A linkage method and system for engineering management BIM system and AI artificial intelligence

Abstract

The present invention discloses a linkage method and system of an engineering management BIM system and AI artificial intelligence, comprising: performing key data mining operations on model component data in the BIM system to generate a target model component data set; calling the target model component data set based on an AI big model, analyzing the target model component data set by the AI ​​big model to generate a linkage analysis result of the target model component data set formed by the BIM system, and visually presenting the linked data of the target model component data set formed after multi-task linkage execution processing; the present invention can mine, cluster, intelligently analyze, linkage schedule and feedback the model component related data in the BIM system after mobilizing the AI ​​model for mining, and simultaneously adds the model analysis of the AI ​​linkage evaluation and the priority scheduling output display for the model component sub-target model parts, thereby improving the linkage processing effect.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering management, and in particular to a method and system for linking an engineering management BIM system with AI artificial intelligence. Background Art

[0002] BIM system, or Building Information Modeling, is a method of building design, construction and management based on digital technology. At the same time, with the gradual application of the latest IT technologies represented by 5G, Internet of Things, and AI technology and high-end manufacturing technologies represented by robotics in the construction industry, the construction industry has accelerated the pace of digital transformation based on BIM technology.

[0003] At present, the BIM collaborative platform can complete intelligent applications in the design, construction and operation and maintenance process through BIM big data and AI engine, especially AI intelligent assisted construction, component design, component intelligent search, AI assisted building layout planning and AI assisted mechanical and electrical design. However, the current BIM collaborative platform is still lacking in the linkage processing and functional application of AI assisted construction site planning, especially the linkage based on AI artificial intelligence big model, lack of detailed training and linkage of model components in the BIM system, and the intelligent analysis of sub-target model components in the model components by the AI ​​expert group. The rationality of the early plan in artificial intelligence analysis and database-based feedback is still lacking, which is reflected in the problems of weak linkage and poor correlation. Therefore, the current BIM collaborative platform lacks the process or method of linkage evaluation based on AI artificial intelligence big model, and thus lacks in construction organization design and resource allocation optimization. To this end, we propose a linkage method and system of engineering management BIM system and AI artificial intelligence. Summary of the invention

[0004] In view of the above problems existing in the existing engineering management BIM system, the present invention is proposed.

[0005] Therefore, one of the objects of the present invention is to provide a method and system for linking an engineering management BIM system with AI artificial intelligence, which can link with the BIM system after mobilizing the AI ​​model for mining, and can mine, cluster, intelligently analyze, link and schedule and feedback the model component related data in the BIM system, and simultaneously add the model analysis of AI linkage evaluation and the priority scheduling output display for the sub-target model parts of the model component, thereby improving the linkage processing effect and further facilitating the construction organization design and resource allocation optimization.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] On the one hand, the present invention provides a method for linking an engineering management BIM system with AI artificial intelligence, comprising:

[0008] By performing key data mining operations on the model component data in the BIM system to output the key data features associated with the model component data in the BIM system, and forming associated behavior data and key data feature data corresponding to the model component data, a target model component data set is generated;

[0009] Based on the AI ​​big model, the target model component data set is called, and the matching model component data, associated behavior data and key data feature data in the target model component data set are aggregated to form the aggregate features of the model component data; the AI ​​big model analyzes the aggregate features of the model component data in the target model component data set, and the big model linkage expert group evaluates it, generates the linkage analysis results of the target model component data set formed by the BIM system, and divides the linkage tasks;

[0010] In response to the linkage task division and linkage analysis results, the target model component data set is subjected to multi-task linkage execution processing to form the linked data of the target model component data set, and linkage scheduling information matching the target model component data set is generated at the same time. Based on the matched linkage scheduling information, the linked data is matched and linked with the target model component data set, and a visual presentation is performed.

[0011] As a preferred solution of the present invention, the linkage task division is specifically based on the associated behavior data and key data feature data in the target model component data set, the model components are screened through the analysis of the AI ​​big model, and the task priority is determined and matched according to the influencing factors of the key data features to generate linkage analysis results of the corresponding level.

