process through the model.

[0078] Example 2, refer to Figures 2-4 As an embodiment of the present invention, a construction system integrating BIM and AI is provided, including a data import module, a scheme generation module, a drawing review and optimization module, and a risk prediction module.

[0079] like Figure 2 As shown, the system can be mapped as a four-layer structure from bottom to top: Model Layer 303, Capability Layer 302, Presentation Layer 301, and Application Layer 300. Different layers interact dynamically through data interfaces to realize intelligent construction and risk management throughout the building's entire lifecycle. The architecture revolves around the overall framework of the AI ​​intelligent body and deeply integrates the four core components: Model Layer 303, Capability Layer 302, Application Layer 301, and Presentation Layer 300. It empowers typical application scenarios such as safety hazard identification, construction method library, technical system consultation, risk identification, and progress analysis.

[0080] AI full-process planning and management, such as Figure 3 As shown, the AI-driven full-process planning and control is guided by the core project objectives, formulating a detailed plan covering all stages and key tasks of the project. It clearly defines the responsible parties, timelines, and deliverable standards for each task, forming a three-dimensional execution loop of "responsibility-time-deliverable," comprising a logic layer (400), a model layer (303), and a data layer (401). An AI agent for progress control is introduced, outputting rich construction simulation analysis results, providing strong support for improving construction efficiency, optimizing resource allocation, and enabling real-time dynamic decision-making.

[0081] AI-assisted image review system, such as Figure 4 As shown, the architecture revolves around the overall system of AI-assisted image review, including review layer 500, calculation layer 501, recognition layer 502, and standardization layer 503.

[0082] The data import module is used to import building data into BIM modeling software, perform structured analysis and transformation of geometric information, material properties and specification parameters, and generate BIM building data.

[0083] The scheme generation module is used to call the AI ​​design model 100, and based on the input BIM building data and user design requirements, it performs automatic generation of building layout and optimization of spatial arrangement to form a building layout scheme, and automatically updates the generated results to the BIM model 200.

[0084] The drawing review optimization module is used to load building code library and user-defined rules on the updated BIM model 200, perform AI drawing review optimization, and automatically check and verify the logical consistency of the model based on the loaded code parameters and user-defined review standards.

[0085] The risk prediction module is used to call the BIM model 200 that has been reviewed and optimized during the construction phase. It combines AI to process real-time construction and operation and maintenance data, perform risk prediction and decision-making simulation, and generate prediction results for construction deviations, structural risks and equipment anomalies, providing support for engineering decision-making.

[0086] This embodiment also provides a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements a construction system integrating BIM and AI as proposed in the above embodiment.

[0087] This embodiment also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a construction system integrating BIM and AI as proposed in the above embodiments.

[0088] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0089] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0090] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0091] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0092] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A construction method integrating BIM and AI, characterized in that, include: Import building data into BIM modeling software to obtain BIM building data; By calling upon the AI ​​design model, based on BIM building data and user design requirements, the system generates building layout schemes and spatial arrangement results, and automatically updates them to the BIM model. AI-powered review and optimization of the updated BIM model; loading of building code library and user-defined rules; and identification of cross-disciplinary conflict points using collision detection algorithm. During the construction phase, the optimized BIM model after review is used, and AI is used to perform risk prediction and decision simulation based on real-time construction and operation and maintenance data. The process of identifying cross-disciplinary conflict points by combining collision detection algorithms includes spatial matching and comparison of architectural, structural and mechanical and electrical professional data based on the updated BIM model, establishing a three-dimensional detection area, calling collision detection algorithms to identify the intersection or overlap between components of different disciplines, determining the conflicting components and their corresponding positions, and classifying and recording the detection results. Using AI to perform risk prediction and decision simulation on real-time construction and operation and maintenance data includes linking construction phase data with BIM model component information to establish a construction and operation and maintenance dataset; the AI ​​model trains a risk identification model based on historical parameters and real-time monitoring data, performs risk prediction, executes multi-scenario decision simulation, outputs risk level and response plan, and updates it to the BIM model simultaneously.

