IFC data analysis method, device, equipment, storage medium and program product

Abstract

This application discloses an IFC data parsing method, apparatus, device, storage medium, and program product, relating to the field of large language model technology. The method includes: constructing a database of IFC data to be searched based on standard documents for various versions of IFC data; searching the database for the IFC data to be parsed, obtained by converting BIM data into an IFC format, to obtain query results; and inputting the IFC data to be parsed and the query results into a large language model for inference to obtain parsed content. This application can improve the efficiency of IFC data parsing.

Description

Technical Field

[0001] The present application relates to the technical field of large language models, and in particular to an IFC data parsing method, an IFC data parsing apparatus, an IFC data parsing device, a storage medium, and a computer program product. Background Art

[0002] Existing methods for parsing IFC (Industry Foundation Classes, a data exchange standard) data include direct parsing and parsing using specialized software. Directly parsing IFC data is difficult and requires professional technicians with specialized knowledge to read. Furthermore, these technicians must refer to the IFC standard document and identify the meaning of the IFC data line by line. This approach is less practical and results in low IFC data parsing efficiency. Using specialized software to parse IFC data also offers limited flexibility and requires checking the attribute information of the IFC data corresponding to the BIM (Building Information Modeling) model one by one. This is cumbersome and labor-intensive, resulting in low IFC data parsing efficiency. Summary of the Invention

[0003] The main purpose of this application is to provide an IFC data parsing method, an IFC data parsing device, an IFC data parsing equipment, a storage medium and a computer program product, aiming to solve the technical problem of low efficiency of IFC data parsing.

[0004] To achieve the above objectives, this application proposes an IFC data parsing method, which includes:

[0005] Build a database of IFC data to be retrieved based on the standard documents of each version of IFC data;

[0006] The IFC data to be parsed obtained by converting the BIM data into the IFC format is searched in the database to be searched to obtain a query result;

[0007] The IFC data to be parsed and the query result are input into a large language model for reasoning to obtain parsed content.

[0008] In one embodiment, the step of constructing a database of IFC data to be searched based on the standard documents of various versions of IFC data includes:

[0009] Preprocess the standard documents of each version of IFC data to obtain the target document;

[0010] The target document is converted into a vector representation, and a database of IFC data to be retrieved is constructed according to the target document represented by the vector.

[0011] In one embodiment, the step of inputting the IFC data to be parsed and the query result into a large language model for inference to obtain parsed content includes:

[0012] The IFC data to be parsed and the query results are spliced according to the preset prompt word template to obtain the prompt words for reasoning in the large language model;

[0013] In the large language model, reasoning is performed on the IFC data to be parsed according to the prompt words to obtain parsed content.

[0014] In one embodiment, after the step of inputting the IFC data to be parsed and the query result into a large language model for inference to obtain parsed content, the following steps are included:

[0015] Determine entities related to the current entity based on the association relationships between entities in the IFC data to be parsed;

[0016] The parsed content of the related entity is added to the parsed content of the current entity to obtain parsed content with an associated relationship.

[0017] In one embodiment, after the step of inputting the IFC data to be parsed and the query result into a large language model for inference to obtain parsed content, the following steps are included:

[0018] Extracting first parsed content from the parsed content;

[0019] Parsing the IFC data to be parsed in the parsing software to obtain second parsed content;

[0020] Comparing the first parsed content and the second parsed content of the same entity to obtain a comparison result;

[0021] According to the comparison results, the IFC data parsing method is optimized.

[0022] In one embodiment, the IFC data parsing method includes:

[0023] Extracting a first parsed content from the parsed content with an association relationship, wherein the association relationship is an association relationship between a current entity and an entity related to the current entity;

[0024] Parsing the IFC data to be parsed in the parsing software to obtain second parsed content;

[0025] Comparing the first parsed content and the second parsed content of the entity with the association relationship to obtain a comparison result;

[0026] According to the comparison results, the IFC data parsing method is optimized.

[0027] In addition, to achieve the above-mentioned purpose, the present application also proposes an IFC data parsing device, which includes: a construction module for constructing a database of IFC data to be searched based on standard documents of various versions of IFC data;

[0028] A retrieval module is used to search the IFC data to be parsed obtained by converting the BIM data into the IFC format in the database to be searched to obtain a query result;

[0029] The parsing module is used to input the IFC data to be parsed and the query result into a large language model for reasoning to obtain parsed content.

