Implementation method of generating substation BIM model using language based on AI model

Abstract

The present invention discloses an implementation method for generating a BIM model of a substation using a language based on an AI model, belongs to the field of BIM technology, and fills the gap in the conversion from language to model: for example, "the main transformer is located on the north side and is connected to the 110kV busbar through a circuit breaker" is directly converted into a compliant transformer BIM model; for the existing family library of the substation, the search function of the intelligent body is realized, and the search tool is accurately called according to the parameters input by the user to obtain the corresponding equipment model file; the 3D model file is output, and the file format is strictly ensured to be able to be imported into the substation 3D software such as Jinbo Chao for intercommunication (such as .obj, .fbx, .glb, etc.).

Description

Technical Field

[0001] The present invention relates to the field of BIM technology, and particularly to a method for implementing the generation of a substation BIM model using language based on an AI model. Background Art

[0002] In the past two decades, parametric models have been widely used in the power field. By constructing parameters, all data and attributes related to, such as power plants, substations, etc., can be associated and adjusted in real time. The parametric model has a more rational and rigorous operating environment, which can save manpower and material resources, facilitate problem troubleshooting and scheme adjustment, so it has an inevitable trend in the industry development. However, due to reasons such as long modeling time, expensive training costs for relevant software, and long training cycles, the development speed of parametric models has been relatively slow, resulting in low productivity.

[0003] The application of artificial intelligence modeling in the power field started relatively early. Some research institutions and enterprises have carried out research and practice on AI modeling in this regard and have released some experimental software. In terms of parametric modeling, many foreign software companies have developed mature software tools, such as Revit, which can achieve automated parametric modeling and design optimization. However, parametric modeling and AI modeling have not been associated.

[0004] With the continuous evolution of AI algorithms and the rapid growth of computing requirements, the era of artificial intelligence has quietly arrived. However, in the power field, the application of artificial intelligence is still in its infancy. Currently, the 3D results generated by AI generation model software on the market, such as those generated by Artefacts.Ai, are still slightly rough, without model information and normals, and cannot fully meet the requirements of BIM forward design. Most popular AI generation model software is mainly applied in the entertainment and media industries, and there is still a certain gap for the parametric models required in the power BIM field.

[0005] Based on this technology, an invention provides a more intuitive and fast model generation and search method, providing an end-to-end solution to automatically generate a substation BIM model that conforms to the power industry specifications through natural language input, significantly improving design efficiency and avoiding human errors. Summary of the Invention

[0006] The purpose of the present invention is to provide a method for implementing the generation of a substation BIM model using language based on an AI model in view of the above deficiencies, solving the problem that the 3D results generated by existing AI generation model software are still slightly rough, without model information and normals, and cannot fully meet the requirements of BIM forward design.

[0007] The present invention is implemented through the following solution:

[0008] An implementation method for generating a substation BIM model using language based on an AI model, comprising the following steps:

[0009] Step S1: Requirement analysis, professionally classifying the BIM model to be generated for the substation;

[0010] Step S2: Parameter determination, dynamically embedding industry specifications through a knowledge graph, and the AI automatically avoiding more than 90% of human errors;

[0011] Step S3: Selection of development technologies, manually evaluating the accuracy and efficiency of different AI algorithms in processing complex building structures, and whether the modeling tool supports the interoperability of mainstream substation software;

[0012] Step S4: System architecture development, developing a natural language parameter extraction tool based on a large model and applying natural language processing technology;

[0013] Step S5: Implementation and testing, generating a 3D model based on the input parameters and applying an optimization algorithm;

[0014] Step S6: Large model training, using the prepared data to train and optimize the established artificial intelligence model, extracting and transforming the features of the input data, optimizing and adjusting the model parameters, and performing performance evaluation and verification; improving the accuracy of the model through iterative training and optimization;

[0015] Step S7: Model output and verification, outputting the BIM model file, strictly ensuring that the file format meets the project requirements; using professional verification tools to comprehensively verify the accuracy and integrity of the model from the dimensions of geometric accuracy and topological structure, and for the problems found, using technical means such as reverse engineering for necessary adjustment and correction until the model meets the substation GIM model delivery standard.

