A bridge design method, device, equipment and medium

Abstract

The present invention provides a bridge design method, device, equipment and medium, and relates to the technical field of bridge design. The method comprises: obtaining basic data of a bridge model, obtaining a natural language instruction input by a user, obtaining a hierarchical instruction, and when the hierarchical instruction is a query instruction, performing a storage check on the hierarchical instruction; when the hierarchical instruction is an operation instruction, performing a function check on the hierarchical instruction; if the storage check succeeds, sending the hierarchical instruction to a data storage and management module for query, obtaining a query result, obtaining an optimized query result and sending the optimized query result to a user based on an information interaction module; if the function check succeeds, obtaining an initial BIM model and an operation result, obtaining an optimized operation result and sending the optimized operation result to a user based on an information interaction module, thereby reducing the workload and error rate of manual operation.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge design, and in particular to a bridge design method, device, equipment and medium. Background Art

[0002] Building Information Modeling (BIM) is a digital method that covers the entire process of engineering project design, construction, and operation management. In recent decades, with the development of BIM technology, the application of BIM in engineering project design has become more and more common, and has greatly changed the design operations and processes of the engineering industry. Take bridge design as an example. In the traditional bridge design process, designers need to manually perform a lot of modeling, calculation, optimization, and drawing work based on a two-dimensional environment, which is inefficient, time-consuming, and error-prone. BIM three-dimensional bridge design can directly use BIM technology to create a three-dimensional information model of the bridge, and carry out in-depth design based on a visual three-dimensional information model, and use the three-dimensional information model to realize the integration of multi-stage and multi-dimensional information in design, construction, and operation. The development of the BIM bridge design system has significantly improved the digitalization and informatization level of bridge design.

[0003] However, the current BIM bridge design system still has shortcomings in terms of intelligence and automation. Although the current BIM bridge design has achieved a leap from two-dimensional design to three-dimensional design, and from scattered design information to information integration, its human-computer interaction is still not friendly enough, and its design process still requires a lot of manual operation. Designers often need to have rich professional knowledge and special training on the BIM technology and operating system used in order to enable designers to operate the BIM design system proficiently. This has brought certain difficulties and challenges to the rapid promotion of the BIM bridge design system and the popularization of the BIM bridge design method.

[0004] In addition, since the BIM three-dimensional information model brings together a large amount of data information from the design to construction, operation and maintenance stages of the model, the data model structure has become increasingly complex, and the data information has become increasingly massive, which has brought great challenges to BIM information retrieval. Users not only need to master complex system interface operations, but also need to be familiar with the professional data structure, attribute terms, and query methods of the BIM model, and have in-depth professional knowledge before they can find the corresponding data information in the large BIM model database. This traditional BIM model information retrieval method has hindered the realization of large-scale, fast, and accurate information retrieval in the design, construction, and operation and maintenance stages. Summary of the invention

[0005] In view of the above technical problems, the technical solution adopted by the present invention is: a bridge design method, which is implemented based on BIM and GPT and includes the following steps:

[0006] Obtaining basic data of the bridge model, and sending the basic data of the bridge model to the data storage and management module for storage;

[0007] Based on the information interaction module, the natural language instructions input by the user are obtained, and the natural language instructions are sent to the GPT intelligent processing module for parsing to obtain the classification instructions;

[0008] When the hierarchical instruction is a query instruction, a storage check is performed on the hierarchical instruction; when the hierarchical instruction is an operation instruction, a function check is performed on the hierarchical instruction, wherein the operation instruction includes a new instruction, a modification instruction, and a deletion instruction;

[0009] If the storage verification is successful, the classification instruction is sent to the data storage and management module for query, the query result is obtained, and the query result is sent to the GPT intelligent processing module for optimization, the optimized query result is obtained and the optimized query result is sent to the user based on the information interaction module, and the process ends;

[0010] If the function verification is successful, the hierarchical instruction is sent to the BIM bridge design and application module. The BIM bridge design and application module calls the function API based on the hierarchical instruction to perform operations and obtain the initial BIM model and operation results.

[0011] Sending the model information corresponding to the initial BIM model to the data storage and management module for storage;

[0012] The operation results are sent to the GPT intelligent processing module for optimization, the optimized operation results are obtained and sent to the user based on the information interaction module.