[0030] In addition, to achieve the above-mentioned purpose, the present application also proposes an IFC data parsing device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the IFC data parsing method described above.

[0031] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium and stores a computer program. When the computer program is executed by a processor, the steps of the IFC data parsing method described above are implemented.

[0032] In addition, to achieve the above-mentioned object, the present application also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps of the IFC data parsing method described above are implemented.

[0033] One or more technical solutions proposed in this application have at least the following technical effects:

[0034] Due to the existing methods of directly parsing IFC data and using specialized software to parse IFC data, the efficiency of IFC data parsing is low. The present application constructs a database of IFC data to be retrieved based on the standard documents of IFC data. During parsing, the parsed content is obtained through retrieval and large language model reasoning. There is no need for professionals to compare standard documents line by line, which reduces the dependence on professional knowledge and improves the efficiency of IFC data parsing. Through retrieval and large language model reasoning, the IFC data to be parsed can be quickly compared and retrieved in the database to be retrieved. The large language model can quickly infer the query results and the data to be parsed, which can reduce the tedious work of checking attribute information one by one, thereby effectively improving the parsing efficiency of IFC data. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0036] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0037] Figure 1 A flowchart of an embodiment of the IFC data parsing method provided in this application;

[0038] Figure 2 A brief flowchart of the IFC data parsing method provided in this application;

[0039] Figure 3 A flowchart of the IFC data parsing method provided in this application;

[0040] Figure 4 This is a schematic diagram of the module structure of the IFC data parsing device of this application;

[0041] Figure 5 This is a schematic diagram of the device structure of the hardware operating environment involved in the IFC data parsing method of this application.

[0042] The purpose, features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0043] It should be understood that the specific embodiments described herein are merely used to explain the technical solutions of the present application and are not intended to limit the present application.

[0044] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0045] In the prior art, directly parsing IFC data means opening the IFC data with a text editor, reading the IFC data line by line against the IFC standard document, and finding the corresponding meaning from the standard. Use professional software such as BIMversion to read the IFC data file, and click on the graphics one by one in the software to view the corresponding attribute information. However, directly parsing IFC files is difficult and requires professional technicians with professional knowledge to read them. It is also inefficient and has poor practicality. The method of reading through BIMversion software is not flexible enough and is too cumbersome. You can only view the attribute information corresponding to the BIM model one by one, which is too cumbersome and labor-intensive. The methods of the prior art make IFC data parsing inefficient.

[0046] Combining IFC data parsing with large models presents several challenges: IFC data is described in the Express language and has a complex format, whereas large models typically accept standardized data formats. Converting IFC data to a format that large models can process requires complex data parsing and conversion. Furthermore, IFC data contains a vast amount of information covering the entire lifecycle of a construction project, resulting in a massive amount of data. Large models require powerful computing resources and efficient algorithms to ensure speed and accuracy when processing such large amounts of data. Furthermore, they require a deep understanding of architectural knowledge to accurately process and analyze IFC data.

[0047] The present application provides an IFC data parsing method based on a large model. During parsing, the parsed content is obtained through retrieval and large language model reasoning. There is no need for professionals to compare standard documents line by line, which reduces the dependence on professional knowledge. It can quickly infer the query results and the data to be parsed, and can reduce the tedious work of checking attribute information one by one, thereby effectively improving the parsing efficiency of IFC data.

[0048] It should be noted that the execution entity of this embodiment may be an IFC data parsing device, or a computing service device with data processing, network communication, and program execution capabilities, such as a tablet computer, personal computer, or mobile phone, or an electronic device or processor capable of implementing the aforementioned functions. This embodiment and the following embodiments will be described below using an IFC data parsing device as an example.

[0049] Based on this, the embodiment of the present application provides an IFC data parsing method, referring to Figure 1 , Figure 1 This is a flowchart of an embodiment of the IFC data parsing method of the present application.

[0050] In this embodiment, the IFC data parsing method includes steps S10 to S30:

[0051] Step S10, constructing a database of IFC data to be searched based on the standard documents of each version of IFC data;

[0052] It's important to note that IFC is a data exchange standard that defines how various types of information are expressed and exchanged throughout the entire lifecycle of a building project, including design, construction, and operations. The standard document is a standardized description of IFC data, including detailed definitions of entities (such as walls, columns, and beams), attributes (such as component size and material), and the relationships between them. The standard document is used to understand and process IFC data. For example, multiple versions of the IFC data standard document are available on the official IFC website.