[0016] In step S4, it specifically includes:

[0017] Step S41, input module; supporting text or voice input;

[0018] Step S42, NLP parsing engine: using a hybrid model that combines BERT and rule templates to solve the ambiguity of professional terms;

[0019] Step S43, establishing a knowledge graph library; storing existing family libraries; establishing dynamic rule examples;

[0020] Step S44, generation module; topology generation: generating an equipment connection diagram based on the GraphSAGE model to ensure the correctness of electrical logic; geometric generation: using an improved Stable Diffusion model, inputting equipment parameters and layout constraints, and outputting a three-dimensional point cloud;

[0021] Step S45, verification and output module; call the BIM platform API to generate an IFC file and automatically add attribute tags.

[0022] In step S43, the specific steps for establishing the knowledge graph library are as follows:

[0023] 1) Data collection and preprocessing: organize equipment, specification, and material data; 2) Knowledge graph modeling: define entities, relationships, and dynamic rules; 3) Knowledge graph storage: select a database and import data; 4) Knowledge graph query and verification: use SPARQL or Cypher for querying and perform real-time verification of design compliance.

[0024] In step S5, it specifically includes:

[0025] Step S51, input description;

[0026] Step S52, generation process; the structured data triggered by NLP parsing undergoes knowledge graph verification. If the busbar height is not specified, the default value is automatically filled; the diffusion model generates the equipment layout to ensure that the circuit breaker is correctly connected to the transformer and the busbar section;

[0027] Step S53, output result; BIM model: including the three-dimensional geometry and electrical properties of the transformer, circuit breaker, and busbar; automatically generate construction drawings: mark the equipment coordinates, installation instructions, and safety distance detection results.

[0028] In step S6, it specifically includes:

[0029] Step S61, data collection; collect existing specification data;

[0030] Step S62, data augmentation; it includes text data augmentation and image / 3D data augmentation: text data augmentation includes synonym replacement and sentence pattern rewriting; image / 3D data augmentation includes rotation, scaling, and noise injection;

[0031] Step S63, data storage and management; use MySQL / PostgreSQL for structured data; use MongoDB / MinIO for unstructured data;

[0032] Step S64, distributed training; data parallelism: split data across multiple GPUs and synchronize gradients; model parallelism: split model layers across different devices; pipeline parallelism: perform stage-by-stage calculations by layer;

[0033] Step S65, optimization of the training process.

[0034] In step S65, it specifically includes: reducing video memory occupancy: mixed-precision training, gradient checkpointing, model parallelism; improving throughput: data parallelism, data loading optimization, gradient accumulation; distributed hyperparameter tuning: using Horovod and DeepSpeed for distributed training, combined with Optuna and Ray Tune for hyperparameter search.

[0035] In step S7, it specifically includes:

[0036] Step S71, BIM model file output

[0037] Step S72, geometric accuracy verification;

[0038] Step S73, topological structure verification;

[0039] Step S74, problem correction;

[0040] Step S75, reverse engineering technology derivation;

[0041] Step S76, AI-assisted correction;

[0042] Step S77, manual review.

[0043] Step S71, BIM model file output specifically means:

[0044] 1) Format selection and specification adaptation: Generate a BIM model file in a standard format to ensure compatibility with downstream tools; 2) File integrity verification to confirm the existence of key components and the completeness of model information.

