Engineering file diagnosis method and system
By using artificial intelligence to perform multimodal analysis and diagnosis of engineering documents, the problems of low accuracy and low efficiency of traditional manual review have been solved. This has enabled automated, consistent, and efficient diagnosis of engineering documents, identified potential risks, and improved engineering quality control.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING QDING INTERCONNECTION TECHNOLOGY CO LTD
- Filing Date
- 2026-03-24
- Publication Date
- 2026-07-07
Smart Images
Figure CN122346802A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the fields of artificial intelligence and other technologies, and in particular to a method and system for diagnosing engineering documents. Background Technology
[0002] The accuracy of installation engineering bidding relies heavily on the consistency of "drawings," "text," and "tables" (i.e., design drawings, bidding documents, and bill of quantities). The traditional review method is a "multi-party review," where engineers from various disciplines such as electrical, HVAC, plumbing, cost, and legal conduct manual cross-checking of the entire set of documents based on their experience. This is a labor-intensive, knowledge-intensive, and error-prone process; therefore, how to review installation engineering documents has become a key research focus.
[0003] Related technologies use text comparison tools (such as Beyond Compare) to check for textual differences between different versions of tender documents or bills of quantities, use CAD (Computer-Aided Design) viewing software (such as AutoCAD Viewer) to manually view drawings, count the number of equipment and identify annotation information with the naked eye, and manually check them against the bill of quantities and tender documents. However, this method has low accuracy and efficiency. Summary of the Invention
[0004] The embodiments of this application aim to at least partially solve one of the technical problems in the related art. Therefore, the purpose of the embodiments of this application is to provide an engineering document diagnostic method, system, device, and medium, which realizes automated diagnostics of engineering documents and improves the efficiency and accuracy of the diagnostics.
[0005] This application provides a method for diagnosing engineering documents, including: acquiring installation engineering documents to be diagnosed and engineering knowledge data; performing deep analysis and multimodal information extraction processing on the installation engineering documents to be diagnosed to obtain target data; performing semantic understanding processing on the target data and engineering knowledge data to perform multidimensional diagnostic processing on the target data to obtain multidimensional diagnostic results, wherein the multidimensional diagnostic results include consistency diagnostic results, compliance diagnostic results, and spatial rationality diagnostic results; and obtaining a risk diagnostic report based on the multidimensional diagnostic results.
[0006] For example, the installation project documents to be diagnosed include tender documents, drawing documents, and bill of quantities. Deep analysis and multimodal information extraction processing are performed on the installation project documents to obtain target data, including: deep analysis and information extraction of the tender documents, using the extracted key technical clauses, equipment brand information, equipment model information, and construction process requirements as the first target data; deep analysis and information extraction of the drawing documents, using the extracted graphic symbol information, graphic symbol quantity information, and text annotation information as the second target data; deep analysis and information extraction of the bill of quantities, using it as the first target information; obtaining material name information, material specification information, material unit information, and material quantity information as the third target data; and obtaining the target data based on the first, second, and third target data.
[0007] For example, semantic understanding processing is performed on the target data and engineering knowledge data to perform multi-dimensional diagnostic processing on the target data and obtain multi-dimensional diagnostic results, including: cross-document consistency diagnostic processing based on the target data to obtain consistency diagnostic results; compliance diagnostic processing based on the target data and engineering knowledge data to obtain compliance diagnostic results; and spatial rationality pre-diagnostic processing based on the target data to obtain spatial rationality diagnostic results.
[0008] For example, the first target data also includes cable specification information; cross-document consistency diagnosis processing includes quantity comparison processing, model comparison processing, and specification comparison processing; cross-document consistency diagnosis processing is performed based on the target data to obtain consistency diagnosis results, including: quantity comparison processing based on graphic symbol quantity information and material quantity information to obtain a first diagnosis result; model comparison processing based on equipment model information and text annotation information to obtain a second diagnosis result; specification comparison processing based on cable specification information and text annotation information to obtain a third diagnosis result; and consistency diagnosis results are obtained based on the first diagnosis result, the second diagnosis result, and the third diagnosis result.
[0009] For example, the target data also includes area data and usage data; compliance diagnosis processing is performed based on the target data and engineering knowledge data to obtain compliance diagnosis results, including: fire protection code judgment based on area data, usage data and engineering knowledge data to obtain compliance diagnosis results.
[0010] For example, the installation project file to be diagnosed is for a first installation project and a second installation project; spatial rationality pre-diagnosis processing is performed based on the target data to obtain spatial rationality diagnosis results, including: performing virtual overlay processing on the target data corresponding to the first installation project and the target data corresponding to the second installation project to perform spatial rationality pre-diagnosis processing on the target data to obtain spatial rationality diagnosis results.
