BIM model information auditing method and device and computer program

Through the combination of pre-trained semantic recognition model and attribute value rule library, the BIM model is deeply audited, which solves the problem of insufficient audit depth in the existing technology, and realizes accurate alignment and constraints of data fields and text descriptions, improving audit accuracy and efficiency.

CN118982329AActive Publication Date: 2024-11-19INST OF COMPUTING TECH CHINA ACAD OF RAILWAY SCI +1

Patent Information

Application Number
CN202411271437.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-11
Publication Date
2024-11-19
Estimated Expiration
2044-09-11

AI Technical Summary

Technical Problem

When faced with complex models, the existing BIM model audit software is insufficient in the audit depth and cannot effectively conduct standardized audits of information values, especially for diversified field annotations and text description information.

Method used

The pre-trained semantic recognition model is used to pre-process and semantic extraction of the BIM model to be reviewed. The model attribute value library is constructed by comparing it with the standard attribute field library, and in-depth verification is carried out based on the attribute value rule library to ensure data fields and text. Accurate alignment and constraints described.

Benefits of technology

It realizes the accurate alignment and constraints of data fields at the semantic layer, enriches the audit content, improves the accuracy and efficiency of audits, and adapts to the audit needs of diversified data forms.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118982329A_ABST
    Figure CN118982329A_ABST
Patent Text Reader

Abstract

The invention provides a BIM model information auditing method and device and a computer program, and the method comprises the steps: carrying out the semantic extraction of the name, function description and standard description of each data field in a to-be-audited building information model after the data cleaning through a pre-trained semantic recognition model, and obtaining a semantic extraction result of the to-be-audited building information model; the semantics of the data field names are used for being compared with the semantics of the standard field names summarized by the standard attribute field library, and the data fields contained in the standard attribute field library in the building information model to be audited are extracted to construct the attribute value library of the model to be audited, so that the accurate alignment of the data fields in the semantic layer is realized; furthermore, based on the constraint recorded by the attribute value rule base, the numerical value or description text recorded by each data field in the to-be-audited model attribute value base is audited, the audit items are enriched, and the constraint mode of the numerical value or the text in the data field in the audit process is deepened.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of construction engineering information technology, and in particular to a BIM model information review method, device and computer program. Background Art

[0002] Building Information Modeling (BIM) is a digital model that is based on a three-dimensional model and contains all relevant information about a building. This model can not only help designers better understand and design buildings, but can also be used in construction, operation and maintenance stages.

[0003] At present, model information review mainly relies on manual inspection, which has the problems of heavy workload and insufficient review accuracy. The goal of BIM model automatic review software is to automatically and quickly review BIM models through computers to meet the requirements of standards and specifications, thereby improving work efficiency and quality. The current BIM model review software mainly conducts information integrity review. However, in the face of increasingly complex model BIM model information, the existing review scheme can only perform integrity verification, and the review depth is obviously insufficient. In the face of diverse field annotations and text description information, it is not enough to achieve normative review of information value. Summary of the invention

[0004] In view of this, the embodiments of the present invention provide a BIM model information review method, device and computer program to eliminate or improve one or more defects existing in the prior art and solve the problem in the prior art that inconsistent text expressions in the BIM model cannot be aligned and deeply checked.

[0005] One aspect of the present invention provides a BIM model information review method, the method comprising the following steps:

[0006] Acquire a building information model to be reviewed, wherein the building information model to be reviewed includes geometric information and non-geometric information recorded based on data fields; the geometric information includes location, shape and size information of the building object; the non-geometric information includes material, specification, cost and construction time information of the building object;

[0007] Performing data preprocessing on the building information model to be reviewed, wherein the data preprocessing includes data cleaning and format standardization;

[0008] Obtain a pre-trained semantic recognition model, and perform semantic extraction on the data field names in the pre-processed building information model to be reviewed to obtain corresponding semantics of the field names to be reviewed;

[0009] Acquire a standard attribute field library for a building information model, compare the semantics of the field name to be reviewed with the semantics of the standard field name of each data field in the standard attribute field library, so as to extract the data fields of the building information model to be reviewed that are included in the standard attribute field library, and construct a model attribute value library to be reviewed, wherein the structure of the model attribute value library to be reviewed is the same as that of the standard attribute field library;

[0010] Acquire an attribute value rule library related to the standard attribute field library, and audit the values ​​of each data field in the audit of the attribute value library of the model to be audited based on the intra-field constraint rules and inter-field relationship constraint rules recorded in the attribute value rule library to obtain a target audit result;

[0011] A summary report on whether the target audit results meet the standards is generated based on the target audit results and sent to the target audit port according to the preset link.