[0045] Step S72, geometric accuracy verification;

[0046] Specifically: Ensure that the generated BIM model meets the design requirements and industry specifications in terms of geometric attributes such as dimensions, positions, and coordinates; Generate a PDF report containing the following: 1) List of non-compliant devices; 2) Geometric error heat map;

[0047] Step S73, topological structure verification;

[0048] Specifically: Topological structure verification; Ensure that the device connection relationships conform to the design logic and industry standards; 1) Electrical connection logic check, including: checking whether the circuit breakers form a closed loop; verifying whether the rated current of the circuit breakers matches the bus load, etc.; 2) Parent-child structure verification; including: checking whether the bus ducts belong to the correct distribution cabinets; verifying whether the child devices inherit the attributes of the parent devices;

[0049] Step S74, problem correction;

[0050] Specific; Problem correction; Classify problems according to verification results and prioritize critical errors to ensure model compliance; Include: automatic correction and manual intervention.

[0051] Step S75, Reverse engineering technology derivation;

[0052] Specifically, reverse engineering technology derivation; Generate key dimensions of the model through measurement, reverse infer design parameters, and optimize the original input; Input the reverse inferred parameters into the generation model and iteratively optimize the design;

[0053] Step S76, AI-assisted correction;

[0054] Specifically, AI-assisted correction; Train a reinforcement learning model to optimize the equipment layout and minimize the number of violations; Extract the equipment dimensions of the generation model; Then match the measured values with the parameters in the equipment library and recommend the closest model;

[0055] Step S77, Manual review;

[0056] The specific steps are as follows: 1) Visual inspection: Check the equipment layout and connection relationship through a 3D view; 2) Parameter verification: Compare whether the model property table is consistent with the design document; 3) Conflict review: Use the Clash Detection tool to recheck structural conflicts.

[0057] In summary, due to the adoption of the above technical solutions, the beneficial effects of the present invention are:

[0058] 1), Traditional substation BIM modeling relies on manual operation by engineers, and it is necessary to repeatedly adjust the equipment layout, parameters and connection relationship, which takes up to several weeks; The accuracy depends on manual verification and requires manual secondary correction, increasing the risk of errors. In view of this deficiency, the present invention provides a new method for solving substation BIM modeling, filling the gap in the conversion from language to model: For example, "The main transformer is located in the north and is connected to the 110kV bus through a circuit breaker" is directly converted into a compliant transformer BIM model; For the existing family library in the substation, the search function of the intelligent agent is realized. According to the parameters input by the user, the search tool is accurately called to obtain the corresponding equipment model file; Output a 3D model file, and strictly ensure that the file format can be imported into substation 3D software such as BCT to be interoperable (such as.obj,.fbx,.glb, etc.).

[0059] 2) Efficiency improvement: The design cycle is shortened from 2-4 weeks of traditional manual modeling to within 10 minutes.

[0060] 3) Error rate reduction: More than 90% of human oversights are avoided through the rule engine, such as insufficient safety distance.

[0061] 4) Enhanced compatibility: Supports standard formats such as.obj,.fbx,.glb, etc., and can be directly imported into substation BIM software platforms (such as Revit, Bochao) and power grid simulation systems. Description of the Drawings

[0062] Figure 1 is the overall flowchart of the present invention;

[0063] Figure 2 is the flowchart of step S4 in the present invention;

[0064] Figure 3 is the flowchart of step S5 in the present invention;

[0065] Figure 4 is the flowchart of step S6 in the present invention;

[0066] Figure 5 is the flowchart of step S7 in the present invention. Detailed Embodiments

[0067] All features disclosed in this specification, or all steps in the disclosed methods or processes, except for mutually exclusive features and / or steps, can be combined in any way.

[0068] Any feature disclosed in this specification (including any additional claims, abstract) can be replaced by other equivalent or similar-purpose alternative features, unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only an example of a series of equivalent or similar features.

[0069] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by terms such as "upper", "lower", "left", "right", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a predetermined orientation, be constructed and operated in a predetermined orientation, and therefore should not be construed as a limitation of the present invention.

[0070] In addition, terms such as "first", "second", etc. are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first", "second", etc. may explicitly or implicitly include one or more of such features.

