Artificial intelligence-based building structure aided design method and device
By acquiring structured data from 3D building models, constructing topological diagrams, and utilizing deep learning neural networks to evaluate whether the design meets standards, the problems of low efficiency and incomplete optimization in traditional building structure design are solved, achieving efficient and accurate automated design evaluation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- JIULING (JIANGSU) DIGITAL INTELLIGENT TECH CO LTD
- Filing Date
- 2021-08-24
- Publication Date
- 2026-06-16
AI Technical Summary
Traditional building structural design relies on human experience, resulting in low adjustment efficiency and incomplete optimization, leading to low work efficiency.
By acquiring structured data from 3D building models, constructing topological diagrams, and utilizing deep learning neural networks to evaluate whether the design between building structures conforms to building standards, automated design evaluation is achieved.
It improves design efficiency and accuracy, reduces reliance on human experience, and enables rapid identification of design errors.
Smart Images
Figure CN115718938B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of architectural design technology, and in particular to an artificial intelligence-based auxiliary design method and device for architectural structures. Background Technology
[0002] With the development of the times, science and technology are gradually becoming more specialized, comprehensive, and quantitative. In architectural structural design, selecting a suitable structural system based on the building's function, layout characteristics, and overall features is crucial. In the traditional architectural structural design process, the design result (i.e., the building structural model) is obtained manually. After completion, the design result is optimized manually based on years of industry experience. For example, depending on the application area of the design result, some regional features are adaptively added or removed, thereby making the design result more compatible with people's needs.
[0003] However, in the traditional design optimization process, relying solely on manual experience to adjust the building structure model may result in low adjustment efficiency and incomplete optimization, leading to low work efficiency. Summary of the Invention
[0004] Therefore, it is necessary to provide an AI-based auxiliary design method and device for building structures to address the aforementioned technical problems.
[0005] In a first aspect, embodiments of this application provide an artificial intelligence-based auxiliary design method for building structures, the method comprising:
[0006] Obtain a three-dimensional building model, generate corresponding structured data based on each building structure in the three-dimensional building model, and extract design relationship data between each building structure;
[0007] A topology diagram is constructed using the design relationship data, and the topology diagram is used to represent the relationship between various building structures;
[0008] The topology diagram is input into a preset deep learning neural network to determine whether the design of each building structure conforms to building standards.
[0009] The building structure includes steel beams, frame beams, steel space frames, ring beams, structural columns, pipe piles, and walls.
[0010] Optionally, generating corresponding structured data based on each building structure in the three-dimensional building model includes:
[0011] Extract the name, floor, material, attribute parameters, type selection, and names of adjacent building structures as structured information for each building structure.
[0012] A list of building components is generated based on the structural information of the building structure according to the floors.
[0013] The data from the list of building components is filled into a structured data table to form the structured data, which is processed based on the floor level.
[0014] Optionally, constructing a topology diagram using the design relationship data includes:
[0015] The design relationship data carries floor marker information. The design relationship data is grouped according to the floor marker information to obtain multiple design relationship group data.
[0016] Determine the correlation strength between the grouped data of the design relationship, and perform clustering based on the correlation strength;
[0017] Multiple preliminary topological structure diagrams are formed based on the clustering and family division, and the multiple preliminary topological structure diagrams are weighted and fused to obtain the topological structure diagram.
[0018] Optionally, determining the correlation strength between the design relationship grouping data includes:
[0019] By parsing the structured information of each building component in the design relationship grouping data, the association relationship between adjacent or related building components can be extracted;
[0020] Optionally, extracting the association relationship between adjacent or related building components based on the association degree includes:
[0021] Based on the structured information, comparison information related to the design standards is extracted according to preset rules;
[0022] The most frequent comparison information in the structured information is extracted first to determine the association. The remaining comparison information is then used for semantic recognition to determine the association. The association strength is then evaluated.
[0023] Secondly, embodiments of this application provide an artificial intelligence-based auxiliary design device for building structures, the device comprising:
[0024] The acquisition module is used to acquire a three-dimensional building model, generate corresponding structured data based on each building structure in the three-dimensional building model, and extract design relationship data between each building structure.