[0012] As a preferred solution of the present invention, the influencing characteristics of the key data feature data include: geometric information characteristics, non-spatial information characteristics, topological relationship characteristics, model complexity characteristics, spatial area characteristics of data, model granularity characteristics, texture characteristics, synergy characteristics, cost attribute characteristics and state characteristics of information output mode.

[0013] As a preferred solution of the present invention, the matching model component data, associated behavior data and key data feature data in the target model component data set are subjected to data aggregation operations to form model component data aggregation features. The aggregation operation specifically adopts a K-means clustering algorithm to perform unsupervised learning and clustering on the key data feature data and associated behavior data.

[0014] As a preferred solution of the present invention, the specific steps of unsupervised learning and clustering between the key data feature data and the associated behavior data are as follows:

[0015] Select the main influencing factors of key data, feature data and associated behavior data as the main influencing features and select several cluster centers; convert the influencing feature data of key data feature data into influencing feature vectors, sort the influencing feature vectors of key data feature data according to the vector size, and select the first K feature vectors as the main influencing factors;

[0016] Based on the distance between the associated behavior data and the cluster center of the main influencing features, the initial key data feature classification is performed;

[0017] Adjust and modify the key data features of the preliminary classification according to the preset principle information;

[0018] Determine whether the classification of key data features is reasonable. If the classification is unreasonable, modify and adjust the classification of key data features again;

[0019] If the classification is reasonable, the final classification of key data features is completed.

[0020] As a preferred solution of the present invention, wherein: after the target model component data set is subjected to multi-task linkage execution processing to form the linked data of the target model component data set, the linked data is determined by constructing a convolutional neural network model, and the specific steps are as follows:

[0021] Based on the tree topological relationship characteristics of the model components, the key data feature data in the target model component data set is gradually decomposed according to the tree topological structure to obtain BIM models of several decomposed sub-target model components corresponding to the target model component;

[0022] The BIM models of the sub-target model components are respectively passed through the convolutional neural network model as a filter to obtain the BIM model feature vectors of the corresponding decomposed sub-target model components; that is, the convolutional neural network model is used to perform convolution processing on the BIM model feature data of the sub-target model components input into the convolutional neural network model in the forward transfer of each layer, and the convolution feature map is obtained after convolution processing;

[0023] Performing mean pooling based on a local feature matrix on the convolution feature map to obtain a pooled feature map; and performing nonlinear activation processing on the pooled feature map to generate an activated feature map;

[0024] According to the activated feature graph, the BIM model feature vector of the new sub-target model component and the BIM model of the new sub-target model component are associated and determined, and output as the linked data;

[0025] Among them, the output of the last layer of the convolutional neural network model is the new feature vector of the BIM model of the sub-target model component after the decomposition, and the input of the first layer of the convolutional neural network model is the BIM model of the sub-target model component after the decomposition.

[0026] As a preferred solution of the present invention, the AI ​​big model analyzes the model component data aggregation features described in the target model component data set, and evaluates according to the big model linkage expert group to generate the linkage analysis results of the target model component data set formed by the BIM system, which are specifically as follows:

[0027] Building a large model linkage expert group based on the model component data aggregation features described in the target model component data set;

[0028] Based on the evaluation of the model component data aggregation features by multiple large model linkage expert groups, the BIM model feature vectors of several trained decomposed sub-target model components of the target model component in the model component data aggregation features and the evaluation matrices of the multiple large model linkage expert groups are calculated to obtain the evaluation matrices of the BIM model feature vectors of the decomposed sub-target model components by the multiple large model linkage expert groups;

[0029] Calculate the global mean of the new feature vector matrix of the BIM model corresponding to each decomposed sub-target model component trained in the evaluation matrix of several large model linkage expert groups, and generate the linkage feature vector of the BIM model of the target model component in the target model component data set determined by the large model linkage expert group;

[0030] Obtaining a linkage analysis result of a target model component BIM model in a target model component data set according to a linkage feature vector and a linkage evaluation model of the BIM model;

[0031] wherein the linkage feature vector of the BIM model is filtered through a classifier to obtain a classification loss function value;

[0032] Based on the classification loss function value and through back propagation of gradient descent, the convolutional neural network model as a filter and the classifier are evaluated and trained by a large model linkage expert group.