[0071] Embodiment 1

[0072] As Figures 1 to 3 shown, the present invention provides a technical solution:

[0073] An implementation method for generating a substation BIM model using language based on an AI model, which at least includes the following steps:

[0074] Step S1: Requirement analysis. Professionally classify the BIM models to be generated for the substation, such as classification into primary electrical, secondary electrical, civil engineering, structure, HVAC, water supply and drainage, etc.

[0075] Step S2: Parameter determination. Dynamically embed industry specifications (such as GB / T 50059) through the knowledge graph, and the AI automatically avoids more than 90% of human errors. The method of automatic avoidance is as follows:

[0076] 1) Analyze the rules (such as safety distances and equipment compatibility) in standard documents such as GB / T 50059 into structured data;

[0077] 2) During the model generation process, call the knowledge graph rule library to verify the design compliance in real time. For example, when generating the main transformer, automatically verify its safety distance from the enclosure (≥5 meters). In the existing technology, the model is usually verified manually, which is very prone to errors. The AI can automatically avoid more than 90% of human errors.

[0078] Step S3: Selection of development technologies. Manually evaluate the accuracy and efficiency of different AI algorithms in processing complex building structures, and whether the modeling tool supports the interoperability of mainstream substation software;

[0079] Step S4: System architecture development. Develop a natural language parameter extraction tool based on the large model, and use advanced natural language processing technologies, specifically, such as the Transformer architecture, etc., to achieve efficient communication and interaction between humans and machines based on natural language;

[0080] Step S5: Implementation and testing. Generate a 3D model based on the input parameters, and use optimization algorithms to ensure that the model has good program performance while maintaining realism and can run smoothly in different hardware environments;

[0081] The optimization algorithms can be mesh simplification, texture compression, rendering optimization or performance testing:

[0082] Mesh simplification can reduce the number of patches and lower the rendering load; texture compression can reduce the texture data volume and optimize the GPU memory occupancy; rendering optimization uses LOD, instance rendering and GPU acceleration technologies to improve the rendering performance; performance testing can test and optimize in different hardware environments to ensure smooth operation;

[0083] Step S6: Large model training. Use the prepared data, which is specifically the data in Step S1 and the data generated in S5, to train and optimize the established artificial intelligence model, extract and transform the features of the input data, optimize and adjust the model parameters, and conduct performance evaluation and verification; Through iterative training and optimization, improve the accuracy of the model;

[0084] Step S7: Model output and verification. Output the BIM model file, strictly ensuring that the file format meets the project requirements (such as.obj,.fbx,.glb, etc.); Use professional verification tools to comprehensively verify the accuracy and integrity of the model from multiple dimensions such as geometric accuracy and topological structure. For the problems found, use technical means such as reverse engineering for necessary adjustments and corrections until the model meets the delivery standard of the substation GIM model.

[0085] Based on the above content, the present invention mainly provides an implementation method for generating a substation BIM model using language based on an AI model, including: developing AI software, which can automatically generate a substation BIM model that meets the specifications of the power industry through natural language input.

[0086] AI search model. Use the search interface of this software to search for existing models in the family library by entering keywords.

[0087] Output of model format. Export a 3D model in a common format through a software interface to achieve interoperability with common 3D software for substations.

[0088] This solution provides an automatic generation of a substation BIM model that meets the specifications of the power industry through natural language input; uses the AI search method to enter keywords to search for existing models in the family library; and the output model format achieves interoperability with common 3D software for substations.

[0089] Embodiment 2

[0090] As Figures 1 to 3 shown, the present invention provides a technical solution:

[0091] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. For steps S1, S2, and S3, which belong to the architecture design and are not within the scope of the invention, the present invention only provides the steps and will not describe them in detail. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art belong to the scope of protection of the present invention.

[0092] An implementation method for generating a substation BIM model using language based on an AI model, the method comprising the following steps:

[0093] S1. Requirement analysis;

[0094] S2. Parameter determination;

[0095] S3. Selection of development technology;

[0096] S4. System architecture development;

[0097] S5. Implement the test;

[0098] S6. Train the large model;

[0099] S7. Model output and verification.