[0011] For example, the multidimensional diagnostic results include risk points; based on the multidimensional diagnostic results, a risk diagnostic report is obtained, including: obtaining the risk diagnostic report based on the risk points and generating optimization suggestions for the risk points; in response to the user's operation instructions for the risk diagnostic report, generating prompt information for the installation project file to be diagnosed for the risk points.
[0012] For example, the engineering document diagnostic method further includes: obtaining reference data from the building information model; and verifying the tender documents and bill of quantities based on the reference data.
[0013] For example, the engineering document diagnostic method further includes: obtaining cost data; when there is a difference between the equipment model information corresponding to the tender document and the text annotation information corresponding to the drawing document, performing a cost difference analysis on the equipment model information and the text annotation information based on the cost data.
[0014] Another embodiment of this application provides an engineering document diagnostic system, which includes: an acquisition module for acquiring installation engineering documents to be diagnosed and engineering knowledge data; a parsing module for performing deep parsing and multimodal information extraction processing on the installation engineering documents to be diagnosed to obtain target data; a processing module for performing semantic understanding processing on the target data and engineering knowledge data to perform multidimensional diagnostic processing on the target data to obtain multidimensional diagnostic results, wherein the multidimensional diagnostic results include consistency diagnostic results, compliance diagnostic results, and spatial rationality diagnostic results; and an acquisition module for obtaining a risk diagnostic report based on the multidimensional diagnostic results.
[0015] Another embodiment of this application provides an electronic device having a computer program stored thereon, which, when executed by a processor, implements the steps of the method of any of the above embodiments.
[0016] Another embodiment of this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method of any of the above embodiments.
[0017] In the above embodiments, the engineering document diagnosis method includes: acquiring the installation engineering documents to be diagnosed and engineering knowledge data; performing deep analysis and multimodal information extraction processing on the installation engineering documents to be diagnosed to obtain target data; performing semantic understanding processing on the target data and engineering knowledge data to perform multi-dimensional diagnostic processing on the target data to obtain multi-dimensional diagnostic results, wherein the multi-dimensional diagnostic results include consistency diagnostic results, compliance diagnostic results, and spatial rationality diagnostic results; and obtaining a risk diagnosis report based on the multi-dimensional diagnostic results. Through deep analysis and multimodal information extraction technology, the installation engineering documents to be diagnosed are transformed into structured target data, and semantic understanding is performed by integrating engineering knowledge data. This achieves automated multi-dimensional comprehensive diagnosis of the consistency, compliance, and spatial rationality of engineering documents, significantly improving the accuracy, comprehensiveness, and efficiency of engineering document review, effectively identifying potential design conflicts, standard deviations, and spatial interference risks, thereby automatically generating risk diagnosis reports, greatly reducing the cost and omissions of manual review, and providing reliable technical support for precise control of engineering quality. Attached Figure Description
[0018] Figure 1 A flowchart of the engineering document diagnosis method provided for the embodiments of this application; Figure 2 Another flowchart of an engineering document diagnostic method provided for embodiments of this application; Figure 3 A block diagram of an engineering document diagnostic system provided for another embodiment of this application; Figure 4 A block diagram of an electronic device provided for another embodiment of this application. Detailed Implementation
[0019] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.
[0020] The accuracy of installation engineering bidding relies heavily on the consistency of "drawings," "text," and "tables" (i.e., design drawings, bidding documents, and bill of quantities). The traditional review method is a "multi-party review," where engineers from various disciplines such as electrical, HVAC, plumbing, cost, and legal conduct manual cross-checking of the entire set of documents based on their experience. This is a labor-intensive, knowledge-intensive, and error-prone process; therefore, how to review installation engineering documents has become a key research focus.
[0021] Related technologies use text comparison tools (such as Beyond Compare) to check for textual differences between different versions of tender documents or bills of quantities, use CAD (Computer-Aided Design) viewing software (such as AutoCAD Viewer) to manually view drawings, count the number of equipment and identify annotation information with the naked eye, and manually check them against the bill of quantities and tender documents. However, this method has low accuracy and efficiency.
[0022] The above-mentioned related technical solutions have fundamental technical defects because they cannot achieve "cross-modal understanding" of information: (1) They cannot automatically perform cross-document and cross-modal content verification: Text comparison tools can only compare text, and CAD software can only display graphics. They are completely separate. Related technologies cannot: automatically read all the symbols of "emergency lighting" on the drawings, count them as "152", and then automatically compare them with the "150" corresponding to the "emergency lighting" item in the bill of quantities (Excel), and highlight the risk of "inconsistent quantity". (2) They lack the ability to deeply analyze embedded domain knowledge: Existing tools do not have a professional knowledge base for installation engineering. They cannot identify that a symbol on the drawing represents a "fire temperature sensor", and they cannot automatically determine whether the sensor layout in the area meets the fire protection code based on the coverage radius of the sensor and the room area. They are just "presenters" of information, not "analysts". (3) The review process is static and cannot make spatial logic predictions: Whether looking at drawings or reading text, it is two-dimensional. The relevant technology cannot virtually overlay drawings from different disciplines (such as electrical and HVAC) in space to automatically identify collisions and conflicts between different pipelines in three-dimensional space.