[0033] As a preferred solution of the present invention, after generating the linkage feature vector of the target model component BIM model in the target model component data set determined by the large model linkage expert group, the linkage evaluation model specifically adopts the following formula for calculation, as follows:

[0034] ;

[0035] ;

[0036] in, For the The label value of the linkage feature vector of each BIM model;

[0037] For the The linkage feature vector of the BIM model The evaluation value of the large model linkage expert group;

[0038] For the The weight of the linkage feature vector of each BIM model to the BIM model feature vector of the decomposed sub-target model component;

[0039] For the The confidence index of the linkage feature vector of each BIM model; For the The total number of items in the linkage eigenvector analysis of the BIM model.

[0040] As a preferred solution of the present invention, wherein: after the target model component data set is subjected to multi-task linkage execution processing to form the linked data of the target model component data set, the multi-task linkage execution is specifically as follows:

[0041] Determine whether the resource pool of the linkage task is full, wherein the resource pool of the linkage task is created according to the configuration of the user on the engineering management BIM system, and the total amount of BIM models of the target model component in the resource pool of the linkage task corresponds to the concurrent amount of the BIM models of the sub-target model component;

[0042] If the resource pool is not full, after multi-task linkage matching is performed for the BIM models of each sub-target model component, linkage scheduling information and the resource pool are allocated for output; wherein the linkage scheduling information includes priority scheduling information of the BIM models of the sub-target model components, that is, according to the priority scheduling information of the BIM models of the sub-target model components, the BIM model linkage task of the sub-target model component with the highest priority in the linked data is inserted into the current matching linkage memory queue, and resources are allocated to the BIM model matched by the linkage feature vector of the BIM model of the target model component with the highest priority in the linked data for matching linkage and visual presentation.

[0043] On the one hand, the present invention provides a system for linking a construction management BIM system with an AI artificial intelligence system, comprising:

[0044] A BIM system association mining module is used to perform key data mining operations on model component data in the BIM system to output associated key data features in the model component data in the BIM system;

[0045] A data processing module, used to form associated behavior data and key data feature data corresponding to the model component data according to the associated key data features, and generate a target model component data set;

[0046] A calling aggregation module is used to call the target model component data set based on the AI ​​big model, and perform data aggregation operations on the matching model component data, associated behavior data and key data feature data in the target model component data set to form model component data aggregation features;

[0047] An analysis module, configured to analyze the model component data aggregation features described in the target model component data set by the AI ​​big model, evaluate according to the big model linkage expert group, generate linkage analysis results of the target model component data set formed by the BIM system, and divide the linkage tasks;

[0048] An execution linkage module is used to respond to the linkage task division and linkage analysis results, perform multi-task linkage execution processing on the target model component data set to form the linked data of the target model component data set, and simultaneously generate linkage scheduling information matching the target model component data set, match and link the linked data with the target model component data set based on the matched linkage scheduling information, and perform visual presentation.

[0049] Beneficial effects of the present invention: The present invention mobilizes the AI ​​model to perform data mining and realizes linkage with the BIM system. It can mine, cluster, intelligently analyze, linkage schedule and feedback the model component related data in the BIM system, and simultaneously adds the model analysis of AI linkage evaluation and the priority scheduling output display for the sub-target model parts of the model component, thereby improving the linkage processing effect and further promoting the optimization of construction organization design and resource allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0051] Figure 1 It is a flow chart of the linkage method between the engineering management BIM system and AI artificial intelligence of the present invention;

[0052] Figure 2 A flowchart of unsupervised learning and clustering between key data feature data and associated behavior data of the present invention;

[0053] Figure 3 A flow chart of the present invention for determining linked data by constructing a convolutional neural network model;

[0054] Figure 4 A flow chart of the present invention for generating linkage analysis results by evaluating the linkage expert group based on the large model;

[0055] Figure 5 It is a flowchart of the multi-task linkage execution of the present invention;

[0056] Figure 6 It is a schematic diagram of the modular structure of the linkage system of the engineering management BIM system and AI artificial intelligence of the present invention. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.