[0100] Please refer to Figure 1 , which is the flowchart of the implementation method for generating a substation BIM model using language based on an AI model provided by an embodiment of the present invention. In this embodiment, the client is developed using the C# language in combination with the WPF (Windows Presentation Foundation) framework; the backend is developed using the Python language, in combination with the FastAPI framework and Blender 3D software. The introduction of this embodiment is carried out on the above premise;

[0101] Step S1, Requirement analysis, professionally classify the BIM model to be generated for the substation, such as primary electrical, secondary electrical, civil engineering, structure, HVAC, water supply and drainage, etc.

[0102] Step S2, Parameter determination, for example: dynamically embed industry specifications through a knowledge graph, such as GB / T 50059, and the AI automatically avoids more than 90% of human errors.

[0103] Step S3, Selection of development technologies, evaluate the accuracy and efficiency of different AI algorithms in processing complex building structures, and whether the modeling tool supports the interoperability of mainstream substation software.

[0104] Step S4, System architecture development, develop a natural language parameter extraction tool based on a large model, and use advanced natural language processing technologies, such as the Transformer architecture, etc., to achieve efficient communication and interaction between humans and machines based on natural language.

[0105] Specifically include:

[0106] Step S41, Input module. Support text or voice input, such as "110kV indoor substation, main transformer capacity 50MVA, ≥8 meters from the west wall";

[0107] Step S42, NLP parsing engine: Use a hybrid model that combines BERT and rule templates to solve the ambiguity of professional terms, such as "busbar" referring to copper bars or cables; output structured data with probability confidence, and request the user to clarify when the confidence is low;

[0108] In this solution, "NLP" is the abbreviation of Natural Language Processing, which is called natural language processing in Chinese. It is a branch of artificial intelligence (AI) and computer science, focusing on enabling computers to understand, interpret, and generate human language. In this solution, "BERT" (Bidirectional Encoder Representation from Transformers) is a pre-trained model proposed by Google AI Research Institute in October 2018.

[0109] Step S43: Establish a knowledge graph library; store existing family libraries, such as: equipment library (transformer, circuit breaker, etc.), specification library (national standards, industry standards), material library (dimensions, electrical parameters); establish dynamic rule examples, such as: IF equipment type = circuit breaker THEN the connection object must include busbars AND transformers.

[0110] This solution establishes dynamic rules, which can transform the organic structure of equipment or specifications, such as parameters like the level and size of equipment, into a database to form specifications. In this way, the generated model conforms to the specifications.

[0111] The specific steps are as follows:

[0112] 1) Data collection and preprocessing: Organize equipment, specification, and material data; 2) Knowledge graph modeling: Define entities, relationships, and dynamic rules; 3) Knowledge graph storage: Select a database and import data; 4) Knowledge graph query and verification: Use SPARQL or Cypher queries to verify the compliance of the design in real time;

[0113] Step S44: Generation module;

[0114] Topology generation: Generate an equipment connection diagram based on the GraphSAGE model to ensure the correctness of electrical logic;

[0115] Geometry generation: Adopt an improved Stable Diffusion model, input equipment parameters and layout constraints, and output a three-dimensional point cloud;

[0116] Step S45: Verification and output module; Call the BIM platform API to generate an IFC file, automatically add attribute tags, such as: equipment ID, operation and maintenance information; output a violation report and correction suggestions, such as: "The distance between the main transformer and the fence is insufficient. It is recommended to move it 2.1 meters to the east";

[0117] Step S5: Implement tests, generate a 3D model based on the input parameters, and use optimization algorithms, such as mesh simplification, texture compression, etc. to ensure that the model has good program performance while maintaining realism and can run smoothly in different hardware environments;

[0118] Specifically, it includes:

[0119] Step S51, input description; for example: "35kV substation, including two dry-type transformers (capacity 10MVA), located on the east side of the factory building, with a spacing of ≥3 meters, connected to the single bus section through a vacuum circuit breaker";

[0120] Step S52, generation process; the structured data triggered by NLP parsing is used to verify the knowledge graph. If the bus height is not specified, the default value of 4.5 meters is automatically filled; the diffusion model generates the equipment layout, and the GNN ensures the correct connection of the circuit breaker to the transformer and the bus section; in this solution, the GNN, that is, the graph neural network, is a deep learning technology for processing graph data;

[0121] Step S53, output result; BIM model (IFC format): including the three-dimensional geometry and electrical properties of the transformer, circuit breaker, and bus; automatically generate construction drawings (PDF): marking the equipment coordinates, installation instructions, and safety distance detection results;

[0122] Step S6, large model training; use the prepared data, which is specifically the data in Step S1 and the data generated after S5, to train and optimize the established artificial intelligence model; through iterative training and optimization, improve the accuracy of the model;

[0123] Specifically

[0124] Step S61, data collection; collect existing standard data, such as power industry standard data and equipment parameter libraries;

[0125] Step S62, data augmentation; text data: synonym replacement, sentence pattern rewriting (applicable to natural language input); image / 3D data: rotation, scaling, noise injection (applicable to BIM model generation).

[0126] Step S63, data storage and management; structured data uses MySQL / PostgreSQL (equipment parameters, design specifications); unstructured data uses MongoDB / MinIO (text descriptions, BIM files).

[0127] Step S64, distributed training; data parallelism: split the data to multiple GPUs and synchronize gradients; model parallelism: split the model layers to different devices; pipeline parallelism: calculate in stages by layer;

[0128] Step S65, training process optimization. For example: reduce video memory occupancy, improve throughput; distributed hyperparameter tuning framework, etc.;

[0129] Specifically:

[0130] Reduce VRAM usage: Mixed-precision training, gradient checkpointing, model parallelism. Improve throughput: Data parallelism, data loading optimization, gradient accumulation. Distributed hyperparameter tuning: Use Horovod and DeepSpeed for distributed training, combined with Optuna and Ray Tune for hyperparameter search.

[0131] Step S7, model output and verification; Output the BIM model file, strictly ensuring that the file format meets the project requirements (such as.obj,.fbx,.glb, etc.); Use professional verification tools to comprehensively verify the accuracy and integrity of the model from multiple dimensions such as geometric accuracy and topological structure; For the problems found, use technical means such as reverse engineering for necessary adjustments and corrections until the model meets the substation GIM model delivery standard.

[0132] Specifically:

[0133] Step S71, BIM model file output; including: 1) Format selection and specification adaptation: Generate a BIM model file in a standard format (such as.obj,.fbx,.glb) to ensure compatibility with downstream tools; 2) File integrity verification, such as: whether key components exist and whether model information is complete.

[0134] Step S72, geometric accuracy verification; such as: dimensions and positions, object coordinates.

[0135] Specifically: Geometric accuracy verification; Ensure that the generated BIM model meets the design requirements and industry specifications in terms of geometric attributes such as dimensions, positions, and coordinates; Generate a PDF report containing the following: 1) A list of non-compliant devices (such as coordinate offsets, dimension mismatches); 2) A heat map of geometric errors (visually showing problem areas).

[0136] Step S73, topological structure verification. Such as: electrical connection logic, parent-child structure.

[0137] Specifically: Topological structure verification; Ensure that the device connection relationships (such as electrical topology, hierarchical structure) conform to the design logic and industry standards; 1) Electrical connection logic check, including: checking whether the circuit breakers form a closed loop (violating the single-bus sectionalized design); verifying whether the rated current of the circuit breakers matches the bus load, etc.; 2) Parent-child structure verification; including: checking whether the bus ducts belong to the correct distribution cabinets; verifying whether the child devices inherit the parent attributes (such as voltage levels).