[0013] A bridge design device, comprising:

[0014] A basic data acquisition module is used to acquire basic data of the bridge model and send the basic data of the bridge model to the data storage and management module for storage;

[0015] A hierarchical instruction acquisition module is used to acquire natural language instructions input by the user based on the information interaction module, and send the natural language instructions to the GPT intelligent processing module for parsing to acquire hierarchical instructions;

[0016] A classification module, used for performing storage verification on the hierarchical instruction when the hierarchical instruction is a query instruction; and performing function verification on the hierarchical instruction when the hierarchical instruction is an operation instruction, wherein the operation instruction includes a new instruction, a modification instruction, and a deletion instruction;

[0017] The query module is used to send the classification instruction to the data storage and management module for query if the storage verification is successful, obtain the query result, and send the query result to the GPT intelligent processing module for optimization, obtain the optimized query result and send the optimized query result to the user based on the information interaction module, and end;

[0018] The operation module is used to send the hierarchical instruction to the BIM bridge design and application module if the function verification is successful. The BIM bridge design and application module calls the function API based on the hierarchical instruction to perform the operation and obtain the initial BIM model and the operation result;

[0019] A storage module is used to send model information corresponding to the initial BIM model to the data storage and management module for storage;

[0020] The optimization module is used to send the operation results to the GPT intelligent processing module for optimization, obtain the optimized operation results and send the optimized operation results to the user based on the information interaction module.

[0021] According to another aspect of the present invention, a non-transitory computer-readable storage medium is provided, in which at least one instruction or at least one program is stored, and the at least one instruction or the at least one program is loaded and executed by a processor to implement the aforementioned method.

[0022] According to yet another aspect of the present invention, an electronic device is provided, comprising a processor and the aforementioned non-transitory computer-readable storage medium.

[0023] The present invention has at least the following beneficial effects: in summary, the basic data of the bridge model is obtained, the natural language instructions input by the user are obtained based on the information interaction module, and the natural language instructions are sent to the GPT intelligent processing module for parsing, and the hierarchical instructions are obtained. When the hierarchical instructions are query instructions, the hierarchical instructions are stored and verified; when the hierarchical instructions are operation instructions, the hierarchical instructions are functionally verified. If the storage verification is successful, the hierarchical instructions are sent to the data storage and management module for query, the query results are obtained, and the query results are sent to the GPT intelligent processing module for optimization, the optimized query results are obtained, and the optimized query results are sent to the user based on the information interaction module; if the functional verification is successful, the hierarchical instructions are sent to the BIM bridge Beam design and application module, BIM bridge design and application module operates based on hierarchical instruction call function API, obtains the initial BIM model and operation results, sends the model information corresponding to the initial BIM model to the data storage and management module for storage, sends the operation results to the GPT intelligent processing module for optimization, obtains the optimized operation results and sends the optimized operation results to the user based on the information interaction module. The present invention integrates BIM and GPT technologies, so that users can directly query BIM-related information through natural language text, and interpret user query information, summarize retrieval results and answer BIM-related questions through GPT functions, thereby reducing the workload and error rate of manual operations and lowering the requirements for the use of queries and other functions. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of 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 work.

[0025] Figure 1 A flow chart of a bridge design method provided by an embodiment of the present invention;

[0026] Figure 2 A schematic structural diagram of a bridge design device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0028] Embodiment 1,

[0029] Embodiment 1 of the present invention provides a bridge design method, which is implemented based on BIM and GPT. Figure 1 As shown, the method comprises the following steps:

[0030] S100, acquiring basic data of a bridge model, and sending the basic data of the bridge model to a data storage and management module for storage.

[0031] Specifically, the basic data of the bridge model includes bridge route information, bridge edge information, terrain information, and geological information. Furthermore, the basic data of the bridge model also includes: bridge-road information, bridge-tunnel boundary information, etc. provided by roadbed professionals and tunnel professionals.

[0032] Specifically, the formats of the bridge model basic data include txt text format, xlsx text format, dwg drawing file format, dxf drawing file format, dgn BIM model format, fbx BIM model format, rvt BIM model format, and ifc BIM model format.

[0033] Furthermore, before S100, it also includes: entering a preset intelligent bridge design system operating environment.

[0034] S200, based on the information interaction module, obtains the natural language instructions input by the user, and sends the natural language instructions to the GPT intelligent processing module for parsing to obtain the classification instructions.