[0058] Reference Figure 1-Figure 6 , is an embodiment of the present invention, which provides a method and system for linking a construction management BIM system with AI artificial intelligence, including:

[0059] like Figure 1 As shown, by performing key data mining operations on the model component data in the BIM system to output the key data features associated with the model component data in the BIM system, and forming associated behavior data and key data feature data corresponding to the model component data, a target model component data set is generated;

[0060] Based on the AI ​​big model, the target model component data set is called, and the matching model component data, associated behavior data and key data feature data in the target model component data set are aggregated to form model component data aggregation features; the AI ​​big model analyzes the model component data aggregation features in the target model component data set, and the big model linkage expert group evaluates it, generates the linkage analysis results of the target model component data set formed by the BIM system, and divides the linkage tasks;

[0061] In response to the linkage task division and linkage analysis results, the target model component data set is subjected to multi-task linkage execution processing to form the linked data of the target model component data set, and linkage scheduling information matching the target model component data set is generated at the same time. Based on the matching linkage scheduling information, the linked data is matched and linked with the target model component data set, and a visual presentation is performed.

[0062] Specifically, in this embodiment, the linkage task division is specifically based on the associated behavior data and key data feature data in the target model component data set, the model components are screened through the analysis of the AI ​​big model, and the priority of the task is determined and matched according to the influencing factors of the key data features, and the linkage analysis results of the corresponding level are generated. In addition, the influencing features of the key data feature data include: geometric information features, non-spatial information features, topological relationship features, model complexity features, spatial area features of data, model granularity features, texture features, synergy features, cost attribute features, and information output mode state features.

[0063] Preferably, in this embodiment, the matching model component data, associated behavior data and key data feature data in the target model component data set are subjected to data aggregation operations to form model component data aggregation features. The aggregation operation specifically uses a K-means clustering algorithm to perform unsupervised learning and clustering between the key data feature data and the associated behavior data.

[0064] like Figure 2 As shown, the specific steps of unsupervised learning and clustering between key data feature data and associated behavior data in this embodiment are as follows:

[0065] Select the main influencing factors of key data feature data and associated behavior data as influencing features and select several cluster centers; that is, convert the influencing feature data of key data feature data into influencing feature vectors and determine them, sort the influencing feature vectors of key data feature data according to the size of the determined vectors, and select the first K feature vectors as the main influencing factors;

[0066] The associated behavior data is classified into initial key data feature categories based on the distance from the cluster center of the main influencing factors affecting the feature;

[0067] Preset the principle information of initial key data feature classification and modify and adjust the key data feature classification accordingly according to the preset principle information;

[0068] Determine whether the classification of key data features is reasonable. If the classification is unreasonable, adjust the classification again;

[0069] If the classification is reasonable, the final classification of key data features is completed.

[0070] like Figure 3 As shown, this embodiment needs to emphasize that after the target model component data set is processed by multi-task linkage execution, the linked data of the target model component data set is formed, and the linked data is determined by building a convolutional neural network model. The specific steps are as follows:

[0071] Based on the tree topological relationship characteristics of the model components, the key data feature data in the target model component data set is gradually decomposed in tree topology to obtain the BIM models of several decomposed sub-target model components corresponding to the target model component;

[0072] The BIM models of the sub-target model components are respectively passed through the convolutional neural network model as a filter to obtain the BIM model feature vectors of the corresponding decomposed sub-target model components; that is, the convolutional neural network model is used to perform convolution processing on the BIM model feature data of the sub-target model components input into the convolutional neural network model in the forward transfer of each layer, and the convolution feature map is obtained after convolution processing;

[0073] Performing mean pooling on the convolution feature map based on the local feature matrix to obtain a pooled feature map; and

[0074] Perform nonlinear activation on the pooled feature map to obtain an activated feature map;

[0075] According to the activated feature graph, the BIM model feature vector of the new sub-target model component and the BIM model of the new sub-target model component are associated and determined, and output as the linked data;

[0076] Among them, the output of the last layer of the convolutional neural network model is the new feature vector of the BIM model of the sub-target model component after decomposition, and the input of the first layer of the convolutional neural network model is the BIM model of the sub-target model component after decomposition.