[0023] This application aims to address the technical problems caused by inconsistencies and mismatches in the core documents of real estate project installation engineering (water supply and drainage, strong and weak current, HVAC, fire protection, etc.) during the bidding stage, due to information inconsistencies and mismatches: (1) Difficulty in cross-document information verification, with the risk of "errors, omissions, clashes, and missing items": Installation engineering involves many types of documents and a large amount of information. The bidding documents describe the material brands and construction processes in words, the design drawings use graphic symbols and annotations to indicate the equipment locations and system routes, and the bill of quantities lists the material quantities in tables. During manual review, it is easy to make mistakes by repeatedly switching and comparing these different document formats. For example, the number of fire sprinkler heads in the bill of quantities does not match the actual number of points on the drawings (errors, omissions), or the cable tray specifications required by the bidding documents cannot accommodate all the cables in that area on the drawings (clashes). These problems are difficult to find during the bidding stage and will bring a lot of changes and cost overruns to the later construction. (2) Difficulty in reviewing the compliance of technical parameters: Installation engineering involves a large number of national regulations and industry standards. Whether the equipment models (such as water pumps and distribution boxes) and material specifications (such as pipe diameter and wire diameter) specified in the bidding documents meet the latest mandatory standards for fire protection, electrical, and energy conservation depends on the personal knowledge of the reviewers. Faced with a massive amount of standard clauses, manual review is difficult to achieve comprehensive coverage and there are compliance risks. (3) Lack of prediction of construction feasibility: Traditional document review is static and isolated. Reviewers cannot discover potential construction conflicts based solely on 2D drawings and text descriptions. For example, the HVAC ducts and fire protection cable trays may intersect at the same elevation on the drawings, which is impossible to achieve in construction. This kind of "conflict on the drawings" problem often only comes to light on the construction site, leading to rework and delays.
[0024] The purpose of this application is to provide a novel intelligent diagnostic method that can simultaneously understand text, tables, and graphical languages, and integrate domain knowledge for automated cross-validation, fundamentally solving the problems of "discrepancies between text and graphics" and "compliance" in installation engineering bidding documents. Therefore, to achieve the above objective, this application provides an engineering document diagnostic method that uses a large AI model to perform multimodal analysis on installation engineering bidding documents and drawings.
[0025] Figure 1 A flowchart illustrating the engineering document diagnostic method provided in this application.
[0026] like Figure 1 As shown, the engineering document diagnosis method 100 provided in this application includes, for example, steps S110-S140.
[0027] Step S110: Obtain the installation project files and project knowledge data to be diagnosed.
[0028] For example, the installation project files to be diagnosed include multiple installation projects, such as electrical engineering, HVAC engineering, water supply and drainage engineering, strong and weak current engineering, fire protection engineering, etc. The installation project files to be diagnosed may include tender documents (e.g., .docx format), drawing files (e.g., .pdf, .dwg format), and bill of quantities (e.g., .xlsx format), etc. The engineering knowledge data includes national and industry standard data (fire protection, electrical, energy conservation, etc.), equipment and material brand data, and historical project problem data.
[0029] Step S120: Perform in-depth analysis and multimodal information extraction processing on the installation project file to be diagnosed to obtain target data.
[0030] For example, deep analysis of drawing documents is based on computer vision (CV) and CAD analysis technologies. The target data includes first target data, second target data, and third target data. The first target data includes key technical clauses, equipment brand information, equipment model information, and construction process requirements obtained through deep analysis and information extraction of the tender documents. The second target data includes graphic symbol information, graphic symbol quantity information, and text annotation information (such as graphic model, specifications, and elevation) obtained through deep analysis and information extraction of the drawing documents. The third target data includes material name information, material specification information, material unit information, and material quantity information obtained through deep analysis and information extraction of the bill of quantities.
[0031] Step S130: Semantic understanding processing is performed on the target data and engineering knowledge data to perform multi-dimensional diagnostic processing on the target data and obtain multi-dimensional diagnostic results, including consistency diagnostic results, compliance diagnostic results and spatial rationality diagnostic results.
[0032] For example, the AI (Artificial Intelligence) intelligent diagnostic engine first performs semantic understanding processing on the target data and engineering knowledge data, and then performs multi-dimensional diagnostic processing on the target data based on the engineering construction knowledge data, thereby obtaining multi-dimensional diagnostic results. The multi-dimensional diagnostic processing includes cross-document consistency diagnostic processing, compliance diagnostic processing, and spatial rationality pre-diagnostic processing. The multi-dimensional diagnostic results include the risk points corresponding to each dimension (document consistency risk points, compliance risk points, and spatial rationality risk points).