[0025] The construction module is used to construct a topology diagram using the design relationship data, the topology diagram being used to characterize the relationships between various building structures;
[0026] An evaluation module is used to input the topology diagram into a preset deep learning neural network to determine whether the design of each building structure conforms to building standards.
[0027] The building structure includes steel beams, frame beams, steel space frames, ring beams, structural columns, pipe piles, and walls.
[0028] Optionally, the acquisition module includes:
[0029] Structured units are used to extract the name, floor, material, attribute parameters, type selection, and names of adjacent building structures as structured information for each building structure.
[0030] The list unit is used to generate a list of building components based on the structured information of the building structure according to the floors.
[0031] The structured unit is used to fill the data in the list of building components into the structured data table to form the structured data, which is obtained by processing based on the floor.
[0032] Optionally, the building module includes:
[0033] A grouping unit is used to group the design relationship data, which carries floor marker information, according to the floor marker information, to obtain multiple design relationship grouping data.
[0034] Clustering unit, used to determine the correlation strength between the grouped data of the design relationship, and to perform clustering based on the correlation strength;
[0035] The processing unit is used to form multiple preliminary topology diagrams based on the clustering and family division, and to perform weighted fusion processing on the multiple preliminary topology diagrams to obtain the topology diagram.
[0036] Furthermore, the clustering unit is also used to extract the association relationship between adjacent or related building components by parsing the structured information of each building component in the design relationship grouping data; and to evaluate the association degree based on the association relationship between the adjacent or related building components to obtain the association strength.
[0037] Thirdly, embodiments of this application provide a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor, when executing the computer program, implements the following steps:
[0038] Obtain a three-dimensional building model, generate corresponding structured data based on each building structure in the three-dimensional building model, and extract design relationship data between each building structure;
[0039] A topology diagram is constructed using the design relationship data, and the topology diagram is used to represent the relationship between various building structures;
[0040] The topology diagram is input into a preset deep learning neural network to determine whether the design of each building structure conforms to building standards.
[0041] The building structure includes steel beams, frame beams, steel space frames, ring beams, structural columns, pipe piles, and walls.
[0042] The aforementioned AI-based building structure auxiliary design method and device acquires design relationship data and component topology diagrams between individual building structures in a 3D building model. Then, it uses a deep learning model for evaluation, determining whether design modifications are necessary based on the evaluation results. By leveraging neural networks to determine design compliance with standards, automated design evaluation can be achieved. This eliminates the need for manual judgment and adjustment of the building structure model, allowing for rapid identification of design errors and significantly improving design efficiency and accuracy. Attached Figure Description
[0043] Figure 1 This is an internal structural diagram of a computer device in one embodiment;
[0044] Figure 2 This is a flowchart illustrating an AI-based building structure assisted design method in one embodiment.
[0045] Figure 3 This is a flowchart illustrating the detailed steps of step S210 in one embodiment;
[0046] Figure 4 This is a flowchart illustrating the detailed steps of step S220 in one embodiment;
[0047] Figure 5 This is a schematic diagram of an AI-based building structure auxiliary design device in one embodiment. Detailed Implementation
[0048] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0049] The AI-based building structure auxiliary design method provided in this application can be applied to... Figure 1The computer device shown includes a processor, memory, network interface, database, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The database stores the neural network described in the following embodiments; a detailed description of the neural network is provided in the following embodiments. The network interface of the computer device can be used to communicate with other external devices via a network connection. Optionally, the computer device can be a server, a desktop computer, a personal digital assistant, or other terminal devices such as tablets, mobile phones, etc., or it can be a cloud or remote server. This application does not limit the specific form of the computer device. The display screen of the computer device can be a liquid crystal display or an e-ink display. The input device can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device casing, or an external keyboard, touchpad, or mouse. Of course, input devices and displays may not be part of the computer equipment; they can be external devices to the computer equipment.
[0050] Those skilled in the art will understand that Figure 1 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0051] The technical solution of this application and how it solves the above-mentioned technical problems will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.