2. The construction method integrating BIM and AI as described in claim 1, characterized in that: The process of obtaining BIM building data includes importing the building's geometric information, material properties, and specification parameters into BIM modeling software to generate a 3D model. The building data is then analyzed and mapped using the BIM model to obtain BIM building data. The BIM data is then transmitted to a central platform via an API interface for data cleaning and standardization.

3. The construction method integrating BIM and AI as described in claim 2, characterized in that: The central platform includes processing steps for storing, parsing, and retrieving BIM building data. It performs the reception, classification, and standardization of BIM building data through the central platform, and updates and compares BIM building data during AI design and AI drawing review optimization processes.

4. The construction method integrating BIM and AI as described in claim 1 or 3, characterized in that: The generation of building layout schemes and spatial arrangement results includes: based on the building land area, plot ratio and functional zoning information in the BIM building data, calling the AI ​​design model to automatically arrange and generate layout schemes; after the layout schemes are generated, the internal space of the building is optimized and divided into areas according to the flow path of people, equipment layout and entrance and exit locations, forming building layout schemes and spatial arrangement results, and automatically synchronizing to the central platform to update the BIM model data.

5. The construction method integrating BIM and AI as described in claim 4, characterized in that: The loading of the building code library and user-defined rules includes establishing a unified code data indexing system for parameterized building code data, calling user-defined rule sets, and setting personalized review parameters for specific building types, project functions, and construction conditions.

6. The construction method integrating BIM and AI as described in any one of claims 1, 2, 3, and 5, characterized in that: The method of identifying cross-disciplinary conflict points by combining collision detection algorithms includes: performing spatial matching and geometric comparison of the data of architecture, structure and MEP in the updated BIM model; establishing a three-dimensional detection area based on the coordinates, dimensions and layout information of the components in the BIM model; calling the collision detection algorithm to calculate the geometric intersection and spatial overlap between components of different disciplines; identifying conflicting components and their corresponding positions; and classifying and recording the detected conflict information after the identification is completed.

7. The construction method integrating BIM and AI as described in claim 6, characterized in that: The process of using AI to perform risk prediction and decision-making simulations on real-time construction and operation data includes: associating and matching progress, quality, safety, and equipment operation data collected during the construction phase with component information in the BIM model to establish a time-series construction and operation dataset; training a risk identification model based on historical construction parameters and real-time monitoring data to make probabilistic predictions of construction deviations, equipment failures, and structural anomalies; after the prediction results are generated, calling the inference engine to perform multi-scenario decision-making simulations, outputting risk levels and response plans, and synchronizing the prediction and simulation results to the BIM model to update the project's full lifecycle data.

8. A construction system integrating BIM and AI, employing the construction method integrating BIM and AI as described in any one of claims 1 to 7, characterized in that: It includes a data import module, a solution generation module, a drawing review and optimization module, and a risk prediction module; The data import module is used to import building data into BIM modeling software, perform structured analysis and transformation of geometric information, material properties and specification parameters, and generate BIM building data. The scheme generation module is used to call the AI ​​design model, and based on the input BIM building data and user design requirements, to automatically generate the building layout and optimize the spatial arrangement to form a building layout scheme, and to automatically update the generated results to the BIM model. The drawing review optimization module is used to load building code library and user-defined rules on the updated BIM model, perform AI drawing review optimization, and automatically check and verify the logical consistency of the model based on the loaded code parameters and user-defined review standards. The risk prediction module is used to call the BIM model optimized by the review drawings during the construction phase, and combine AI to process real-time construction and operation and maintenance data, perform risk prediction and decision inference, and generate prediction results of construction deviation, structural risk and equipment anomaly, so as to provide support for engineering decision-making.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the construction method integrating BIM and AI as 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 construction method integrating BIM and AI as described in any one of claims 1 to 7.

Images