[0077] like Figure 4 As shown, in this embodiment, the AI ​​big model further analyzes the model component data aggregation characteristics in the target model component data set, evaluates according to the big model linkage expert group, and generates the linkage analysis results of the target model component data set formed by the BIM system, which are as follows:

[0078] Construct a large model linkage expert group based on the model component data aggregation features in the target model component data set;

[0079] Based on the evaluation of the model component data aggregation features by multiple large model linkage expert groups, the BIM model feature vectors of several trained decomposed sub-target model components of the target model component in the model component data aggregation features and the evaluation matrices of the multiple large model linkage expert groups are calculated to obtain the evaluation matrices of the BIM model feature vectors of the decomposed sub-target model components by the multiple large model linkage expert groups;

[0080] Calculate the global mean of the new feature vector matrix of the BIM model corresponding to each decomposed sub-target model component trained in the evaluation matrix of several large model linkage expert groups, and generate the linkage feature vector of the BIM model of the target model component in the target model component data set determined by the large model linkage expert group;

[0081] According to the linkage feature vector and linkage evaluation model of the BIM model, a linkage analysis result of the BIM model of the target model component in the target model component data set is obtained;

[0082] Among them, the linkage feature vector of the BIM model is filtered through a classifier to obtain the classification loss function value; and based on the classification loss function value and through the back propagation of gradient descent, the convolutional neural network model and classifier as the filter are evaluated and trained by the large model linkage expert group.

[0083] After the linkage evaluation model of this embodiment generates the linkage feature vector of the target model component BIM model in the target model component data set determined by the large model linkage expert group, the linkage evaluation model of this embodiment specifically uses the following formula for calculation, as follows:

[0084] ;

[0085] ;

[0086] in, For the The label value of the linkage feature vector of each BIM model;

[0087] For the The linkage feature vector of the BIM model The evaluation value of the large model linkage expert group;

[0088] For the The weight of the linkage feature vector of each BIM model on the BIM model feature vector of the decomposed sub-target model component;

[0089] For the The confidence index of the linkage feature vector of each BIM model; For the The total number of items in the linkage eigenvector analysis of the BIM model.

[0090] like Figure 5 As shown, in this implementation, the target model component data set is preferably processed by multi-task linkage execution to form the linked data of the target model component data set. The multi-task linkage execution is specifically as follows:

[0091] Determine whether the resource pool of the linkage task is full, wherein the resource pool of the linkage task is created according to the user's configuration on the engineering management BIM system, and the total amount of BIM models of the target model component in the resource pool of the linkage task corresponds to the concurrent amount of BIM models of the sub-target model component;

[0092] If the resource pool is not full, after multi-task linkage matching is performed for the BIM models of each sub-target model component, linkage scheduling information and the resource pool are allocated for output; the linkage scheduling information includes the priority scheduling information of the BIM models of the sub-target model components, that is, according to the priority scheduling information of the BIM models of the sub-target model components, the BIM model linkage task of the sub-target model component with the highest priority in the linkage data is inserted into the current matching linkage memory queue, and resources are allocated to the BIM model matched by the linkage feature vector of the BIM model of the target model component with the highest priority in the linkage data for matching linkage and visual presentation.

[0093] like Figure 6 As shown, based on the above method, this embodiment also provides a linkage system of a project management BIM system and AI artificial intelligence, including:

[0094] The BIM system association mining module 101 is used to perform key data mining operations on the model component data in the BIM system to output the key data features associated with the model component data in the BIM system.