[0033] Step S140: Based on the multidimensional diagnostic results, a risk diagnostic report is obtained.
[0034] For example, the AI big model generates optimization suggestions based on multi-dimensional diagnostic results. The risk diagnosis report includes the risk points corresponding to each dimension and the optimization suggestions corresponding to each risk point.
[0035] In the above embodiments, deep analysis and multimodal information extraction technologies are used to transform the installation engineering documents to be diagnosed into structured target data. By integrating engineering knowledge data for semantic understanding, automated multi-dimensional comprehensive diagnosis of the consistency, compliance, and spatial rationality of engineering documents is achieved. This significantly improves the accuracy, comprehensiveness, and efficiency of engineering document review, effectively identifies potential design conflicts, standard deviations, and spatial interference risks, and automatically generates risk diagnosis reports. This greatly reduces the cost and omissions of manual review and provides reliable technical support for the precise control of engineering quality.
[0036] In one example, the installation project documents to be diagnosed include tender documents, drawings, and a bill of quantities. Deep analysis and multimodal information extraction are performed on these documents to obtain target data, including: deep analysis and information extraction of the tender documents, using key technical clauses, equipment brand information, equipment model information, and construction process requirements as the first target data; deep analysis and information extraction of the drawings, using graphic symbol information, graphic symbol quantity information, and text annotation information as the second target data; deep analysis and information extraction of the bill of quantities as the first target information; and obtaining material name information, material specification information, material unit information, and material quantity information as the third target data. Based on the first, second, and third target data, the final target data is obtained.
[0037] Specifically, deep analysis and multimodal information extraction processing are implemented based on a multimodal information extraction module (including text parsing, image parsing, and table parsing units). The text parsing unit deeply parses the text of the tender document, extracting key technical clauses, specified equipment brands (equipment brand information) and models (equipment model information), construction process requirements, etc. The image / CAD parsing unit uses computer vision (CV) and CAD file parsing technology to analyze drawings. It can identify standard electrical, water supply and drainage, and fire protection graphic symbols (graphic symbol information), count their quantities (graphic symbol quantity information), and read adjacent textual annotations (such as model, specifications, elevation, etc.). The table parsing unit parses the Bill of Quantities (BOQ) Excel file, converting it into a structured database containing fields such as material name information, specifications (material specification information), unit (material unit information), and quantity (material quantity information).
[0038] In one example, semantic understanding processing is performed on the target data and engineering knowledge data to perform multi-dimensional diagnostic processing on the target data, resulting in multi-dimensional diagnostic results, including: cross-document consistency diagnostic processing based on the target data to obtain consistency diagnostic results; compliance diagnostic processing based on the target data and engineering knowledge data to obtain compliance diagnostic results; and spatial rationality pre-diagnostic processing based on the target data to obtain spatial rationality diagnostic results.
[0039] Specifically, the structured information (target data) is fed into the AI intelligent diagnostic engine, which utilizes large model capabilities and a professional knowledge base (engineering knowledge data) to perform multi-dimensional diagnostics. The AI intelligent diagnostic engine first performs semantic understanding processing on the target data and engineering construction knowledge data, and then performs multi-dimensional diagnostic processing on the target data based on the engineering knowledge data, thereby obtaining multi-dimensional diagnostic results. Multi-dimensional diagnostic processing includes cross-document consistency diagnostic processing, compliance diagnostic processing, and spatial rationality pre-diagnostic processing. The multi-dimensional diagnostic results include document consistency risk points corresponding to consistency diagnostic results, compliance risk points corresponding to compliance diagnostic results, and spatial rationality risk points corresponding to spatial rationality diagnostic results.
[0040] In one example, the first target data also includes cable specification information; cross-document consistency diagnostic processing includes quantity comparison processing, model comparison processing, and specification comparison processing; based on the target data, cross-document consistency diagnostic processing is performed to obtain consistency diagnostic results, including: quantity comparison processing based on graphic symbol quantity information and material quantity information to obtain a first diagnostic result; model comparison processing based on equipment model information and text annotation information to obtain a second diagnostic result; specification comparison processing based on cable specification information and text annotation information to obtain a third diagnostic result; based on the first diagnostic result, the second diagnostic result, and the third diagnostic result, a consistency diagnostic result is obtained.