[0035] Specifically, a preset field list set is obtained, and the natural language instructions are parsed using the GPT intelligent processing module based on the preset field list set to obtain hierarchical instructions. In one embodiment of the present invention, the preset fields include: preset execution fields, preset first-level fields, preset attribute field lists, and preset operation fields. The preset attribute field list includes: center pile number, span, and structure type.

[0036] In an exemplary description of the present invention, the natural language instruction is: build a new 3*30mT beam bridge with a center pile number of K5+135.0, and the GPT intelligent processing module is used for parsing to obtain the parsing result:

[0037] Preset Execution Field: New

[0038] Preset first level field: Bridge

[0039] Preset attribute fields: [Center pile number: K5+135.0], [Span: 3*30m], [Structure type: T beam]

[0040] Default operation field: empty

[0041] Based on the analysis results, a combination is performed to obtain hierarchical instructions.

[0042] In another exemplary description of the present invention, the natural language instruction is: query the average pile length of Bridge No. 1, and use the GPT intelligent processing module to parse and obtain the parsing result:

[0043] Preset execution field: Calculation

[0044] Preset first level field: Bridge No. 1

[0045] Preset attribute fields: [Center pile number: empty], [Span: empty], [Structure type: empty]

[0046] Default operation field: Mean

[0047] Based on the analysis results, a combination is performed to obtain hierarchical instructions.

[0048] S300, when the hierarchical instruction is a query instruction, perform storage verification on the hierarchical instruction; when the hierarchical instruction is an operation instruction, perform function verification on the hierarchical instruction, wherein the operation instruction includes a new creation instruction, a modification instruction, and a deletion instruction.

[0049] Specifically, the storage verification of the hierarchical instructions can be understood as verifying whether the hierarchical instructions are sufficient, whether they can close a query process, and whether the data storage and management module stores relevant data information.

[0050] Specifically, functional verification of hierarchical instructions can be understood as verifying whether the hierarchical instructions are sufficient, whether they can close an operation process, and whether the BIM bridge design and application module has the functional API related to the operation instruction.

[0051] S400, if the storage verification is successful, send the grading instruction to the data storage and management module for query, obtain the query result, and send the query result to the GPT intelligent processing module for optimization, obtain the optimized query result and send the optimized query result to the user based on the information interaction module, and end.

[0052] Specifically, the data storage and management module queries and processes in sequence according to the hierarchical instructions. In an exemplary description of the present invention, the length information of all pile foundations of Bridge No. 1 is queried and obtained, and then the obtained pile foundation length information is averaged to obtain 36.4m as the query result; the GPT intelligent processing module, with its powerful natural language processing and context understanding capabilities, organizes the result information of the instruction completion into context-related and coherent natural language text, "The average pile length of the bridge is 36.4m", and sends it to the user based on the information interaction module.

[0053] S500: If the function verification is successful, the hierarchical instruction is sent to the BIM bridge design and application module. The BIM bridge design and application module calls the function API based on the hierarchical instruction to perform operations and obtain the initial BIM model and operation results.

[0054] Specifically, the BIM bridge design and application module calls the relevant functional APIs in this module in sequence according to the hierarchical instructions to complete the corresponding operations. In one embodiment of the present invention, the functional APIs such as "bridge span arrangement", "obtaining upper and lower structure library templates", and "bridge 3D model creation" are called in sequence to create a 3*30m bridge 3D information model with a T-beam upper structure at the center pile number K5+135.0 as the initial BIM model; the operation result is "command completed".

[0055] Specifically, the model information corresponding to the initial BIM model includes: the model information corresponding to the initial BIM model includes: position information, attribute information, wherein the attribute information includes size information, material information, etc.

[0056] S600: Send model information corresponding to the initial BIM model to the data storage and management module for storage.

[0057] S700, sending the operation result to the GPT intelligent processing module for optimization, obtaining the optimized operation result and sending the optimized operation result to the user based on the information interaction module.

[0058] In an exemplary description of the present invention, the optimized operation result is "the instruction is completed and the bridge is created successfully".