[0012] In some embodiments, the non-geometric information further includes the text description information, and the text description information includes a functional description and a specification description of the building;

[0013] Performing semantic extraction on the functional description to obtain the functional semantics to be reviewed, and performing semantic extraction on the specification description to obtain the specification semantics to be reviewed;

[0014] The functional semantics to be audited corresponding to the textual description information in the model attribute value library to be audited are compared for consistency with the standard functional semantics recorded in the standard attribute field library; the standard semantics to be audited corresponding to the textual description information in the model attribute value library to be audited are compared for consistency with the standard standard semantics recorded in the standard attribute field library, and the target audit result is expanded based on the comparison result.

[0015] In some embodiments, the pre-training step of the semantic recognition model includes:

[0016] Acquire a training sample set, the training sample set comprising a plurality of positive samples and negative samples, the positive samples comprising semantically consistent text pairs in the field related to the building information model, the negative samples comprising semantically inconsistent text pairs in the field related to the building information model, and labels of whether the semantics are consistent are added to the positive samples and the negative samples;

[0017] Acquire an initial neural network model, wherein the initial neural network model adopts a skip-garm model and a similarity calculation module, wherein the skip-garm model performs semantic extraction on the text pairs in the positive sample or the negative sample, and adopts the similarity calculation module to calculate the similarity of the text pairs and obtain a prediction result of the consistency comparison;

[0018] The initial neural network model is trained using the training sample set, a cross entropy loss function is calculated according to the prediction result and the label, parameters of the initial neural network model are updated according to the cross entropy loss function, and the updated skip-garm model is used as a semantic recognition model.

[0019] In some embodiments, the intra-field constraint rules include:

[0020] Constraint rules within string fields to check the length, format, value range, uniqueness, mandatory field, and presence of prohibited characters of the string;

[0021] Numeric field constraints to check the range, precision, scale and unit of the numerical value in the data field;

[0022] Constraint rules within enumeration fields to check whether the enumeration value in the data field is included in a preset range;

[0023] Boolean field constraint rules to check the value format, non-emptiness and dependency on the target field of the Boolean value in the data field;

[0024] The constraint rules in the date type field are used to check the format and sequence of the date type data in the data field.

[0025] In some embodiments, the method further includes: calculating the entity length and entity position of the building information model to be reviewed based on the geometric information, and comparing them with the attribute length and attribute position of the standard building information model, generating a position review result and sending it to the target review port according to the preset link.

[0026] In some embodiments, the method also includes: when the target audit result contains a problem data field that does not comply with the field constraint rules and the inter-field relationship constraint rules, searching for other data field contents that generated the problem data field, and regenerating and updating the problem data field.

[0027] In some embodiments, the method is further used to calculate a comprehensive accuracy rate as a review result, including:

[0028] For each item in the target audit result, a weight is assigned according to the audit difficulty and workload, and the calculation formula is:

[0029]

[0030] Among them, ω i represents the weight of the i-th entry, x i represents the building difficulty of the i-th item, yi represents the workload of the i-th item; n represents the number of model files;

[0031] The precision rate is calculated for the target engineering project, and the calculation formula is:

[0032]

[0033] Where P is the precision of the target project, P ij represents the precision of the i-th entry in the j-th model file; m represents the total number of entries in the j-th model file;

[0034] The recall rate is calculated for the target engineering project, and the calculation formula is:

[0035]

[0036] Among them, R is the recall rate of the target project, P ij represents the recall rate of the ith entry in the jth model file; m represents the total number of entries in the jth model file;

[0037] Calculate the comprehensive accuracy F of the target engineering project, the calculation formula is:

[0038]

[0039] On the other hand, the present invention also provides a BIM model information review device, including a processor, a memory, and a computer program / instructions stored in the memory, wherein the processor is used to execute the computer program / instructions, and when the computer program / instructions are executed, the device implements the steps of the above method.

[0040] On the other hand, the present invention further provides a computer-readable storage medium having a computer program / instruction stored thereon, wherein the computer program / instruction implements the steps of the above method when executed by a processor.