[0052] It should be noted that the executing entity of the following method embodiments can be an AI-based auxiliary design device for building structures. This device can be implemented as part or all of the aforementioned computer equipment through software, hardware, or a combination of software and hardware. The following method embodiments are described using a computer equipment as an example of the executing entity.
[0053] Figure 2This is a flowchart illustrating an AI-based assisted design method for building structures, provided as one embodiment. This embodiment relates to the process of using a computer device with a neural network to assist in the design of a building structure model. Figure 2 As shown, it includes:
[0054] Step S210: Obtain a three-dimensional building model, generate corresponding structured data based on each building structure in the three-dimensional building model, and extract the design relationship data between each building structure.
[0055] In this embodiment of the invention, the three-dimensional building model is a model data carrier containing various building information. The building structure may include steel beams, frame beams, steel space frames, ring beams, structural columns, pipe piles, walls, etc. Each building structure encompasses a lot of information, such as walls, including interior and exterior walls, wall height, wall thickness, wall material, wall type, etc.
[0056] The data of each building structure stored in the three-dimensional building model is processed into structured data. This structured data contains the relationship data between each building structure and other building structures. This relationship data is not limited to the connection method, but is also related to the building structure selection, floors, region, etc.
[0057] Based on the type of each building structure and the rules determined by prior knowledge, the design relationship data of each building structure is extracted. For example, the design relationship data extracted for load-bearing walls and non-load-bearing walls are different. For specific details, please refer to the national standard.
[0058] Step S220: Construct a topology diagram using the design relationship data. The topology diagram is used to characterize the relationships between various building structures.
[0059] Specifically, the design relationship data and building structure names will be used as nodes in the topology diagram; the building structure names, floors, and connecting edges between adjacent building structures will be used to represent the relationships between building structures; and the weights of the relationships between building structures and floors will be used as the values on the connecting edges.
[0060] It can be understood that if the weight of the relationship between two building structures is not zero, it indicates that there is a connection between them. In this case, in the topology graph, the nodes of these two building structures are connected by an edge, and the value on this edge is the weight of the relationship between the two building structures. If the weight of the relationship between two building structures is zero, it indicates that there is no connection between them, and therefore, no edge is constructed between these two building structure nodes in the topology graph. Following these graph construction rules, by constructing connecting edges between nodes whose relationship weight is not zero and using the relationship weight as the value on the connecting edge, the topology graph can be constructed.
[0061] In this embodiment of the invention, the above-mentioned relationship represents the relationship between different building structures. It can also be the relationship between the attributes (e.g., the relationship between steel beams and walls) or the assignment (the relationship between frame beams and walls) of different building structures. This embodiment does not limit this relationship.
[0062] Step S230: Input the topology diagram into a preset deep learning neural network to determine whether the design of each building structure conforms to building standards.
[0063] The deep learning network determines, at least by processing the topology graph, whether the design of each building structure conforms to building standards.
[0064] Deep learning neural networks are multi-layered convolutional neural networks that can aggregate the feature vectors of a node's neighboring nodes. Building structures are not isolated entities; the density feature of the relationships between building structures measures the degree of correlation between any two building structures.
[0065] It is understandable that the relationships between building structures can be considered from different perspectives:
[0066] There are definitely fixed, strong relationships between building structures located on or adjacent to each other, such as direct fixation, embedding, and reinforcement. These relationships may differ between individual floors, or even between some floors. The closer two building structures are in the neural network, the stronger their relationship.
[0067] Therefore, the design relationships between the derived building structures can be used to compare building standards, thereby achieving the purpose of auxiliary design of building structures.
[0068] The above-mentioned use of neural networks to identify the relationships in the topology diagram can automatically determine whether the design meets building standards. Designers can adjust the building structure model based on the identification and evaluation results, thus eliminating the need to rely on human experience to adjust the building structure model, making the design more efficient and accurate.
[0069] In one embodiment, such as Figure 3 The diagram shown is a detailed flowchart of step S210, including:
[0070] Step S211: Extract the name, floor, material, attribute parameters, selection type, and names of adjacent building structures as structured information for each building structure.