[0095] The data processing module 102 is used to form associated behavior data and key data feature data corresponding to the model component data according to the associated key data features, and generate a target model component data set;

[0096] The calling aggregation module 103 is used to call the target model component data set based on the AI ​​big model, and perform data aggregation operations on the matching model component data, associated behavior data and key data feature data in the target model component data set to form model component data aggregation features;

[0097] The analysis module 104 is used to analyze the aggregation characteristics of the model component data in the target model component data set by the AI ​​big model, evaluate it according to the big model linkage expert group, generate the linkage analysis results of the target model component data set formed by the BIM system, and divide the linkage tasks;

[0098] The execution linkage module 105 is used to respond to the linkage task division and linkage analysis results, perform multi-task linkage execution processing on the target model component data set to form the linked data of the target model component data set, and simultaneously generate linkage scheduling information that matches the target model component data set. Based on the matched linkage scheduling information, the linked data is matched and linked with the target model component data set, and a visual presentation is performed.

[0099] Further, this embodiment also specifically provides a case of matching linkage recommendation of building model components based on AI artificial intelligence and engineering management BIM system, as follows:

[0100] Step S201: Compliance and rationality verification is performed based on AI artificial intelligence. According to the basic principles of construction organization design and the built-in national standard requirements, the standardization and rationality of the site layout are verified, and adjustments and optimizations are made according to the verification results. Based on the optimization results, two-dimensional drawings, site deployment plans and material requirements are intelligently output;

[0101] Step S202, creating a building planning document online, obtaining data on functional zoning, rooms and spaces, such as span, depth, clear height, type, and suitable number of floors;

[0102] Step S203, based on the data obtained in the previous step, create a geometric model of a specific functional partition, room and space in a web browser;

[0103] Step S204, calling the target model component data set based on the AI ​​big model, and having the AI ​​big model analyze the model component data aggregation features in the target model component data set, and automatically generating many combinations of design layouts through the enhanced deep learning algorithm of the AI ​​big model, each design layout having a corresponding automatic evaluation index value;

[0104] Step S205, determining the high-quality options that meet the requirements and carrying out the specific configuration of the next step of AI recommendation model construction;

[0105] Step S206, import the optimal layout of rooms and spaces into the BIM system, determine the design details, and finally complete the construction drawing design.

[0106] Based on the above cases, we can know that in the design stage of building facilities using BIM system, AI can help with modeling design and early conceptual plan generation, and achieve design goals by controlling renderings and generating architectural floor plans. AI can play a role in the optimization design stage, and BIM system can be combined with AI to improve drawing efficiency by intelligently identifying and matching building peripheral components. In addition, combining BIM models, point cloud models, 720 panoramic technology and other visualization methods can create a richer and more intuitive value experience, bringing convenience and efficiency to the whole process of construction management.

[0107] In summary, the present invention, by mobilizing the AI ​​model for mining and then linking with the BIM system, can mine, cluster, intelligently analyze, link schedule and feedback the model component related data in the BIM system, and simultaneously adds the model analysis of AI linkage evaluation and the priority scheduling output display for the sub-target model parts of the model component, thereby improving the linkage processing effect, and further facilitating the construction organization design and resource allocation optimization.

[0108] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium.

[0109] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples, unless they are contradictory.

[0110] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0111] Any process or method description in the flow chart or otherwise described herein can be understood to represent a module, fragment or portion of a code including one or more executable instructions for implementing the steps of a specific logical function or process. And the scope of the preferred embodiment of the present application includes other implementations, in which the functions may not be performed in the order shown or discussed, including in a substantially simultaneous manner or in a reverse order according to the functions involved.

[0112] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, which can be embodied in any computer-readable medium for use by an instruction execution system, apparatus or device (such as a computer-based system, a system including a processor or other system that can fetch instructions from an instruction execution system, apparatus or device and execute instructions), or used in combination with these instruction execution systems, apparatuses or devices.

[0113] It should be understood that the various parts of the present application can be implemented with hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented with software or firmware stored in a memory and executed by a suitable instruction execution system. All or part of the steps of the above embodiment method can be completed by instructing the relevant hardware through a program, which can be stored in a computer-readable storage medium, and when the program is executed, it includes one of the steps of the method embodiment or a combination thereof.

[0114] In addition, each functional unit in each embodiment of the present application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into one module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of a software functional module. If the above-mentioned integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The storage medium can be a read-only memory, a disk or an optical disk, etc.