[0053] Building a searchable database using various versions of IFC standard documents involves converting various information within the IFC standard documents (such as text descriptions and data structures) into vector representations, such as by using an embedding model to convert the standard documents into numeric vectors. The goal of vectorization is to leverage the characteristics of vector operations, such as efficient similarity calculations (e.g., cosine similarity).

[0054] First, build a vector database, such as Faiss, Annoy, the vector search plug-in for Elasticsearch, Milvus, or ChromaDB, to store the vectorized data. Store the vector data corresponding to different versions of the IFC standard document in the established vector database for subsequent query and retrieval operations.

[0055] In one embodiment, the database to be searched is a vector database, which stores vectorized information of various versions of IFC data standard documents. The vector representation of various aspects of knowledge about IFC data structure, entities, attributes, etc. is equivalent to a knowledge base and has certain reference value when parsing the IFC data to be parsed.

[0056] In one embodiment, step S10 includes steps A10 to A20:

[0057] Step A10, pre-processing the standard documents of each version of IFC data to obtain a target document;

[0058] It should be noted that the standard documents for each version of IFC data are constructed and stored separately. Preprocessing includes data processing, document segmentation, and content extraction. Data cleaning involves cleaning the collected IFC standard documents, removing irrelevant information, correcting errors, and unifying the format. This includes steps such as removing HTML tags, deleting stop words, word segmentation, and removing special characters. Document segmentation involves splitting the IFC data standard documents into smaller parts due to model input limitations (such as token number restrictions). These parts should maintain semantic integrity as much as possible, such as by segmenting by paragraph or chapter. Content extraction involves extracting key information from the IFC standard documents, such as titles, paragraphs, keywords, data rules, etc., to facilitate subsequent vector generation and storage.

[0059] Step A20: convert the target document into a vector representation, and construct a database of IFC data to be retrieved based on the target document represented by the vector.

[0060] It should be noted that the target document obtained from the above preprocessing is converted into a numerical vector using an embedding model. The embedding model can be a pretrained model such as BERT, GPT, or a customized model specific to the task. Using the embedding model involves calling its API or loading it into a local environment. The cleaned and segmented target document or fragment of the target document is sent to the embedding model as input, and the embedding model returns a vector representation of the target document. The vector captures the deep semantic features of the text. Finally, the generated vector is stored in a vector database, thereby constructing a database of IFC data to be retrieved.

[0061] In this implementation, preprocessing various versions of IFC data standard documents can improve data quality. Different versions of IFC data may have inconsistent formats or different representations. Preprocessing can convert this data into a more unified format, resulting in more accurate search results. A database to be searched based on vector representations can more accurately locate relevant IFC data by calculating the similarity between the query vector and the target document represented by the vectors in the database. This improves search accuracy and enhances the efficiency of parsing IFC data.

[0062] Step S20, performing IFC data conversion on the BIM data to obtain the IFC data to be parsed, searching the database to be searched, and obtaining the query result;

[0063] It should be noted that this step is equivalent to data preparation, which requires collecting and organizing the IFC data to be extracted, confirming that the IFC data can be opened normally, and confirming the IFC version. If the BIM data has not been converted to IFC format, it must be exported to IFC format using BIM software. BIM, or Building Information Modeling, is a digital architectural design and management method. Converting BIM data to IFC format means converting the data generated by BIM software into a format that complies with the IFC standard so that it can be used in other software or systems that support IFC. For example, in Revit software, the building model can be saved as an IFC file through the "Export" function. The IFC data to be parsed contains a large amount of building information, but it needs to be parsed before it can be understood and applied.

[0064] Before retrieving the IFC data to be parsed in the database to be retrieved, the IFC data to be parsed needs to be converted into a vector representation through the embedded model. Through vectorization, the originally complex IFC data can be converted into a form that is easier for computers to process, thus preparing for subsequent parsing.

[0065] The vector database has stored vectorized data of standard documents of various versions of IFC data; in the vector database, the vectorized IFC data to be parsed can be retrieved, and the vector data most similar to the vector data to be retrieved can be found in the vector database by calculating the similarity measure between the vector data to be retrieved and the vector data stored in the database; then the top 100 most similar vector data are used as query results, where the query results are content in the IFC standard document related to the IFC data to be parsed. In addition, there is no specific limit on the specific number of the top vector data to be used as the query result.

[0066] Step S30: Input the IFC data to be parsed and the query result into the large language model for inference to obtain parsed content.