[0138] Step S74, problem correction. Such as: problem classification and priority ranking.

[0139] Specifically; Problem correction; Classify problems according to the verification results and prioritize the handling of critical errors to ensure model compliance; including: automatic correction and manual intervention.

[0140] Step S75, reverse engineering technology derivation; for example, by measuring the key dimensions of the generated model to reverse-derive the design parameters.

[0141] Specifically, for reverse engineering technology derivation: by measuring the key dimensions of the generated model to reverse-derive the design parameters, optimizing the original input; inputting the reverse-derived parameters into the generated model for iterative design optimization.

[0142] Step S76, AI-assisted correction. Train a reinforcement learning model to optimize the layout and minimize the number of violations.

[0143] Specifically, for AI-assisted correction: train a reinforcement learning model to optimize the equipment layout and minimize the number of violations; for example, extract the equipment dimensions of the generated model (such as the length, width, and height of the transformer); then match the measured values with the parameters in the equipment library and recommend the closest model.

[0144] Step S77, manual review. Reviewed by BIM engineers.

[0145] Specifically, for manual review: reviewed by BIM engineers; the steps are as follows: 1) Visual inspection: Check the equipment layout and connection relationships through 3D views; 2) Parameter verification: Compare whether the model property table is consistent with the design document; 3) Conflict recheck: Use the Clash Detection tool to recheck for structural conflicts.

[0146] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for generating a substation BIM model using a language based on an AI model, characterized in that: The following steps are involved: Step S1: Demand analysis, professional classification of the BIM models that need to be generated for the substation; Step S2: Parameter determination, dynamically embedding industry specifications through knowledge graphs, and AI automatically avoids more than 90% of human errors; Step S3: Development technology selection, manual evaluation of the accuracy and efficiency of different AI algorithms in processing complex building structures, and whether the modeling tools support interoperability with mainstream substation software; Step S4: System architecture development, developing a natural language parameter extraction tool based on a large model, using natural language processing technology; In step S4, it specifically includes: Step S41, input module: supports text or voice input; Step S42, NLP parsing engine: using a hybrid model that integrates BERT and rule templates to resolve professional terminology ambiguity; Step S43, establishing a knowledge graph library; storing existing family libraries; establishing dynamic rule examples; Step S44, generating modules; topology generation: generating a device connection diagram based on the GraphSAGE model to ensure the correctness of electrical logic; geometry generation: using Stable Diffusion to improve the model, input device parameters and layout constraints, and output a three-dimensional point cloud; Step S45, verification and output module: call BIM platform API to generate IFC file and automatically add attribute tags Step S5: Conducting testing, generating a 3D model based on the input parameters, and applying an optimization algorithm; In step S5, it specifically includes: Step S51, input description; Step S52, generation process; after NLP parsing, the structured data triggers the knowledge graph verification. If the bus height is not specified, the default value is automatically filled in; the diffusion model generates the equipment layout to ensure that the circuit breaker is correctly connected to the transformer and bus segment; Step S53, output results; BIM model: including the three-dimensional geometry and electrical properties of transformers, circuit breakers, and busbars; automatically generate construction drawings: annotate equipment coordinates, installation instructions, and safety distance detection results Step S6: large model training, using the prepared data to train and tune the established artificial intelligence model, feature extraction and conversion of input data, optimization and adjustment of model parameters, and performance evaluation and verification; improve the accuracy of the model through iterative training and tuning; Step S7: Model output and verification, output BIM model files, and strictly ensure that the file format meets the project requirements; use professional verification tools to comprehensively verify the accuracy and completeness of the model from the dimensions of geometric accuracy and topological structure, and use reverse engineering technology to make necessary adjustments and corrections to the problems found until the model meets the substation GIM model delivery standards.