[0115] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for linking an engineering management BIM system with AI artificial intelligence, characterized in that: The following steps are involved: By performing key data mining operations on the model component data in the BIM system, the associated key data features of the model component data in the BIM system are extracted, the associated behavior data and key data feature data corresponding to the model component data are formed, and the target model component data set is generated by integration; Based on the AI ​​big model, the target model component data set is called, and the matching model component data, associated behavior data and key data feature data in the target model component data set are aggregated to form aggregate features of the model component data; The AI ​​big model analyzes the aggregation features of the above-mentioned model component data, generates linkage analysis results of the target model component data set in the BIM system based on the evaluation of the big model linkage expert group, and divides the linkage tasks; In response to the linkage task division and linkage analysis results, after multi-task linkage execution processing is performed on the target model component data set, the linked data of the target model component data set is formed, and linkage scheduling information matching the target model component data set is generated at the same time. Based on the matched linkage scheduling information, the linked data is matched and linked with the target model component data set, and visual presentation is achieved.

2. The method for linking a project management BIM system with AI artificial intelligence as claimed in claim 1, characterized in that: The linkage task division is based on the associated behavior data and key data feature data in the target model component data set. The model components are screened through the analysis of the AI ​​big model, and the task priority is determined according to the influencing factors of the key data features to generate linkage analysis results of the corresponding level.

3. A method for linking a project management BIM system with AI artificial intelligence as claimed in claim 2, characterized in that: The influencing characteristics of the key data feature data include: geometric information characteristics, non-spatial information characteristics, topological relationship characteristics, model complexity characteristics, data space area characteristics, model granularity characteristics, texture characteristics, synergy characteristics, cost attribute characteristics and state characteristics of information output mode.

4. The method for linking a BIM system for engineering management with AI artificial intelligence as claimed in claim 3, characterized in that: The matched model component data, associated behavior data and key data feature data in the target model component data set are aggregated by a K-means clustering algorithm to form aggregate features of the model component data, and unsupervised learning and clustering are performed on the key data features and associated behavior data.

5. The method for linking a project management BIM system with AI artificial intelligence as claimed in claim 4, characterized in that: The specific steps of unsupervised learning and clustering between the key data feature data and the associated behavior data include: Select the main influencing factors in the key data feature data and the associated behavior data as the main features, and select several cluster centers based on these features; convert the influencing features of the key data features into influencing feature vectors, and sort the key data feature data according to the numerical value of the vectors, and select the first K feature vectors as the main influencing factors; Perform preliminary classification of key data features based on the distance between the associated behavior data and the cluster center of the main features; Adjust and modify the key data features of the preliminary classification according to the preset rules; Determine whether the classification of key data features is reasonable. If the classification is unreasonable, adjust the classification again; If the classification is reasonable, the final classification of key data features is completed.

6. The method for linking a project management BIM system with AI artificial intelligence as claimed in claim 1, characterized in that: After the target model component data set is subjected to multi-task linkage execution processing, the linked data of the target model component data set is formed, and the linked data is determined by constructing a convolutional neural network model. The specific steps are as follows: Based on the tree-like topological relationship characteristics of the model components, the key data in the target model component data set is gradually decomposed according to the tree-like topological structure to obtain the BIM models of several target model components; The BIM models of the sub-target model components are respectively input into the convolutional neural network model as a filter to obtain the BIM model feature vectors of the decomposed sub-target model components; that is, the convolutional neural network model is used to perform convolution processing in the forward transfer of each layer to obtain the convolution feature map of the BIM model features of the sub-target model components; Performing mean pooling processing based on a local feature matrix on the convolution feature map to obtain a pooled feature map, and performing nonlinear activation processing on the pooled feature map to generate an activated feature map; According to the activated feature graph, the BIM model feature vector of the new sub-target model component and the BIM model of the new sub-target model component are associated and determined, and output as the linked data; Among them, the output of the last layer of the convolutional neural network model is the new feature vector of the BIM model of the sub-target model component after decomposition, and the input of the first layer is the BIM model of the sub-target model component after decomposition.