[0067] It's important to note that the IFC data to be parsed and the query results are provided as input to the large language model. The large language model then infers the IFC data to be parsed based on its own algorithms and learned knowledge, combined with the reference knowledge provided by the vector database, to produce the parsed content. For example, the large model's input could be "Input content: IFC data, Reference knowledge: Query results." Alternatively, it could be "Based on the following input content and reference knowledge, generate an answer that correctly parses and translates the input content. The answer must be easy to read and understand." The resulting parsed content can be used for subsequent architectural design, construction management, operations, and other related tasks.

[0068] In this embodiment, a database of IFC data to be retrieved is constructed based on the standard documentation for IFC data. During parsing, the parsed content is obtained through retrieval and inference using a large language model. This eliminates the need for professionals to compare line-by-line with the standard documentation, reducing reliance on specialized knowledge and improving the efficiency of IFC data parsing. Through retrieval and inference using a large language model, IFC data to be parsed can be quickly compared and retrieved from the database to be retrieved. The large language model can quickly infer query results and the data to be parsed, reducing the tedious work of individually reviewing attribute information, thereby effectively improving the efficiency of IFC data parsing.

[0069] In one embodiment, step S30 includes steps B10 to B20:

[0070] Step B10: splicing the IFC data to be parsed and the query results with reference to a preset prompt word template to obtain prompt words for reasoning in the large language model;

[0071] It should be noted that the IFC data to be parsed, obtained based on format conversion, and the query results retrieved from the vector database based on the IFC data to be parsed are spliced together with the query results using a "preset prompt word template" to construct "prompt words" suitable for reasoning in the large language model. In the large language model, prompt words serve as instructions or guidance for the large language model, telling it what information to base its reasoning, analysis, or answer. By splicing the IFC data and query results into prompt words according to the template, the large language model can use this data related to the construction field to perform specific reasoning tasks, thereby obtaining results that meet the requirements.

[0072] Step B20: Inferring the IFC data to be parsed based on the prompt words in the large language model to obtain parsed content.

[0073] It should be noted that the large language model infers the IFC data to be parsed based on the received prompt words, and uses the language knowledge, semantic information, and logical relationships learned during its pre-training process to conduct in-depth analysis of the IFC data.

[0074] For example, parsing IFC data involves translating the IFC data EXPRESS language into a readable and understandable language. Based on a knowledge base, IFC data is parsed using RAG (Retrieval-augmented Generation) technology. IFC data is text data and can be parsed row by row. For example, consider this row of IFC data:

[0075] IFCWALLSTANDARDCASE('2A_TnvSvT3Ju6Ruoo5s7HB',#268437024,'IfcWall-4',$,$,#268678164,#268678331,$”)”; The IFC data to be parsed and the query results are then combined into a context according to the template to form an enhanced prompt (prompt word) for the large language model to answer the question. An example template is: "According to the requirements, based on the content and reference knowledge in the following input, generate an answer that correctly parses and translates the input content. The answer must be easy to read and understand. If the reference knowledge provided does not provide assistance in answering the question, form an answer based on your own knowledge." The above prompt is input into the large language model, which then performs inference and outputs the parsed content, such as:

[0076] “#268678167: The globally unique identifier of this node;

[0077] IFCWALLSTANDARDCASE: instantiated entity type, representing a standard wall;

[0078] 2A_TnvSvT3Ju6Ruoo5s7HB: ID of the standard wall;

[0079] #268437024: The #268437024 node in this file. For its specific meaning, see the content of this node.

[0080] IfcWall-4: component name;

[0081] $: indicates an empty value, this position should be a description;

[0082] $: represents a null value, this position should be the type;

[0083] #268678164: The #268678164 node in this file. For its specific meaning, see the content of this node.

[0084] #268678331: The #268678331 node in this file. For its specific meaning, see the content of this node.

[0085] $: Indicates an empty value, a label should be placed at this position. ".

[0086] In this embodiment, by splicing the IFC data to be parsed and the query results according to the preset prompt word template, the large language model can perform reasoning under a complete information framework, so that the large language model can quickly locate relevant information and perform reasoning, avoiding information fragmentation, thereby improving the accuracy of reasoning, obtaining the parsed content more quickly, and thus improving the efficiency of parsing.