[0041] Specifically, the cross-document consistency diagnosis unit performs cross-document consistency diagnosis processing to obtain consistency diagnosis results. Quantity comparison processing, for example, automatically compares the equipment quantity (graphic symbol quantity information) parsed from the drawing file (e.g., 120 smoke detectors) with the quantity (material quantity information) in the bill of quantities table (e.g., 118), marking discrepancies. Model comparison processing, for example, compares the equipment model information specified in the tender document (e.g., "XX brand A model water pump") with the model information (text annotation information) marked on the drawing (e.g., "XX brand B model water pump"), marking discrepancies. Specification comparison processing, for example, compares the cable specification information described in the tender document (e.g., "ZR-YJV-4*50") with the specifications (text annotation information) on the electrical system diagram, marking conflicts.
[0042] In one example, the target data also includes area data and usage data; compliance diagnosis is performed based on the target data and engineering knowledge data to obtain compliance diagnosis results, including: fire protection code judgment based on area data, usage data and engineering knowledge data to obtain compliance diagnosis results.
[0043] Specifically, the technical parameter compliance diagnostic unit will extract technical parameters from documents (tender documents) and drawings and verify them against national / industry standards in the knowledge base (engineering knowledge data). For example, based on the room area (area data) and usage data parsed from the drawing documents, it will determine whether the arrangement density of fire sprinkler heads or smoke detectors in the room complies with fire protection regulations (engineering knowledge data).
[0044] In one example, the installation project file to be diagnosed targets a first installation project and a second installation project; spatial rationality pre-diagnosis processing is performed based on the target data to obtain spatial rationality diagnosis results, including: performing virtual overlay processing on the target data corresponding to the first installation project and the target data corresponding to the second installation project to perform spatial rationality pre-diagnosis processing on the target data to obtain spatial rationality diagnosis results.
[0045] Specifically, the spatial rationality pre-diagnosis unit virtually overlays pipeline information (graphic symbol information) from drawings of different disciplines (such as HVAC (first installation project) and electrical (second installation project)), and uses algorithms to detect whether there are obvious intersections and collisions on the two-dimensional plane and issue early warnings.
[0046] In one example, the multidimensional diagnostic results include risk points; based on the multidimensional diagnostic results, a risk diagnostic report is obtained, including: based on the risk points, obtaining the risk diagnostic report and generating optimization suggestions for the risk points; in response to the user's operation instructions for the risk diagnostic report, generating prompt information for the installation project file to be diagnosed for the risk points.
[0047] Specifically, the AI intelligent diagnostic engine summarizes all the discovered problems (risk points) and generates a comprehensive diagnostic report (risk diagnostic report), and provides intelligent suggestions (optimization suggestions), such as: (1) Problem list (risk diagnostic report): clearly lists all inconsistencies (document consistency risk points), non-compliance items (compliance risk points) and potential collision points (spatial rationality risk points). (2) Visual positioning: when the user clicks on any problem (risk point) in the (operation instruction) list (risk diagnostic report), it can automatically jump to the corresponding file (tender document paragraph, drawing file location, bill of quantities line) and highlight it (prompt information), realizing "click and see". (3) Intelligent suggestions (optimization suggestions): for inconsistency problems: "Risk (risk point): the number of smoke detectors in fire protection drawing F-03 is 120, but the bill of quantities is 118. Suggestion (optimization suggestion): please review the design and make unified modifications based on one of them." for non-compliance problems: "Risk (Risk Point): The cable trays in the power supply shaft are made of non-fireproof materials, which does not comply with the mandatory provisions of Article X of the 'Code for Fire Protection Design of Buildings' GB50016-2014. Recommendation (Optimization Suggestion): Change the cable tray material to fireproof cable trays." In one example, the engineering document diagnostic method also includes: obtaining reference data from the building information model; and verifying the tender documents and bill of quantities based on the reference data.
[0048] Specifically, for projects that have adopted BIM (Building Information Modeling) forward design, they can directly interface with the BIM model to extract the most accurate 3D information (reference data) of equipment and pipelines. This information can then be used as a benchmark (Ground Truth) to verify the accuracy of the tender documents and bill of quantities, achieving a higher-dimensional consistency guarantee between the model and the documents.
[0049] In one example, the engineering document diagnostic method also includes: obtaining cost data; and when there is a difference between the equipment model information corresponding to the tender document and the text annotation information corresponding to the drawing document, performing a cost difference analysis on the equipment model information and the text annotation information based on the cost data.
[0050] Specifically, the AI intelligent diagnostic engine can incorporate the company's cost database (cost data). When it discovers differences between the tender documents and drawing documents in terms of material (textual annotation information) or equipment (equipment model information), it can not only point out the technical inconsistencies but also automatically estimate the potential cost differences between these two different choices, providing tender decision-makers with a more intuitive business impact analysis.
[0051] Figure 2Another flowchart of an engineering document diagnostic method provided for the implementation of this application is shown below. Figure 2 As shown, the diagnostic methods for engineering documents include S201-S207.
[0052] S201, User Input.
[0053] For example, users upload a series of installation files (including tender documents, drawing files, and bill of quantities).
[0054] S202, Multimodal Information Extraction.