[0041] On the other hand, the present invention also provides a computer program product, comprising a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0042] The beneficial effects of the present invention are at least:

[0043] The BIM model information audit method, device and computer program of the present invention, for the building information model to be audited after data cleaning, uses a pre-trained semantic recognition model to perform semantic extraction on the name, function description and specification description of each data field in the building information model to be audited, the semantics of the data field name is used to compare with the semantics of the standard field name summarized in the standard attribute field library, and extracts the data fields contained in the building information model to be audited in the standard attribute field library, constructs an attribute value library of the model to be audited, and realizes accurate alignment of the data fields at the semantic layer; further based on the constraints recorded in the attribute value rule library, the numerical values ​​or description texts recorded in each data field in the attribute value library of the model to be audited are checked, the audit items are enriched, and the constraint mode of the numerical values ​​or texts in the data fields in the audit process is deepened.

[0044] Furthermore, corresponding constraint rules are configured for different forms of numerical values ​​in data fields, which can adapt to the audit requirements of diverse data forms in BIM model audit scenarios and improve efficiency and audit accuracy.

[0045] Additional advantages, purposes, and features of the present invention will be described in part in the following description, and will become apparent to those skilled in the art after studying the following, or may be learned from the practice of the present invention. The purposes and other advantages of the present invention may be achieved and obtained by the structures specifically indicated in the specification and the accompanying drawings.

[0046] Those skilled in the art will appreciate that the objectives and advantages that can be achieved with the present invention are not limited to the above specific description, and the above and other objectives that can be achieved by the present invention will be more clearly understood from the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The drawings described herein are used to provide a further understanding of the present invention, constitute a part of the present application, and do not constitute a limitation of the present invention. In the drawings:

[0048] Figure 1 The present invention is a flowchart of a BIM model information review method according to an embodiment of the present invention.

[0049] Figure 2 The figure is a logical relationship diagram of the BIM model information review method according to an embodiment of the present invention.

[0050] Figure 3 A schematic diagram of the review format of the BIM model information review method according to an embodiment of the present invention.

[0051] Figure 4 This is a software architecture diagram of a BIM model automatic review tool according to another embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0053] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, only structures and / or processing steps closely related to the solutions according to the present invention are shown in the accompanying drawings, while other details that are not closely related to the present invention are omitted.

[0054] It should be emphasized that the term “include / comprises” when used herein refers to the presence of features, elements, steps or components, but does not exclude the presence or addition of one or more other features, elements, steps or components.

[0055] The automated audit of BIM models is an urgent need in this field. The current BIM model audit software mainly conducts information integrity audits. By building a standard requirement database, all the information fields required to be included in various types of components are sorted out. Subsequently, the BIM information is extracted to establish an audit model database, and the information fields that have been added to various models are sorted out. By comparing and verifying the two databases, the missing or incorrect attribute information in the BIM model is reported. However, in the face of the current building information model with diverse data content forms and logical relationships, the difficulty of auditing lies in the fact that in the process of compilation, use and recording by multiple parties, the names and descriptions of some matters are colloquial and non-standardized, and the order and structure of data fields may also be different. This leads to the audit process relying on manual labor. Even if automated auditing is achieved to a certain extent, it cannot cope with the problems of information extraction difficulties caused by non-standardized structure data fields, diverse comparison rules, and audit difficulties caused by non-standard text records.

[0056] The present invention provides a BIM model information review method, such as Figure 1 As shown, the method includes the following steps S101 to S106:

[0057] Step S101: Acquire a building information model to be reviewed, where the building information model to be reviewed includes geometric information and non-geometric information recorded based on data fields; the geometric information includes the location, shape and size information of the building object; the non-geometric information includes the material, specification, cost and construction time information of the building object.

[0058] Step S102: performing data preprocessing on the building information model to be reviewed, wherein the data preprocessing includes data cleaning and format standardization.

[0059] Step S103: obtaining a pre-trained semantic recognition model, performing semantic extraction on the data field names in the pre-processed building information model to be reviewed to obtain the corresponding semantics of the field names to be reviewed.

[0060] Step S104: Refer to Figure 2 , obtain the standard attribute field library about the building information model, compare the semantics of the field name to be reviewed with the standard field name semantics of each data field in the standard attribute field library, so as to extract the data fields contained in the standard attribute field library in the building information model to be reviewed, and construct it as the attribute value library of the model to be reviewed. The structure of the attribute value library of the model to be reviewed is the same as that of the standard attribute field library.