[0087] In another embodiment, step S30 includes steps E10 to E20:

[0088] Step E10, determining entities related to the current entity based on the association relationship between entities in the IFC data to be parsed;

[0089] It should be noted that in IFC data, entities are abstract representations of various objects in the field of construction engineering. For example, in the IFC data of a construction project, entities can be building components such as walls, columns, and beams in a building, or they can be elements such as people and materials in the project. These entities contain various attribute information related to themselves, such as the height, thickness, and material type of the wall. Various associations exist between entities, and the associations reflect the actual connections between different objects in the construction project. For example, there may be a connection relationship between a wall entity and a column entity, and the wall may be supported by the column; or there may be a load-bearing relationship between a beam entity and a wall entity, and the beam may bear the weight of the wall.

[0090] Based on the association relationships between these entities, we can find other entities related to it. For example, if the current entity is a window, based on the association relationships, we can determine that its related entities include walls (because windows are usually installed on walls), rooms (the room where the window is located), etc.

[0091] In step E20, the parsed content of the related entity is added to the parsed content of the current entity to obtain parsed content with an associated relationship.

[0092] It's important to note that both the related entity and the current entity have their own parsed content. Adding the parsed content of the related entity to the parsed content of the current entity means integrating the related entity with the current entity's information system, resulting in parsed content with the current entity. Parsed content with the current entity is more comprehensive and rich, encompassing not only the current entity's own information but also information about the impact of other entities related to the current entity on the current entity.

[0093] For example, since there is an association relationship between nodes in IFC data, the associated attribute information is associated, and the association can be performed using methods such as keyword matching and association query. After the association is performed, the parsed content with the association relationship is “#268678167: the globally unique identifier of the node;

[0094] IFCWALLSTANDARDCASE: instantiated entity type, representing a standard wall;

[0095] 2A_TnvSvT3Ju6Ruoo5s7HB: ID of the standard wall;

[0096] #268437024: This standard wall history was created by the modifier;

[0097] IfcWall-4: component name;

[0098] $: indicates an empty value, this position should be a description;

[0099] $: represents a null value, this position should be the type;

[0100] #268678164: IFCLOCALPLACEMENT(#122,#134); The relative position of this standard wall in the entire building space coordinate system;

[0101] #268678331: IFCPRODUCTDEFINITIONSHAPE($,$,(#141,#159)); The style, shape, and shape generation method of this standard wall;

[0102] $: Indicates an empty value, a label should be placed at this position. ".

[0103] In this implementation, related entities are identified based on the relationships between them. This avoids viewing individual entities in isolation, helps grasp the layout and functional relationships of the building as a whole, constructs a more complete spatial relationship network, and improves the completeness of the understanding of the entire IFC data.

[0104] Based on the above embodiment of the present application, in another embodiment of the present application, the same or similar contents as the above embodiment can be referred to the above introduction and will not be described in detail later. After step S30, the IFC data parsing method further includes:

[0105] Extracting first parsed content from the parsed content;

[0106] Parsing the IFC data to be parsed in parsing software to obtain second parsed content; comparing the first parsed content with the second parsed content of the same entity to obtain a comparison result;

[0107] According to the comparison results, the IFC data parsing method is optimized.

[0108] It should be noted that the first parsed content is extracted from the parsed content derived from the large model inference. The first parsed content is a portion of the results obtained by parsing the IFC data. The IFC data to be parsed is parsed in specialized parsing software. In one feasible implementation, the parsing can be automated using a script. The result obtained by parsing the IFC data using the parsing software is the second parsed content.

[0109] For the same entity, the first parsing content and the second parsing content are compared. Specifically, the attribute values of the entities can be compared. For example, the length of a beam entity in the first parsing content is 5 meters, while the length of the beam entity in the second parsing content is 4.9 meters. This is the difference in attribute values. Based on the comparison results, the IFC data parsing method is optimized. If the comparison results show that there are certain inaccuracies in the first parsing content (for example, there are large deviations in the entity attribute values), the prompt words can be readjusted, the knowledge base can be updated, the vector retrieval model can be replaced, etc. For example, if it is found that the first parsing content is unique in the qualitative analysis of certain entity attributes, and the second parsing content of the parsing software is more accurate in quantitative analysis, then the two can be combined to optimize the parsing method so that it can perform both accurate qualitative analysis and precise quantitative analysis, thereby improving the overall quality of IFC data parsing.

[0110] In this embodiment, by extracting the first parsed content from the parsed content and obtaining the second parsed content through the parsing software, and then comparing them, the difference in accuracy between the two parsing methods can be found. After optimizing the parsing method based on the difference, the parsing efficiency can be improved.