[0059] In summary, the basic data of the bridge model is obtained, the natural language instructions input by the user are obtained based on the information interaction module, and the natural language instructions are sent to the GPT intelligent processing module for parsing to obtain the hierarchical instructions. When the hierarchical instructions are query instructions, the hierarchical instructions are stored and verified; when the hierarchical instructions are operation instructions, the hierarchical instructions are functionally verified. If the storage verification is successful, the hierarchical instructions are sent to the data storage and management module for query, and the query results are obtained. The query results are sent to the GPT intelligent processing module for optimization, and the optimized query results are obtained and sent to the user based on the information interaction module; if the functional verification is successful, the hierarchical instructions are sent to the BIM bridge design and application module. Block, the BIM bridge design and application module operates based on the hierarchical instruction call function API, obtains the initial BIM model and operation results, sends the model information corresponding to the initial BIM model to the data storage and management module for storage, sends the operation results to the GPT intelligent processing module for optimization, obtains the optimized operation results and sends the optimized operation results to the user based on the information interaction module. The present invention integrates BIM and GPT technologies, so that users can directly query BIM-related information through natural language text, and interpret user query information, summarize retrieval results and answer BIM-related questions through the GPT function, thereby reducing the workload and error rate of manual operations and lowering the requirements for the use of queries and other functions.

[0060] Furthermore, if the storage verification fails, perform the following steps:

[0061] Get the storage verification result.

[0062] If the storage verification result shows that the information is insufficient, the information to be supplemented is obtained, and the information to be supplemented is optimized using the GPT intelligent processing module. The optimized information to be supplemented is sent to the user through the information interaction module to remind the user to supplement it.

[0063] Specifically, the natural language instruction supplemented by the user is obtained, and storage verification is performed until the storage verification is successful.

[0064] If the storage verification result is an unsupported result, the GPT intelligent processing module is used to optimize the storage verification result, and the optimized storage verification result is sent to the user through the information interaction module to remind the user to modify the natural language instruction.

[0065] Specifically, the natural language instruction modified by the user is obtained, and storage verification is performed until the storage verification is successful.

[0066] Furthermore, if the functional verification fails, perform the following steps:

[0067] Get the function verification result.

[0068] If the function verification result is insufficient information, obtain target supplementary information, use the GPT intelligent processing module to optimize the target supplementary information, and send the optimized target supplementary information to the user through the information interaction module to remind the user to supplement it.

[0069] If the function verification result is that the problem is not supported, the GPT intelligent processing module is used to optimize the function verification result, and the optimized function verification result is sent to the user through the information interaction module to remind the user to modify the natural language instruction.

[0070] Specifically, in S200, the natural language instruction is sent to the GPT intelligent processing module for parsing to obtain the hierarchical instruction, including:

[0071] A preset combination list set and a preset instruction corresponding to each preset combination list are obtained, wherein the preset combination list set includes a plurality of preset combination lists, each preset combination list includes a plurality of preset field values, and the preset instruction is a statement consisting of the preset field values ​​included in the corresponding preset combination list.

[0072] The GPT intelligent processing module is used to parse the natural language instruction to obtain a target field value list, wherein the target field value list includes a plurality of target field values, and the target field value is a value corresponding to a preset field in the natural language instruction.

[0073] The target field value list and each preset combination list in the preset combination list set are used for matching to obtain an initial matching value, thereby obtaining an initial matching value list, wherein the initial matching value list includes a plurality of initial matching values.

[0074] The target field value list and each preset list in the preset combination list set are used for matching to obtain an initial matching value, including: the initial matching value is the number of target field values ​​in the preset combination list.

[0075] In another embodiment of the present invention, different preset field values ​​have different weights, and a weighted sum is taken as the initial matching value based on the weights of the preset field values.

[0076] The maximum initial matching value is obtained as the target matching value, and the preset instruction corresponding to the preset combination list corresponding to the target matching value is used as the grading instruction.

[0077] In summary, a preset combination list set and preset instructions corresponding to each preset combination list are obtained, GPT is used to parse the natural language instructions, a target field value list is obtained, the target field value list and each preset list in the preset combination list set are used for matching, an initial matching value is obtained, the maximum initial matching value is obtained as the target matching value, and the preset instructions corresponding to the preset combination list corresponding to the target matching value are used as hierarchical instructions. The present invention standardizes the natural language instructions input by the user through the preset combination list and the preset instructions, thereby improving the accuracy of the model generated by GPT.

[0078] Furthermore, when the number of target matching values ​​is greater than 1, the following steps are performed:

[0079] The preset combination list corresponding to the target matching value is used as the target combination list.