[0071] The 3D building model stores data on various building components, specifically in the form of a data list, but not limited to this. Each component may contain one or more building structures, called building structure objects. The data of each building structure object includes at least its name, floor, material, attribute parameters, selection type, and the names of adjacent building structures. This data is structured for easy retrieval.
[0072] Structured data, also known as row data, is data logically expressed and implemented using a two-dimensional table structure. It strictly adheres to data format and length specifications and is primarily stored and managed through relational databases. Therefore, data stored in a three-dimensional building model can be processed and applied.
[0073] Step S212: Generate a list of building components based on the structured information of the building structure according to the floors.
[0074] Since buildings consist of multiple floors, the relationships between building structures are strongly correlated with the floors. Therefore, it is necessary to first determine the building components within each floor, and then create a list of building components based on the building structure. This results in a better data foundation, which is conducive to forming a better data structure and creating a structural topology diagram.
[0075] Step S213: Fill the data in the building component list into the structured data table to form the structured data, which is obtained by processing based on the floor.
[0076] To facilitate the creation of high-quality topology diagrams and adapt to the computational model of convolutional neural networks, structured data for each or related layers is generated based on floor levels. The relationships between floors are relatively simple, namely, upper and lower floors; the walls of each floor, except for load-bearing walls, can be flexible. Steel beam structures, however, require consideration of stress requirements and standards.
[0077] Furthermore, based on structured data, relevant data can be extracted according to the requirements of the component topology diagram.
[0078] In this embodiment of the invention, the computer device can traverse a three-dimensional building model and obtain data on different building structures based on the traversed structure. This data is then filtered and organized to obtain usable structured data. This enables the automatic acquisition of structural data from a complete three-dimensional building model, thus automating data acquisition and organization with high efficiency.
[0079] In one embodiment, such as Figure 4 The diagram shown is a detailed flowchart of step S220, in which the generation of corresponding structured data based on each building structure in the 3D building model specifically includes:
[0080] Step S221: The design relationship data carries floor marking information. The design relationship data is grouped according to the floor marking information to obtain multiple design relationship group data.
[0081] As mentioned above, the structured data is formed based on the floors. The extracted design relationship data uses the floors as marker information and is further grouped based on this marker information. This can form the process grouping data of the corresponding building, or merge floors with similar or identical designs to reduce the amount of calculation.
[0082] Step S222: Determine the correlation strength between the grouped data of the design relationship, and perform clustering based on the correlation strength.
[0083] Understandably, the design relationships between building structures are diverse, and even the connections between building structures can vary greatly. Therefore, targeted cluster analysis is necessary to better determine the design relationships between various building structures.
[0084] Specifically, by parsing the structured information of each building component in the design relationship grouping data, the association relationships between adjacent or related building components are extracted. Then, the association degree is evaluated based on the association relationships between the adjacent or related building components to obtain the association strength.
[0085] For example, the relationships include the relationship between frame beams and loads, the relationship between span structures and column grid spacing, the relationship between bathroom structural design, the relationship between span spatial structures and spatial grid structures, and the relationship between the cross-sectional height and span of frame beams.
[0086] Specifically, the constraints on the relationship between frame beams and loads (the bending moment at the beam ends is the largest under both horizontal and vertical loads; based on the constraints formed by excavation, the frame beams are designed as variable cross-section haunched beams), the constraints on the relationship between span structure and column grid spacing (large-span factory structures are sensitive to snow loads; the calculation benchmark value for snow load (basic snow pressure) should be a 100-year recurrence period value; based on the constraints formed by excavation, the corresponding column grid spacing is selected to achieve the optimal steel quantity; according to the constraints, solid-web steel beams are used for spans greater than 30 meters, and truss steel beams are used for spans of 10 meters, etc.), and the constraints on the relationship between bathroom structure design (in bathrooms...) In structural design, based on the constraints obtained from excavation, the waterproofing height in the shower area is typically 1.8m, and in non-shower areas, it is 0.3m. Attention should be paid to the drainage methods of individual bathroom fixtures. For example, if the bathtub is brick-built, waterproofing treatment needs to be considered; if it is a prefabricated bathtub, the drainage point needs to be considered. This should be clearly stated during the plumbing and electrical handover. Pre-installation of 40mm or 50mm branch pipes is also important. Imported bathtubs often have a drain outlet that is flush with the floor drain. Other constraints include the relationship between span spatial structures and spatial grid structures (spatial grid structures are widely used in large-span, large-space structures, offering high material utilization), and the relationship between the cross-sectional height of the frame beams and the span (the cross-sectional height of the frame beams is taken as 1 / 20 to 1 / 22 of the span).