[0111] In one embodiment, the IFC data parsing method further includes:

[0112] Extracting a first parsed content from the parsed content with an association relationship, wherein the association relationship is an association relationship between a current entity and an entity related to the current entity;

[0113] Parsing the IFC data to be parsed in the parsing software to obtain second parsed content;

[0114] Comparing the first parsed content and the second parsed content of the entity with the association relationship to obtain a comparison result;

[0115] According to the comparison results, the IFC data parsing method is optimized.

[0116] It should be noted that the analysis content obtained by large model reasoning can be compared with the analysis content obtained by the analysis software; the analysis content with association relationships can also be compared with the analysis content obtained by the analysis software.

[0117] An association relationship is the relationship between the current entity and related entities. The first parsed content is extracted from the parsed content with the association relationship. Then, the IFC data to be parsed is put into specialized parsing software for parsing to obtain the second parsed content. Next, for the entities with the association relationship, the first parsed content and the second parsed content are compared, taking into account the relationship between the entities and the attributes of the entities themselves. Finally, the IFC data parsing method is optimized based on the comparison results.

[0118] For example, the incorrectly parsed data is analyzed to determine the causes of the parsing errors, such as inaccurate prompts, failure to find corresponding text in the retrieval, etc. Based on the analysis results, the IFC data parsing method is optimized, prompts are adjusted, the knowledge base is updated, and the vector retrieval model is replaced to increase retrieval accuracy. The above steps are repeated for multiple experiments to continuously improve the accuracy of IFC data parsing.

[0119] In this implementation, by extracting a first parsed content from the parsed content with associated relationships and comparing it with a second parsed content generated by the parsing software, differences in entity attribute parsing can be identified. The parsing method can then be optimized to address these differences, thereby improving the accuracy of entity attribute parsing. By comparing and optimizing the parsing method, such associations can be parsed more accurately, making the parsing process more targeted, reducing repetitive parsing work, and improving the efficiency of IFC data parsing.

[0120] For example, to help understand the brief process of the IFC data parsing method obtained by combining this embodiment with the above embodiments, please refer to Figure 2 , Figure 2 A brief flowchart of an IFC data parsing method is provided. Specifically, the entire IFC data parsing process may include: knowledge base construction, namely, pre-processing the standard document of IFC data, converting it into a numerical vector, and storing it in a vector library; data preparation, namely, converting the format of BIM data to obtain the IFC data to be parsed, converting the IFC data to be parsed into a numerical vector, searching in a vector database, and obtaining a query result; data parsing, namely, splicing the IFC data to be parsed and the query result into prompt words, and reasoning the IFC data with reference to the knowledge base to obtain parsed content; result collation, namely, adding related entities with an association relationship with the current entity to the parsed content of the current entity; and continuous optimization, namely, comparing the parsed content with the results obtained by the parsing software, and optimizing the IFC data parsing process according to the comparison result.

[0121] For example, please refer to Figure 3 , Figure 3 A flow chart of an IFC data parsing method is provided, specifically: after vectorizing standard documents of various versions of IFC data, the documents are stored in a vector database respectively; after vectorizing the IFC data to be parsed, a search query is performed in the vector database to obtain a query result; using the standard documents of various versions as a reference knowledge base and the concatenation of the IFC data to be parsed and the query result as a prompt word, the IFC data to be parsed is inferred in a large language model to obtain parsed content; the parsed content obtained by the large model reasoning can be compared with the parsed content obtained by the parsing software to obtain a comparison result, thereby optimizing the IFC data parsing method; and the parsed content with an associated relationship can be compared with the parsed content obtained by the parsing software to obtain a comparison result, thereby optimizing the IFC data parsing method.

[0122] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the IFC data parsing method of the present application. Based on this technical concept, more forms of simple transformations, such as the interaction and combination of various embodiments, are all within the scope of protection of the present application.

[0123] This application also provides an IFC data parsing device, please refer to Figure 4 , the IFC data parsing device includes:

[0124] A construction module 10 is used to construct a database of IFC data to be searched based on standard documents of various versions of IFC data;

[0125] A retrieval module 20 is configured to perform a search on the IFC data to be parsed obtained by converting the BIM data into the IFC format, and obtain a query result in the database to be searched.

[0126] The parsing module 30 is configured to input the IFC data to be parsed and the query result into a large language model for inference to obtain parsed content.