[0080] If the number of preset field values ​​included in the target combination list is equal to the number of target field values ​​in the target field value list, the preset instruction corresponding to the target combination list is used as the grading instruction, otherwise, the next step is executed.

[0081] A target difference value list is obtained, wherein the target difference value list includes a plurality of target difference values, and the target difference value is the difference between the number of preset field values ​​included in the target combination list and the number of target field values ​​in the target field value list.

[0082] The minimum target difference is obtained as the final difference, and the preset instruction corresponding to the target combination list corresponding to the final difference is used as the classification instruction.

[0083] In summary, when the number of target matching values ​​is greater than 1, the preset combination list corresponding to the target matching value is used as the target combination list. If the number of preset field values ​​contained in the target combination list is equal to the number of target field values, the preset instructions corresponding to the target combination list are used as grading instructions. Otherwise, the target difference list is obtained, and the minimum target difference is obtained as the final difference. The preset instructions corresponding to the target combination list corresponding to the final difference are used as grading instructions, so as to find the most matching preset combination list.

[0084] Embodiment 2:

[0085] Embodiment 2 of the present invention provides a bridge design device, such as Figure 2 As shown, the device comprises:

[0086] The basic data acquisition module 10 is used to acquire the basic data of the bridge model and send the basic data of the bridge model to the data storage and management module for storage.

[0087] The hierarchical instruction acquisition module 20 is used to acquire the natural language instructions input by the user based on the information interaction module, and send the natural language instructions to the GPT intelligent processing module for parsing to acquire hierarchical instructions.

[0088] The classification module 30 is used to perform storage verification on the classified instructions when the classified instructions are query instructions; and perform functional verification on the classified instructions when the classified instructions are operation instructions, wherein the operation instructions include new creation instructions, modification instructions, and deletion instructions.

[0089] The query module 40 is used to send the grading instruction to the data storage and management module for query if the storage verification is successful, obtain the query result, and send the query result to the GPT intelligent processing module for optimization, obtain the optimized query result and send the optimized query result to the user based on the information interaction module, and end.

[0090] The operation module 50 is used to send the hierarchical instruction to the BIM bridge design and application module if the function verification is successful. The BIM bridge design and application module calls the function API based on the hierarchical instruction to perform operations and obtain the initial BIM model and operation results.

[0091] The storage module 60 is used to send the model information corresponding to the initial BIM model to the data storage and management module for storage.

[0092] The optimization module 70 is used to send the operation result to the GPT intelligent processing module for optimization, obtain the optimized operation result and send the optimized operation result to the user based on the information interaction module.

[0093] Specifically, the device further includes:

[0094] The storage verification result acquisition module is used to obtain the storage verification result.

[0095] The module for acquiring information to be supplemented is used to obtain the information to be supplemented if the storage verification result is insufficient information, use the GPT intelligent processing module to optimize the information to be supplemented, and send the optimized information to be supplemented to the user through the information interaction module to remind the user to supplement it.

[0096] The storage verification module is used to optimize the storage verification result using the GPT intelligent processing module if the storage verification result is an unsupported result, and send the optimized storage verification result to the user through the information interaction module to remind the user to modify the natural language instruction.

[0097] Specifically, the device further includes:

[0098] The function verification result acquisition module is used to obtain the function verification result.

[0099] The target supplementary information acquisition module is used to obtain the target supplementary information if the function verification result is insufficient information, use the GPT intelligent processing module to optimize the target supplementary information, and send the optimized target supplementary information to the user through the information interaction module to remind the user to supplement it.

[0100] The function verification module is used to optimize the function verification result using the GPT intelligent processing module if the function verification result is an unsupported result, and send the optimized function verification result to the user through the information interaction module to remind the user to modify the natural language instruction.

[0101] The hierarchical instruction acquisition module 20 includes:

[0102] The preset instruction acquisition submodule is used to obtain a preset combination list set and a preset instruction corresponding to each preset combination list, wherein the preset combination list set includes a plurality of preset combination lists, each preset combination list includes a plurality of preset field values, and the preset instruction is a statement composed of the preset field values ​​contained in the corresponding preset combination list.

[0103] The target field value acquisition submodule is used to use the GPT intelligent processing module to parse the natural language instructions and obtain a target field value list, wherein the target field value list includes several target field values, and the target field value is the value corresponding to the preset field in the natural language instruction.