[0087] In an embodiment of the present invention, for example, after the correlation is identified, it is found that a large-span factory building uses 32-meter solid-web steel beams. Matching this correlation with the building standard (solid-web steel beams for spans greater than 30 meters) will yield a design result that meets the requirements.
[0088] Furthermore, the following methods can be used to assist in achieving the above steps:
[0089] Based on the structured information, comparison information related to design standards is extracted according to preset rules. In one embodiment, preset rules corresponding to specific requirements can be set, for example, rules for walls, walls and steel beams, or one or more other objects.
[0090] The comparison information that appears most frequently in the structured information is extracted first to determine the association relationship. The remaining comparison information is then used for semantic recognition, and the association relationship is determined based on the semantic recognition results.
[0091] For example, semantic association refers to the semantic relationship between building structure names, such as whether a semantic association exists and the degree of semantic association. Specifically, semantic features containing contextual semantic information can be extracted from each building structure name. Then, by calculating the similarity between the semantic features of each building structure name, pairs of building structure names with semantic association are identified. If the similarity between the semantic features of the building structure names is high enough, they are considered to have a semantic association. The number of building structure name pairs contained in the structured information is counted and normalized; the final normalized result is used as the semantic relationship weight between the building structure name and the design.
[0092] Long Short-Term Memory (LSTM) networks can be used to extract semantic features of building structure names, thereby identifying pairs of building structure names with semantic relationships, and determining the semantic relationship weights between building structures included in case keyword pairs.
[0093] According to the above processing scheme, the semantic relationship weights between building structures with semantic association can be determined, and the semantic relationship weights between building structures without semantic association can be set to zero. This allows the determination of the semantic relationship weights between any two building structures.
[0094] By obtaining the semantic relationship weights between multiple building structures, the importance or frequency of occurrence of the building structures can be determined, and the association strength can be obtained. Based on the association strength, clustering can be performed.
[0095] Step S223: Based on the clustering and family formation, multiple preliminary topological structure diagrams are formed, and the multiple preliminary topological structure diagrams are weighted and fused to obtain the topological structure diagram.
[0096] Based on the above clustering and grouping, multiple topology diagrams with different design relationships can be formed, and these multiple topology diagrams represent different design features. By performing a weighted fusion process on these multiple topology diagrams, the aforementioned topology diagram can be obtained.
[0097] Finally, the topology diagram is input into a pre-set deep learning neural network to determine whether the design of each building structure conforms to building standards. Users can modify the design based on outputs that do not meet the building standards. The deep learning neural network consists of pre-trained relationships. For example, the constraint condition for the relationship between coupling beams and vertical loads; under vertical loads, the coupling beams are removed.
[0098] In this embodiment of the invention, a computer device inputs the topological diagram of the aforementioned building structure into a neural network. This neural network can identify multiple building structures and various relationships between different building structure objects. The computer device then performs matching and judgment on the obtained relationships according to the benchmark building labels to obtain the identification and evaluation results. This enables automated design evaluation, eliminating the need for manual judgment and adjustment of the building structure model, quickly locating design errors, and significantly improving design efficiency and accuracy.
[0099] It should be understood that, although Figure 2-4 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 2-4 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.
[0100] In one embodiment, such as Figure 5 As shown, an artificial intelligence-based building structure auxiliary design device is provided, comprising:
[0101] The acquisition module 501 is used to acquire a three-dimensional building model, generate corresponding structured data based on each building structure in the three-dimensional building model, and extract design relationship data between each building structure; wherein, the building structure includes steel beams, frame beams, steel space frames, ring beams, structural columns, pipe piles, and walls.