2. According to claim 1, a method for generating a substation BIM model using a language based on an AI model is characterized by: In step S43, the specific steps of establishing a knowledge graph library are as follows: 1) Data collection and preprocessing: organize equipment, specifications, and material data; 2) Knowledge graph modeling: define entities, relationships, and dynamic rules; 3) Knowledge graph storage: select a database and import data; 4) Knowledge graph query and verification: use SPARQL or Cypher queries to verify design compliance in real time.

3. The method for generating a substation BIM model using a language based on an AI model according to claim 1 or 2, characterized in that: In step S6, it specifically includes: Step S61, data collection: collecting existing normative data; Step S62, data enhancement; which includes text data enhancement and image / 3D data enhancement: text data enhancement includes synonym replacement and sentence rewriting; image / 3D data enhancement includes rotation, scaling, and noise injection; Step S63, data storage and management; structured data uses MySQL / PostgreSQL; unstructured data uses MongoDB / MinIO; Step S64, distributed training; data parallelism: split data to multiple GPUs, synchronize gradients; model parallelism: split model layers to different devices; pipeline parallelism: calculate in stages by layer; Step S65, training process optimization.

4. According to claim 3, a method for generating a substation BIM model using a language based on an AI model is characterized in that: In step S65, specifically including: reducing video memory usage: mixed precision training, gradient checkpoints, model parallelism; Improve throughput: data parallelism, data loading optimization, and gradient accumulation; Distributed hyperparameter tuning: Use Horovod and DeepSpeed ​​for distributed training, and combine Optuna and Ray Tune for hyperparameter search.

5. The method for generating a substation BIM model using a language based on an AI model according to claim 1 or 2, characterized in that: In step S7, it specifically includes: Step S71, BIM model file output; Step S72, geometric accuracy verification; Step S73, topology verification; Step S74, problem correction; Step S75, reverse engineering technology derivation; Step S76, AI-assisted correction; Step S77, manual review.

6. According to claim 5, a method for generating a substation BIM model using a language based on an AI model is characterized by: Step S71, BIM model file output is specifically as follows: 1) Format selection and specification adaptation: Generate BIM model files that conform to the standard format to ensure compatibility with downstream tools; 2) File integrity verification to confirm whether key components exist and whether the model information is complete.

7. The method for generating a substation BIM model using a language based on an AI model according to claim 6 is characterized in that: Step S72, geometric accuracy verification; Specifically: ensure that the geometric properties of the generated BIM model in terms of size, position, and coordinates meet the design requirements and industry specifications; generate a PDF report containing the following: 1) a list of illegal equipment; 2) a heat map of geometric errors; Step S73, topology verification; Specifically: topology verification; ensuring that the equipment connection relationship complies with the design logic and industry standards; 1) electrical connection logic check, including: checking whether the circuit breaker forms a closed loop; verifying whether the rated current of the circuit breaker matches the bus load; 2) parent-child structure verification; including: checking whether the bus duct belongs to the correct distribution cabinet; verifying whether the child device inherits the parent attribute; Step S74, problem correction; Specific; problem correction; classify problems based on validation results and prioritize critical errors to ensure model compliance; includes: automatic correction and manual intervention.

8. The method for generating a substation BIM model using a language based on an AI model according to claim 7 is characterized in that: Step S75, reverse engineering technology derivation; Specifically, reverse engineering technology is used to derive; by measuring the key dimensions of the generated model, the design parameters are inferred and the original input is optimized; the inferred parameters are input into the generated model and the design is optimized iteratively; Step S76, AI-assisted correction; Specifically, AI-assisted correction; Train a reinforcement learning model to optimize device layout and minimize the number of violations; extract device dimensions for the generated model; then match the measurements with the device library parameters and recommend the closest model; Step S77, manual review; The specific steps are as follows: 1) Visual inspection: check the equipment layout and connection relationship through 3D view; 2) Parameter verification: compare whether the model attribute table is consistent with the design document; 3) Conflict review: use the Clash Detection tool to review structural conflicts.

Images