[0104] The initial matching value acquisition submodule is used to use the target field value list and each preset combination list in the preset combination list set to perform matching to obtain an initial matching value, thereby obtaining an initial matching value list, wherein the initial matching value list includes a plurality of initial matching values.

[0105] The target matching value acquisition submodule is used to acquire the maximum initial matching value as the target matching value, and use the preset instruction corresponding to the preset combination list corresponding to the target matching value as the grading instruction.

[0106] Among them, the target matching value acquisition submodule includes:

[0107] The target combination list acquisition unit is used to use the preset combination list corresponding to the target matching value as the target combination list.

[0108] The grading instruction acquisition unit is used to use the preset instruction corresponding to the target combination list as the grading instruction if the number of preset field values ​​contained in the target combination list is equal to the number of target field values ​​in the target field value list; otherwise, the target difference list acquisition unit is executed.

[0109] The target difference list acquisition unit is used to acquire a target difference list, wherein the target difference list includes a plurality of target differences, and the target difference value is the difference between the number of preset field values ​​included in the target combination list and the number of target field values ​​in the target field value list.

[0110] The final difference acquisition unit is used to acquire the minimum target difference as the final difference, and use the preset instruction corresponding to the target combination list corresponding to the final difference as the classification instruction.

[0111] Embodiment 3,

[0112] Embodiment three of the present invention also provides a non-transitory computer-readable storage medium, which can be set in an electronic device to store at least one instruction or at least one program related to implementing a bridge design method in embodiment one. The at least one instruction or the at least one program is loaded and executed by the processor to implement the method provided in the above embodiment.

[0113] An embodiment of the present invention further provides an electronic device, comprising a processor and the aforementioned non-transitory computer-readable storage medium.

[0114] Although some specific embodiments of the present invention have been described in detail by way of example, it should be understood by those skilled in the art that the above examples are only for illustration, not for limiting the scope of the present invention. It should also be understood by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the present invention.

Claims

1. A bridge design method, which is implemented based on BIM and GPT, characterized in that: The method comprises the following steps: Obtaining basic data of the bridge model, and sending the basic data of the bridge model to the data storage and management module for storage; Based on the information interaction module, the natural language instructions input by the user are obtained, and the natural language instructions are sent to the GPT intelligent processing module for parsing to obtain the classification instructions; When the hierarchical instruction is a query instruction, a storage check is performed on the hierarchical instruction; when the hierarchical instruction is an operation instruction, a function check is performed on the hierarchical instruction, wherein the operation instruction includes a new instruction, a modification instruction, and a deletion instruction; If the storage verification is successful, the classification instruction is sent to the data storage and management module for query, the query result is obtained, and the query result is sent to the GPT intelligent processing module for optimization, the optimized query result is obtained and the optimized query result is sent to the user based on the information interaction module, and the process ends; If the function verification is successful, the hierarchical instruction is sent to the BIM bridge design and application module. The BIM bridge design and application module calls the function API based on the hierarchical instruction to perform operations and obtain the initial BIM model and operation results. Sending the model information corresponding to the initial BIM model to the data storage and management module for storage; The operation results are sent to the GPT intelligent processing module for optimization, the optimized operation results are obtained and sent to the user based on the information interaction module; The natural language instructions are sent to the GPT intelligent processing module for parsing to obtain the hierarchical instructions, including: Obtaining a preset combination list set and a preset instruction corresponding to each preset combination list, wherein the preset combination list set includes a plurality of preset combination lists, each preset combination list includes a plurality of preset field values, and the preset instruction is a statement consisting of the preset field values ​​included in the corresponding preset combination list; Use the GPT intelligent processing module to parse the natural language instruction to obtain a target field value list, wherein the target field value list includes a plurality of target field values, and the target field value is a value corresponding to a preset field in the natural language instruction; Using the target field value list and each preset combination list in the preset combination list set to perform matching, obtaining an initial matching value, thereby obtaining an initial matching value list, wherein the initial matching value list includes a plurality of initial matching values; The maximum initial matching value is obtained as the target matching value, and the preset instruction corresponding to the preset combination list corresponding to the target matching value is used as the grading instruction.

2. The bridge design method according to claim 1, characterized in that: The basic data of the bridge model includes bridge route information, bridge edge information, terrain information, and geological information.