[0102] The construction module 502 is used to construct a topology diagram using the design relationship data, the topology diagram being used to characterize the relationships between various building structures.
[0103] Evaluation module 503 is used to input the topology diagram into a preset deep learning neural network to determine whether the design of each building structure conforms to building standards.
[0104] In one embodiment, the acquisition module 501 includes:
[0105] Structured units are used to extract the name, floor, material, attribute parameters, type selection, and names of adjacent building structures as structured information for each building structure.
[0106] The list unit is used to generate a list of building components based on the structured information of the building structure according to the floors.
[0107] The structured unit is used to fill the data in the list of building components into the structured data table to form the structured data, which is obtained by processing based on the floor.
[0108] In one embodiment, the building module 502 includes:
[0109] A grouping unit is used to group the design relationship data, which carries floor marker information, according to the floor marker information, to obtain multiple design relationship grouping data.
[0110] Clustering unit, used to determine the correlation strength between the grouped data of the design relationship, and to perform clustering based on the correlation strength;
[0111] The processing unit is used to form multiple preliminary topology diagrams based on the clustering and family division, and to perform weighted fusion processing on the multiple preliminary topology diagrams to obtain the topology diagram.
[0112] In one embodiment, the clustering unit is further configured to extract the association relationships of adjacent or related building components by parsing the structured information of each building component in the design relationship grouping data; and to evaluate the association degree based on the association relationships of the adjacent or related building components to obtain the association strength.
[0113] Specific limitations regarding AI-based building structure auxiliary design devices can be found in the above-mentioned limitations on intelligent building structure design evaluation methods, and will not be repeated here. Each module in the aforementioned intelligent building structure design evaluation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.
[0114] As shown in Figure 6, in one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:
[0115] Obtain a three-dimensional building model, generate corresponding structured data based on each building structure in the three-dimensional building model, and extract design relationship data between each building structure;
[0116] A topology diagram is constructed using the design relationship data, and the topology diagram is used to represent the relationship between various building structures;
[0117] The topology diagram is input into a preset deep learning neural network to determine whether the design of each building structure conforms to building standards.
[0118] The building structure includes steel beams, frame beams, steel space frames, ring beams, structural columns, pipe piles, and walls.
[0119] In one embodiment, when the processor executes the computer program, it further implements the step of "generating corresponding structured data based on each building structure in the three-dimensional building model", including:
[0120] Extract the name, floor, material, attribute parameters, type selection, and names of adjacent building structures as structured information for each building structure.
[0121] A list of building components is generated based on the structural information of the building structure according to the floors.
[0122] The data from the list of building components is filled into a structured data table to form the structured data, which is processed based on the floor level.
[0123] In one embodiment, when the processor executes the computer program, it further implements the step of "constructing a topology diagram using the design relationship data", including:
[0124] The design relationship data carries floor marker information. The design relationship data is grouped according to the floor marker information to obtain multiple design relationship group data.
[0125] Determine the correlation strength between the grouped data of the design relationship, and perform clustering based on the correlation strength;
[0126] Multiple preliminary topological structure diagrams are formed based on the clustering and family division, and the multiple preliminary topological structure diagrams are weighted and fused to obtain the topological structure diagram.
[0127] Specifically, determining the correlation strength between the grouped design relationship data includes:
[0128] By parsing the structured information of each building component in the design relationship grouping data, the association relationship between adjacent or related building components can be extracted;
[0129] The correlation strength is determined by assessing the relationship between adjacent or related building components.
[0130] The extraction of the relationships between adjacent or related building components includes:
[0131] Based on the structured information, comparison information related to the design standards is extracted according to preset rules;
[0132] The comparison information that appears most frequently in the structured information is extracted first to determine the association relationship. The remaining comparison information is then used for semantic recognition, and the association relationship is determined based on the semantic recognition results.
[0133] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:
[0134] Obtain a three-dimensional building model, generate corresponding structured data based on each building structure in the three-dimensional building model, and extract design relationship data between each building structure;
[0135] A topology diagram is constructed using the design relationship data, and the topology diagram is used to represent the relationship between various building structures;
[0136] The topology diagram is input into a preset deep learning neural network to determine whether the design of each building structure conforms to building standards.