7. The method for linking a BIM system for engineering management with AI artificial intelligence as claimed in claim 6, characterized in that: The AI ​​big model analyzes the model component data aggregation features described in the target model component data set, and generates the linkage analysis results of the target model component data set in the BIM system based on the evaluation of the big model linkage expert group. The specific steps are as follows: Building a large model linkage expert group based on the model component data aggregation features described in the target model component data set; Based on the evaluation of the model component data aggregation features by multiple large model linkage expert groups, the BIM model feature vectors of several sub-target model components after training decomposition of the target model component in the model component data aggregation features are calculated, and the evaluation matrices of the multiple large model linkage expert groups are calculated to obtain the evaluation matrices of the BIM model feature vectors of the decomposed sub-target model components by several large model linkage expert groups; Calculate the global mean of the BIM model feature vector matrix of each decomposed sub-target model component in the evaluation matrix of several large model linkage expert groups respectively, and generate the linkage feature vector of the BIM model of the target model component in the target model component data set determined by the large model linkage expert group; Obtaining a linkage analysis result of a target model component BIM model in a target model component data set according to the linkage feature vector and linkage evaluation model of the BIM model; Among them, the linkage feature vector of the BIM model is filtered by a classifier to obtain a classification loss function value; and based on the classification loss function value, the convolutional neural network model and the classifier as the filter are evaluated and trained by the large model linkage expert group through back propagation of gradient descent.

8. The method for linking a project management BIM system with AI artificial intelligence as claimed in claim 7, characterized in that: After generating the BIM model linkage feature vector of the target model component in the target model component data set determined by the large model linkage expert group, the linkage evaluation model specifically uses the following formula for calculation: ； ； in, For the The label value of the linkage feature vector of each BIM model; For the The linkage feature vector of the BIM model The evaluation value of the large model linkage expert group; For the The weight of the linkage feature vector of each BIM model to the BIM model feature vector of the decomposed sub-target model component; For the The confidence index of the linkage feature vector of each BIM model; For the The total number of items in the linkage eigenvector analysis of the BIM model.

9. The method for linking a BIM system for engineering management with AI artificial intelligence according to claim 1, characterized in that: After the target model component data set is subjected to multi-task linkage execution processing, the linked data of the target model component data set is formed. The specific multi-task linkage execution steps are as follows: Determine whether the resource pool of the linkage task is full, wherein the resource pool of the linkage task is created according to the configuration of the user in the engineering management BIM system, and the total amount of BIM models of the target model component in the resource pool of the linkage task corresponds to the concurrent amount of the BIM models of the sub-target model component; If the resource pool is not full, multi-task linkage matching is performed for the BIM models of each sub-target model component, and linkage scheduling information and resource pool are allocated for output; The linkage scheduling information includes priority scheduling information of the BIM model of the sub-target model component. According to the priority scheduling information of the BIM model of the sub-target model component, the BIM model linkage task of the sub-target model component with the highest priority in the linked data is inserted into the current matching linkage memory queue, and resources are allocated to the BIM model matched by the linkage feature vector of the BIM model of the target model component with the highest priority in the linked data, and matching linkage and visual presentation are performed.

10. A system applied to the linkage method of a project management BIM system and AI artificial intelligence as claimed in claim 1, characterized in that: include: The BIM system association mining module is used to extract the associated key data features in the model component data in the BIM system by performing key data mining operations on the model component data in the BIM system; A data processing module, used to form associated behavior data and key data feature data corresponding to the model component data according to the associated key data features, and generate a target model component data set; A calling aggregation module is used to call the target model component data set based on the AI ​​big model, and perform data aggregation operations on the matched model component data, associated behavior data and key data feature data in the target model component data set to form model component data aggregation features; An analysis module, configured to analyze the model component data aggregation features described in the target model component data set by the AI ​​big model, generate linkage analysis results of the target model component data set in the BIM system based on the evaluation of the big model linkage expert group, and divide the linkage tasks; An execution linkage module is used to respond to the linkage task division and linkage analysis results, perform multi-task linkage execution processing on the target model component data set, form the linked data of the target model component data set, and simultaneously generate linkage scheduling information matching the target model component data set, and match and link the linked data with the target model component data set based on the matching linkage scheduling information to achieve visual presentation.

Images