[0055] For example, the text parsing unit performs deep analysis of the tender document text, extracting key technical clauses, specified equipment brands (equipment brand information) and models (equipment model information), construction process requirements, etc. The image / CAD parsing unit utilizes computer vision (CV) and CAD file parsing technology to analyze drawings. It can identify standard electrical, water supply and drainage, fire protection, and other graphic symbols (graphic symbol information), count their quantities (graphic symbol quantity information), and read adjacent textual annotations (such as model, specifications, elevation, etc.). The table parsing unit parses the Bill of Quantities (BOQ) Excel file, converting it into a structured database containing fields such as material name information, specifications (material specification information), units (material unit information), and quantity (material quantity information), using the extracted data as the target data.
[0056] S203, Obtain engineering knowledge data.
[0057] For example, engineering knowledge data includes national and industry standard data (fire protection, electrical, energy conservation, etc.), equipment and material brand data, and historical project issue data.
[0058] S204, AI intelligent diagnostic engine.
[0059] For example, the AI intelligent diagnostic engine first performs semantic understanding processing on the target data and engineering construction knowledge data, and then performs multi-dimensional diagnostic processing on the target data based on the engineering knowledge data to obtain multi-dimensional diagnostic results. The multi-dimensional diagnostic processing includes cross-document consistency diagnostic processing (cross-document consistency diagnostic unit), compliance diagnostic processing (technical parameter compliance diagnostic unit), and spatial rationality pre-diagnostic processing (spatial rationality pre-diagnostic unit). The multi-dimensional diagnostic results include the risk points corresponding to each dimension (document consistency risk points, compliance risk points, and spatial rationality risk points).
[0060] S205, Risk diagnosis report and optimization suggestions are generated.
[0061] For example, the AI-powered intelligent diagnostic engine summarizes all the identified problems (risks) and generates a comprehensive diagnostic report (risk diagnostic report), and provides intelligent suggestions (optimization suggestions). S206, front-end visual report presentation.
[0062] S207, Users view the report and adopt suggestions.
[0063] The engineering document diagnostic method proposed in this application achieves the following: (1) Cross-modal semantic alignment and verification technology of "figure-text-table": It breaks down the data barriers between unstructured text, semi-structured CAD drawings and structured tables. Through AI's deep understanding of multimodal information and entity alignment, it realizes for the first time the automated and integrated consistency verification of the core bidding documents of installation engineering. (2) Deeply embedding domain knowledge into the diagnostic process: It constructs an engineering knowledge base containing national norms, industry standards and enterprise historical data, and uses it as the "brain" of the AI diagnostic engine. This makes the diagnosis no longer a simple information comparison, but an "intelligent review" that incorporates expert experience and compliance requirements, which greatly improves the professionalism and accuracy of the diagnosis. (3) It realizes the transformation from "passive problem discovery" to "proactive risk warning": Through comprehensive diagnosis of quantity, model, specifications, norms and even spatial rationality, it can discover a large number of hidden risks that are easily missed in traditional manual review and will lead to major changes and cost overruns in the later stage, providing a powerful pre-risk control capability for project management.
[0064] Figure 3 A block diagram of an engineering document diagnostic system provided for another embodiment of this application.
[0065] This specification provides an engineering document diagnostic system 300. Please refer to [link / reference]. Figure 3 The engineering document diagnostic system 300 includes: an acquisition module 310, a parsing module 320, a processing module 330, and an acquisition module 340.
[0066] The acquisition module 310 is used to acquire the installation project files and project knowledge data to be diagnosed.
[0067] The parsing module 320 is used to perform in-depth parsing and multimodal information extraction processing on the installation project files to be diagnosed, so as to obtain the target data.
[0068] The processing module 330 is used to perform semantic understanding processing on the target data and engineering knowledge data to perform multi-dimensional diagnostic processing on the target data and obtain multi-dimensional diagnostic results, including consistency diagnostic results, compliance diagnostic results and spatial rationality diagnostic results.
[0069] Module 340 is used to obtain a risk diagnosis report based on the multidimensional diagnostic results.
[0070] For example, the installation project documents to be diagnosed include tender documents, drawing documents, and bill of quantities; the parsing module 320 is also used to perform in-depth parsing and information extraction on the tender documents, and use the key technical clause information, equipment brand information, equipment model information, and construction process requirement information obtained from the information extraction as the first target data; to perform in-depth parsing and information extraction on the drawing documents, and use the graphic symbol information, graphic symbol quantity information, and text annotation information obtained from the information extraction as the second target data; to perform in-depth parsing and information extraction on the bill of quantities, and use it as the first target information; to obtain material name information, material specification information, material unit information, and material quantity information, and use it as the third target data; and to obtain the target data based on the first target data, the second target data, and the third target data.