[0137] The building structure includes steel beams, frame beams, steel space frames, ring beams, structural columns, pipe piles, and walls.
[0138] It should be clear that the process of the processor executing the computer program in the above embodiments is consistent with the execution process of each step in the above method, as can be seen in the description above.
[0139] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0140] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0141] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.
Claims
1. An AI-based auxiliary design method for building structures, characterized in that, The method includes: A three-dimensional building model is obtained, and corresponding structured data is generated based on each building structure in the three-dimensional building model. Design relationship data between each building structure is extracted; the design relationship data carries floor marking information. The design relationship data is grouped according to the floor marking information to obtain multiple design relationship grouping data; By parsing the structured information of each building component in the design relationship grouping data, comparison information related to design standards is extracted based on the structured information according to preset rules; The comparison information that appears most frequently in the structured information is extracted first to determine the association relationship. The remaining comparison information is used for semantic recognition, and the association relationship between adjacent or related building components is determined based on the semantic recognition results. The correlation degree of the adjacent or related building components is assessed to determine the correlation strength. Clustering and grouping are performed based on the correlation strength; Multiple preliminary topological structure diagrams are formed based on the clustering and family division. The multiple preliminary topological structure diagrams are then weighted and fused to obtain the topological structure diagram, which is used to characterize the relationship between various building structures. The topology diagram is input into a preset deep learning neural network to determine whether the design of each building structure conforms to building standards.
2. The method according to claim 1, characterized in that, The step of generating corresponding structured data based on each building structure in the three-dimensional building model includes: Extract the name, floor, material, attribute parameters, type selection, and names of adjacent building structures as structured information for each building structure. A list of building components is generated based on the structural information of the building structure according to the floors. The data from the list of building components is filled into a structured data table to form the structured data, which is processed based on the floor level.
3. The method according to claim 1 or 2, characterized in that, The relationships mentioned include the relationship between frame beams and loads, the relationship between span structures and column grid spacing, the relationship between bathroom structural design, the relationship between span spatial structures and spatial grid structures, and the relationship between the cross-sectional height of frame beams and span.
4. An AI-based auxiliary design device for building structures, characterized in that, The device includes: The acquisition module is used to acquire a three-dimensional building model, generate corresponding structured data based on each building structure in the three-dimensional building model, and extract design relationship data between each building structure; the design relationship data carries floor marking information; A construction module is used to group the design relationship data according to the floor marking information to obtain multiple design relationship group data; by parsing the structured information of each building component in the design relationship group data, and extracting comparison information related to design standards based on the structured information according to preset rules; prioritizing the extraction of the comparison information with the highest frequency in the structured information to determine the association relationship, performing semantic recognition on the remaining comparison information, and determining the association relationship of adjacent or related building components based on the semantic recognition results; evaluating the association degree of the adjacent or related building components to obtain the association strength; performing clustering based on the association strength; forming multiple preliminary topology diagrams based on the clustering; and performing weighted fusion processing on the multiple preliminary topology diagrams to obtain the topology diagram, which is used to represent the association relationship between various building structures. The evaluation module is used to input the topology diagram into a preset deep learning neural network to determine whether the design of each building structure conforms to building standards.
5. The apparatus according to claim 4, characterized in that, The acquisition module includes: Structured units are used to extract the name, floor, material, attribute parameters, type selection, and names of adjacent building structures as structured information for each building structure. The list unit is used to generate a list of building components based on the structured information of the building structure according to the floors. The structured unit is used to fill the data in the list of building components into the structured data table to form the structured data, which is obtained by processing based on the floor.
6. The apparatus according to claim 4 or 5, characterized in that, The relationships mentioned include the relationship between frame beams and loads, the relationship between span structures and column grid spacing, the relationship between bathroom structural design, the relationship between span spatial structures and spatial grid structures, and the relationship between the cross-sectional height of frame beams and span.
7. 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 AI-based auxiliary design method for building structures as described in any one of claims 1 to 3.