[0061] Step S105: Refer to Figure 2 and Figure 3 , obtain the attribute value rule library related to the standard attribute field library, and audit the values ​​of each data field in the audited model attribute value library based on the intra-field constraint rules and inter-field relationship constraint rules recorded in the attribute value rule library to obtain the target audit result.

[0062] Step S106: Generate a standard compliance summary report based on the target audit results and send it to the target audit port according to the preset link.

[0063] In step S101, the position in the geometric information refers to the specific coordinate position of the building in space, usually expressed in XYZ coordinates. The shape of the building object represents the three-dimensional form of the building, including the specific shape and structure of the walls, roofs, windows, doors, etc. The size represents the specific size of the building or its components, such as length, width, height, etc.

[0064] Among non-geometric information, materials represent the types and properties of materials used in buildings, such as reinforced concrete, glass, wood, etc. Specifications represent the specific specifications of materials, such as the diameter of steel bars, the thickness of glass, etc. Costs represent various cost information related to buildings, including material costs, labor costs, etc. Construction time refers to the time schedule for building construction recorded in the model, including the start time, end time, and progress plan for each stage.

[0065] In step S102, in some embodiments, the data preprocessing includes filling missing data with specified values ​​and performing unit conversion on the values ​​recorded in each data field.

[0066] Specifically, the purpose of data cleaning is to remove or repair erroneous, redundant or inconsistent data in the BIM model to improve the quality and availability of the data. Specifically, it includes the following aspects: 1) Data integrity check, check whether there are missing attributes or information in the model, such as missing geometric information, material properties or component information. Complete or mark the missing data. 2) Data consistency check, ensure that the same type of data uses consistent naming specifications and units, such as the same type of components using the same naming rules and the same unit system (meters, millimeters, etc.). 3) Duplicate data processing, identify and delete duplicate components or redundant data in the model to reduce data redundancy. 4) Error data correction, find and correct errors in the data, such as geometric errors (such as face interlacing, non-closed faces), attribute errors (such as incorrect material properties), etc. 5) Data filtering, according to the audit requirements, filter out the data that needs to be audited and eliminate irrelevant or unnecessary data.

[0067] The purpose of format standardization is to ensure that model data meets certain standards and specifications so that it can be correctly transferred and used between different software and platforms. Specifically, it includes the following: 1) Unit and precision standardization, converting all data in the model to a unified audit system and precision requirements. 2) Attribute name standardization, unifying the attribute names of components, following industry standards or project standards, such as using unified material names, classification codes, etc. 3) File format conversion, converting model data into the file format required for auditing to ensure consistency comparison during the audit process. 4) Geometric accuracy adjustment, adjusting the geometric accuracy of the model according to standards or audit requirements, such as adjusting the level of modeling details, simplifying or refining the geometric structure. 5) Coordinate system and reference point standardization, unifying the coordinate system and reference point settings of the model to ensure that the model can be correctly positioned and aligned in different environments.

[0068] In step S103, in some embodiments, the pre-training step of the semantic recognition model includes S1031 to S1033:

[0069] Step S1031: Obtain a training sample set, which includes multiple positive samples and negative samples. The positive samples include semantically consistent text pairs in the field related to building information modeling, and the negative samples include semantically inconsistent text pairs in the field related to building information modeling. Labels indicating whether the positive samples and the negative samples are semantically consistent are added.

[0070] Step S1032: Obtain an initial neural network model. The initial neural network model uses a skip-garm model and a similarity calculation module. The skip-garm model performs semantic extraction on text pairs in positive samples or negative samples, and uses a similarity calculation module to calculate the similarity of text pairs and obtain a prediction result of consistency comparison.

[0071] Step S1033: Use the training sample set to train the initial neural network model, calculate the cross entropy loss function according to the prediction result and the label, update the parameters of the initial neural network model according to the cross entropy loss function, and use the updated skip-garm model as the semantic recognition model.

[0072] Steps S1031 to S1033 of the present application train a semantic recognition model for technical terms in the field related to building information modeling. The Skip-Gram used in this process is a neural network model for generating word embeddings, which is a representation method for converting words into continuous vectors. These vectors are located in a high-dimensional space and retain the semantic relationship of words to a certain extent. Here, the Skip-Gram model is used to extract semantic features from the text pairs in each positive sample or negative sample. The similarity calculation module receives the semantic expression obtained by the word embedding operation of the Skip-Gram model on the text, and calculates the similarity of the text pairs. Through this similarity, the model can generate a prediction result to determine whether the pair of texts is semantically consistent or inconsistent. In the process of identifying positive and negative samples, a cross entropy loss is constructed to update the model parameters, and the updated Skip-Gram model is used as a semantic recognition model to perform word embedding for semantic extraction and comparison.