[0071] For example, the processing module 330 is also used to perform cross-document consistency diagnosis processing based on the target data to obtain consistency diagnosis results; perform compliance diagnosis processing based on the target data and engineering knowledge data to obtain compliance diagnosis results; and perform spatial rationality pre-diagnosis processing based on the target data to obtain spatial rationality diagnosis results.
[0072] For example, the first target data also includes cable specification information; the cross-document consistency diagnosis process includes quantity comparison processing, model comparison processing, and specification comparison processing; the processing module 330 is also used to perform quantity comparison processing based on graphic symbol quantity information and material quantity information to obtain a first diagnosis result; perform model comparison processing based on equipment model information and text annotation information to obtain a second diagnosis result; perform specification comparison processing based on cable specification information and text annotation information to obtain a third diagnosis result; and obtain a consistency diagnosis result based on the first diagnosis result, the second diagnosis result, and the third diagnosis result.
[0073] For example, the target data also includes area data and usage data; the processing module 330 is also used to make fire protection code judgments based on the area data, usage data and engineering knowledge data to obtain compliance diagnosis results.
[0074] For example, the installation project file to be diagnosed is for the first installation project and the second installation project; the processing module 330 is also used to perform virtual overlay processing based on the target data corresponding to the first installation project and the target data corresponding to the second installation project, so as to perform spatial rationality pre-diagnosis processing on the target data and obtain spatial rationality diagnosis results.
[0075] For example, the multidimensional diagnostic results include risk points; the obtaining module 340 is also used to obtain a risk diagnostic report based on the risk points and generate optimization suggestions for the risk points; in response to the user's operation instructions for the risk diagnostic report, it generates prompt information for the installation project file to be diagnosed for the risk points.
[0076] For example, the engineering document diagnostic system 300 further includes: a verification module for acquiring reference data of the building information model; and verifying the tender documents and bill of quantities based on the reference data.
[0077] For example, the engineering document diagnostic system 300 further includes: an analysis module for acquiring cost data; when there is a difference between the equipment model information corresponding to the tender document and the text annotation information corresponding to the drawing document, a cost difference analysis is performed on the equipment model information and the text annotation information based on the cost data.
[0078] Figure 4 A block diagram of an electronic device provided for another embodiment of this application.
[0079] Another embodiment of this application provides an electronic device having a computer program stored thereon, which, when executed by a processor, implements the steps of the method of any of the above embodiments.
[0080] like Figure 4 As shown, for ease of understanding, embodiments of this application illustrate a specific electronic device 400.
[0081] Electronic device 400 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device 400 may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.
[0082] like Figure 4 As shown, the electronic device 400 includes a computing unit 401, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 402 or a computer program loaded from a storage unit 408 into a random access memory (RAM) 403. The RAM 403 may also store various programs and data required for the operation of the electronic device 400. The computing unit 401, ROM 402, and RAM 403 are interconnected via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.
[0083] Multiple components in electronic device 400 are connected to input / output (I / O) interface 405. These components include: input unit 406, such as a keyboard or mouse; output unit 407, such as various types of displays or speakers; storage unit 408, such as a hard disk or optical disk; and communication unit 409, such as a network interface card (NIC), modem, or wireless transceiver. Communication unit 409 allows electronic device 400 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.
[0084] The computing unit 401 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 401 performs the various methods described above. For example, in some embodiments, any one or more of the various methods described above can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 408. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 400 via ROM 402 and / or communication unit 409. When the computer program is loaded into RAM 403 and executed by the computing unit 401, one or more steps of any one or more of the various methods described above can be performed. Alternatively, in other embodiments, the computing unit 401 can be configured to perform any one or more of the various methods described above by any other suitable means (e.g., by means of firmware).
[0085] This application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method in any of the above embodiments.
[0086] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this application, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.
[0087] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.
[0088] In the description of this application, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this application, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.
[0089] In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicating the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.
[0090] Furthermore, the terms "first," "second," etc., used in the embodiments of this application are for descriptive purposes only and should not be construed as indicating or implying relative importance, or implicitly specifying the number of technical features indicated in this embodiment. Therefore, features defined with terms such as "first" and "second" in the embodiments of this application can explicitly or implicitly indicate that the embodiment includes at least one of those features. In the description of this application, the word "multiple" means at least two or more, such as two, three, four, etc., unless otherwise explicitly and specifically defined in the embodiments.
[0091] In this application, unless otherwise explicitly specified or limited in the embodiments, the terms "installation," "connection," "joining," and "fixing" appearing in the embodiments should be interpreted broadly. For example, a connection can be a fixed connection, a detachable connection, or an integral part; it can also be a mechanical connection, an electrical connection, etc. Of course, it can also be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication between two components, or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific implementation.
[0092] In this application, unless otherwise expressly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "on top of," and "over" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.
[0093] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.