3. The bridge design method according to claim 1, characterized in that: The formats of the bridge model basic data include txt text format, xlsx text format, dwg drawing file format, dxf drawing file format, dgn BIM model format, fbx BIM model format, rvt BIM model format, and ifc BIM model format.

4. The bridge design method according to claim 1, characterized in that: The model information corresponding to the initial BIM model includes: location information and attribute information, wherein the attribute information includes size information and material information.

5. The bridge design method according to claim 1, characterized in that: If the storage verification fails, perform the following steps: Get the storage verification result; If the storage verification result is that the information is insufficient, obtain the information to be supplemented, use the GPT intelligent processing module to optimize the information to be supplemented, and send the optimized information to be supplemented to the user through the information interaction module to remind the user to supplement; If the storage verification result is an unsupported result, the GPT intelligent processing module is used to optimize the storage verification result, and the optimized storage verification result is sent to the user through the information interaction module to remind the user to modify the natural language instruction.

6. The bridge design method according to claim 1, characterized in that: When the number of target matching values ​​is greater than 1, perform the following steps: The preset combination list corresponding to the target matching value is used as the target combination list; If the number of preset field values ​​included in the target combination list is equal to the number of target field values ​​in the target field value list, the preset instruction corresponding to the target combination list is used as the grading instruction, otherwise, the next step is executed; Obtain a target difference value list, wherein the target difference value list includes a plurality of target difference values, wherein the target difference value is a difference between the number of preset field values ​​included in the target combination list and the number of target field values ​​in the target field value list; The minimum target difference is obtained as the final difference, and the preset instruction corresponding to the target combination list corresponding to the final difference is used as the classification instruction.

7. A bridge design device, characterized in that: The device comprises: A basic data acquisition module is used to acquire basic data of the bridge model and send the basic data of the bridge model to the data storage and management module for storage; A hierarchical instruction acquisition module is used to acquire natural language instructions input by the user based on the information interaction module, and send the natural language instructions to the GPT intelligent processing module for parsing to acquire hierarchical instructions; A classification module, used for performing storage verification on the hierarchical instruction when the hierarchical instruction is a query instruction; and performing function verification on the hierarchical instruction when the hierarchical instruction is an operation instruction, wherein the operation instruction includes a new creation instruction, a modification instruction, and a deletion instruction; The query module is used to send the classification instruction to the data storage and management module for query if the storage verification is successful, obtain the query result, and send the query result to the GPT intelligent processing module for optimization, obtain the optimized query result and send the optimized query result to the user based on the information interaction module, and end; The operation module is used to send the hierarchical instruction to the BIM bridge design and application module if the function verification is successful. The BIM bridge design and application module calls the function API based on the hierarchical instruction to perform the operation and obtain the initial BIM model and the operation result; A storage module is used to send model information corresponding to the initial BIM model to the data storage and management module for storage; The optimization module is used to send the operation results to the GPT intelligent processing module for optimization, obtain the optimized operation results and send the optimized operation results to the user based on the information interaction module; Among them, the hierarchical instruction acquisition module includes: A preset instruction acquisition submodule is used to acquire a preset combination list set and a preset instruction corresponding to each preset combination list, wherein the preset combination list set includes a plurality of preset combination lists, each preset combination list includes a plurality of preset field values, and the preset instruction is a statement composed of the preset field values ​​included in the corresponding preset combination list; A target field value acquisition submodule is used to parse the natural language instruction using the GPT intelligent processing module to obtain a target field value list, wherein the target field value list includes a plurality of target field values, and the target field value is a value corresponding to a preset field in the natural language instruction; An initial matching value acquisition submodule is used to use the target field value list and each preset combination list in the preset combination list set to perform matching to obtain an initial matching value, thereby obtaining an initial matching value list, wherein the initial matching value list includes a plurality of initial matching values; The target matching value acquisition submodule is used to acquire the maximum initial matching value as the target matching value, and use the preset instruction corresponding to the preset combination list corresponding to the target matching value as the grading instruction.

8. A non-transitory computer-readable storage medium, wherein at least one instruction or at least one program is stored in the storage medium, characterized in that: The at least one instruction or the at least one program is loaded and executed by the processor to implement the bridge design method as described in any one of claims 1-6.

9. An electronic device, characterized in that: The invention comprises a processor and the non-transitory computer-readable storage medium as claimed in claim 8.

Images