[0127] Optionally, the construction module 10 is further configured to pre-process the standard documents of each version of the IFC data to obtain a target document;

[0128] The target document is converted into a vector representation, and a database of IFC data to be retrieved is constructed according to the target document represented by the vector.

[0129] Optionally, the parsing module 30 is further configured to combine the IFC data to be parsed and the query result with reference to a preset prompt word template to obtain prompt words for reasoning in the large language model;

[0130] In the large language model, reasoning is performed on the IFC data to be parsed according to the prompt words to obtain parsed content.

[0131] Optionally, the parsing module 30 is further configured to determine entities related to the current entity based on association relationships between entities in the IFC data to be parsed;

[0132] The parsed content of the related entity is added to the parsed content of the current entity to obtain parsed content with an associated relationship.

[0133] Optionally, the parsing module 30 is further configured to extract first parsed content from the parsed content;

[0134] Parsing the IFC data to be parsed in the parsing software to obtain second parsed content;

[0135] Comparing the first parsed content and the second parsed content of the same entity to obtain a comparison result;

[0136] According to the comparison results, the IFC data parsing method is optimized.

[0137] Optionally, the parsing module 30 is further configured to extract a first parsed content from parsed content with an association relationship, wherein the association relationship is an association relationship between the current entity and entities related to the current entity;

[0138] Parsing the IFC data to be parsed in the parsing software to obtain second parsed content;

[0139] Comparing the first parsed content and the second parsed content of the entity with the association relationship to obtain a comparison result;

[0140] According to the comparison results, the IFC data parsing method is optimized.

[0141] The IFC data parsing device provided in this application, utilizing the IFC data parsing method described in the aforementioned embodiments, can address the technical issue of low IFC data parsing efficiency. Compared to the prior art, the IFC data parsing device provided in this application achieves the same beneficial effects as the IFC data parsing method described in the aforementioned embodiments. Other technical features of the IFC data parsing device are the same as those disclosed in the aforementioned embodiments and are not further elaborated here.

[0142] The present application provides an IFC data parsing device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the IFC data parsing method of the first embodiment described above.

[0143] Reference below Figure 5 , which shows a schematic diagram of the structure of an IFC data parsing device suitable for implementing embodiments of the present application. The IFC data parsing device in the embodiments of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 5 The IFC data parsing device shown is merely an example and should not limit the functions and scope of use of the embodiments of the present application.

[0144] like Figure 5As shown, the IFC data parsing device may include a processing device 1001 (e.g., a central processing unit, graphics processing unit, etc.), which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 1002 or programs loaded from a storage device 1003 into a random access memory (RAM) 1004. RAM 1004 also stores various programs and data required for the operation of the IFC data parsing device. Processing device 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems may be connected to I / O interface 1006: input devices 1007, such as a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008, such as a liquid crystal display (LCD), speaker, vibrator, etc.; storage device 1003, such as a magnetic tape or hard disk; and communication devices 1009. The communication device 1009 can allow the IFC data parsing device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows an IFC data parsing device with various systems, it should be understood that it is not required to implement or have all of the systems shown. More or fewer systems may be implemented or have alternatively.

[0145] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0146] The IFC data parsing device provided in this application utilizes the IFC data parsing method described in the aforementioned embodiment to address the technical issue of low IFC data parsing efficiency. Compared to the prior art, the beneficial effects of the IFC data parsing device provided in this application are the same as those of the IFC data parsing method described in the aforementioned embodiment. Other technical features of the IFC data parsing device are the same as those disclosed in the aforementioned embodiment and are not further elaborated here.

[0147] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0148] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0149] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, a computer program) stored thereon, wherein the computer-readable program instructions are used to execute the IFC data parsing method in the above embodiment.

[0150] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0151] The computer-readable storage medium may be included in the IFC data parsing device, or may exist independently without being incorporated into the IFC data parsing device.

[0152] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the IFC data parsing device, the IFC data parsing device: constructs a database of IFC data to be searched based on standard documents of various versions of IFC data;

[0153] The IFC data to be parsed obtained by converting the BIM data into the IFC format is searched in the database to be searched to obtain a query result;

[0154] The IFC data to be parsed and the query result are input into a large language model for reasoning to obtain parsed content.

[0155] Computer program code for performing the operations of the present application may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0156] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0157] The modules described in the embodiments of the present application may be implemented in software or hardware, wherein the name of a module does not necessarily limit the unit itself.