Claims
1. A method for diagnosing engineering documents, characterized in that, The method includes: Obtain the installation project files and project knowledge data to be diagnosed; The installation project file to be diagnosed is subjected to deep analysis and multimodal information extraction processing to obtain target data; Semantic understanding processing is performed on the target data and the engineering knowledge data to perform multi-dimensional diagnostic processing on the target data and obtain multi-dimensional diagnostic results, wherein the multi-dimensional diagnostic results include consistency diagnostic results, compliance diagnostic results and spatial rationality diagnostic results; Based on the multidimensional diagnostic results, a risk diagnostic report is obtained.
2. The method according to claim 1, characterized in that, The installation project documents to be diagnosed include tender documents, drawings, and a bill of quantities; the deep analysis and multimodal information extraction processing of the installation project documents to be diagnosed yields target data, including: The tender documents were analyzed in depth and information was extracted. The key technical clauses, equipment brand information, equipment model information and construction process requirements obtained from the information extraction were used as the primary target data. The drawing file is subjected to in-depth analysis and information extraction. The graphic symbol information, graphic symbol quantity information and text annotation information obtained from the information extraction are used as the second target data. The bill of quantities is subjected to in-depth analysis and information extraction as the first target information; material name information, material specification information, material unit information, and material quantity information are obtained as the third target data. Target data is obtained based on the first target data, the second target data, and the third target data.
3. The method according to claim 2, characterized in that, The semantic understanding processing of the target data and the engineering knowledge data, to perform multi-dimensional diagnostic processing on the target data and obtain multi-dimensional diagnostic results, includes: Based on the target data, cross-document consistency diagnostic processing is performed to obtain the consistency diagnostic results; Compliance diagnosis is performed based on the target data and the engineering knowledge data to obtain the compliance diagnosis result; Based on the target data, a spatial rationality pre-diagnosis process is performed to obtain the spatial rationality diagnosis result.
4. The method according to claim 3, characterized in that, The first target data also includes cable specification information; the cross-document consistency diagnostic processing includes quantity comparison processing, model comparison processing, and specification comparison processing; the cross-document consistency diagnostic processing based on the target data to obtain the consistency diagnostic result includes: A quantity comparison process is performed based on the quantity information of the graphic symbols and the quantity information of the materials to obtain a first diagnostic result; A model comparison process is performed based on the device model information and the text label information to obtain a second diagnostic result; Based on the cable specification information and the text annotation information, a specification comparison process is performed to obtain a third diagnostic result; Based on the first diagnostic result, the second diagnostic result, and the third diagnostic result, the consistency diagnostic result is obtained.
5. The method according to claim 3, characterized in that, The target data also includes area data and usage data; the compliance diagnosis process based on the target data and the engineering knowledge data to obtain the compliance diagnosis result includes: Based on the area data, the usage data, and the engineering knowledge data, fire safety regulations are assessed to obtain the compliance diagnosis results.
6. The method according to claim 3, characterized in that, The installation project files to be diagnosed pertain to both the first and second installation projects; the spatial rationality pre-diagnosis processing based on the target data, to obtain the spatial rationality diagnosis results, includes: Virtual overlay processing is performed on the target data corresponding to the first installation project and the target data corresponding to the second installation project to perform spatial rationality pre-diagnosis processing on the target data, and obtain the spatial rationality diagnosis result.
7. The method according to claim 1, characterized in that, The multidimensional diagnostic results include risk points; The risk diagnosis report obtained based on the multidimensional diagnostic results includes: Based on the risk points, a risk diagnosis report is obtained, and optimization suggestions are generated for the risk points. In response to the user's operation command for the risk diagnosis report, a prompt message is generated for the installation project file to be diagnosed for the risk point.
8. The method according to any one of claims 2-6, characterized in that, The method further includes: Obtain reference data for the building information model; The bidding documents and the bill of quantities are verified based on the reference data.
9. The method according to any one of claims 2-6, characterized in that, The method further includes: Obtain cost data; When there is a difference between the equipment model information corresponding to the tender document and the text annotation information corresponding to the drawing document, a cost difference analysis is performed on the equipment model information and the text annotation information based on the cost data.
10. An engineering document diagnostic system, characterized in that, The system includes: The acquisition module is used to acquire the installation project files and project knowledge data to be diagnosed. The parsing module is used to perform deep parsing and multimodal information extraction processing on the installation project file to be diagnosed to obtain target data; The processing module is used to perform semantic understanding processing on the target data and the engineering knowledge data, so as to perform multi-dimensional diagnostic processing on the target data and obtain multi-dimensional diagnostic results, wherein the multi-dimensional diagnostic results include consistency diagnostic results, compliance diagnostic results and spatial rationality diagnostic results; The module is used to obtain a risk diagnosis report based on the multidimensional diagnostic results.