[0158] The computer-readable storage medium provided in this application stores computer-readable program instructions (i.e., a computer program) for executing the aforementioned IFC data parsing method, thereby resolving the technical issue of low IFC data parsing efficiency. Compared to the prior art, the beneficial effects of the computer-readable storage medium provided in this application are similar to those of the IFC data parsing method provided in the aforementioned embodiment and are not further elaborated here.

[0159] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned IFC data parsing method when executed by a processor.

[0160] The computer program product provided in this application can solve the technical problem of low efficiency in IFC data parsing. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the IFC data parsing method provided in the above embodiment, and will not be repeated here.

[0161] The above description is only part of the embodiments of the present application and does not limit the patent scope of the present application. All equivalent structural transformations made by using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect application in other related technical fields are included in the patent protection scope of the present application.

Claims

1. An IFC data parsing method, characterized in that: The IFC data parsing method includes: Build a database of IFC data to be retrieved based on the standard documents of each version of IFC data; The IFC data to be parsed obtained by converting the BIM data into the IFC format is searched in the database to be searched to obtain a query result; Inputting the IFC data to be parsed and the query result into a large language model for reasoning to obtain parsed content; Determining entities related to the current entity based on association relationships between entities in the IFC data to be parsed, wherein the association relationships are association relationships between the current entity and entities related to the current entity, reflecting actual connections between different objects in a construction project, wherein the actual connections include connection relationships, support relationships, or bearing relationships between entities; adding parsed content of the related entities to the parsed content of the current entity to obtain parsed content with association relationships, wherein the parsed content with association relationships includes information about the current entity itself and information about the impact of other entities associated with the current entity on the current entity; Extracting first parsed content from parsed content with an associated relationship; parsing the IFC data to be parsed in parsing software to obtain second parsed content; comparing the first parsed content with the second parsed content of the entity with the associated relationship to obtain a comparison result; and optimizing the IFC data parsing method based on the comparison result.

2. The IFC data parsing method according to claim 1, wherein: The step of constructing a database of IFC data to be searched based on the standard documents of various versions of IFC data includes: Preprocess the standard documents of each version of IFC data to obtain the target document; The target document is converted into a vector representation, and a database of IFC data to be retrieved is constructed according to the target document represented by the vector.

3. The IFC data parsing method according to claim 1, wherein: The step of inputting the IFC data to be parsed and the query result into a large language model for inference to obtain parsed content includes: The IFC data to be parsed and the query results are spliced according to the preset prompt word template to obtain the prompt words for reasoning in the large language model; In the large language model, reasoning is performed on the IFC data to be parsed according to the prompt words to obtain parsed content.

4. The IFC data parsing method according to claim 1, wherein: After the step of inputting the IFC data to be parsed and the query result into a large language model for inference to obtain parsed content, the method includes: Extracting first parsed content from the parsed content; Parsing the IFC data to be parsed in the parsing software to obtain second parsed content; Comparing the first parsed content and the second parsed content of the same entity to obtain a comparison result; According to the comparison results, the IFC data parsing method is optimized.

5. An IFC data parsing device, characterized in that: The IFC data parsing device includes: A construction module is used to construct a database of IFC data to be retrieved based on the standard documents of various versions of IFC data; A retrieval module is used to search the IFC data to be parsed obtained by converting the BIM data into the IFC format in the database to be searched to obtain a query result; A parsing module, configured to input the IFC data to be parsed and the query result into a large language model for reasoning to obtain parsed content; The parsing module is further configured to determine entities related to the current entity based on association relationships between entities in the IFC data to be parsed, wherein the association relationships are association relationships between the current entity and entities related to the current entity, reflecting actual connections between different objects in a construction project, wherein the actual connections include connection relationships, support relationships, or bearing relationships between entities; and add parsed content of the related entities to the parsed content of the current entity to obtain parsed content with association relationships, wherein the parsed content with association relationships includes information about the current entity itself and information about the impact of other entities associated with the current entity on the current entity. Extracting first parsed content from parsed content with an associated relationship; parsing the IFC data to be parsed in parsing software to obtain second parsed content; comparing the first parsed content with the second parsed content of the entity with the associated relationship to obtain a comparison result; and optimizing the IFC data parsing method based on the comparison result.

6. An IFC data parsing device, characterized in that: The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the IFC data parsing method according to any one of claims 1 to 4.

7. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the IFC data parsing method according to any one of claims 1 to 4 are implemented.

8. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the IFC data parsing method according to any one of claims 1 to 4 are implemented.

Images