[0073] In step S104, based on the semantic extraction, the semantics of the field name to be reviewed are compared with the semantics of the standard field name in the standard attribute field library, and the data fields contained in the standard attribute field library of the building information model to be reviewed are extracted to construct the model attribute value library to be reviewed. This is mainly to align the data fields and keep them consistent with the structure of the standard attribute field library to facilitate subsequent review. In step S105, the attribute value rule library is used to record various constraint rules for the standard attribute field library so as to conduct review according to the corresponding constraint rules. Among them, for the data field in numerical form, the constraint rules in the field are configured as follows:

[0074] Constraint rules within string fields are used to check the length, format, value range, uniqueness, mandatory nature, and presence of prohibited characters of the string.

[0075] Constraint rules within numeric fields are used to check the range, precision, scale, and unit of the values ​​within the data fields.

[0076] Constraint rules within enumeration fields are used to check whether the enumeration value in the data field is included in the preset range.

[0077] Boolean field constraint rules are used to check the value format, non-emptiness, and dependency on the target field of the Boolean value in the data field.

[0078] Constraint rules within date type fields are used to check the format and sequence of date type data within data fields.

[0079] The relationship constraint rules between fields mainly include constraints formed between different data fields based on logical relationships, operational relationships, normative comparison relationships, etc.

[0080] Logical relationship constraints refer to the relationships between different data fields that must be followed based on business logic or actual application scenarios. This type of constraint ensures that the data between fields conforms to specific business logic. For example: conditional dependency, the value of a field determines the optional range or value of another field. Mutual exclusivity, there is a mutual exclusivity relationship between certain fields, that is, when one field has a value, the other field must be empty.

[0081] An operation relation constraint means that different data fields must satisfy a certain mathematical or arithmetic relationship. Such constraints are usually applied to numeric fields to ensure that the operation relation between the data is reasonable. For example, the value of the "Total Amount" field should be equal to the result of multiplying the "Unit Price" by the "Quantity". Or the value of a field must be within the range of another field value, for example, the value of the "Age" field must be greater than or equal to the difference between the "Year of Birth" and the current year.

[0082] Normative comparison constraints refer to the comparison between different data fields based on industry standards, regulations or internal specifications. This constraint ensures that the data meets specific standards or specifications. For example, in the code comparison table, the values ​​of certain fields need to be compared and verified with the standardized code table or reference table. Format consistency, field values ​​must conform to specific formats or specifications, such as the format of postal codes, the format of telephone numbers, etc. If a field value conforms to a specific format, the associated fields must also be adjusted accordingly. For example, if an "ID number" field is 18 digits, then the "date of birth" field should conform to the year information contained in the ID number.

[0083] Furthermore, for the description information in text form, similarity comparison can be performed directly by extracting semantics. Specifically, non-geometric information also includes text description information, which includes functional description and specification description of the building; the functional description is a detailed description of the function of the building, such as whether a building is used as an office building, residence or shopping mall, etc. The specification description is a description of technical specifications related to building design and construction, such as earthquake-resistant design specifications, energy-saving standards, etc.

[0084] Perform semantic extraction on the functional description to obtain the functional semantics to be reviewed, and perform semantic extraction on the specification description to obtain the specification semantics to be reviewed;

[0085] The functional semantics to be audited corresponding to the text description information in the attribute value library of the model to be audited are compared for consistency with the standard functional semantics recorded in the standard attribute field library; the standard semantics to be audited corresponding to the text description information in the attribute value library of the model to be audited are compared for consistency with the standard standard semantics recorded in the standard attribute field library, and the target audit result is expanded based on the comparison result.

[0086] In step S106, based on the review of all constraints recorded in the attribute value rule library in step S105, the audit result of each audit item is obtained, that is, whether it meets or does not meet the requirements, and further statistical analysis is performed to obtain the corresponding audit result.

[0087] In order to integrate the audit results of each item into the final audit result, the following strategies can be adopted to adapt to different audit scenarios and strictness:

[0088] Single item rejection strategy is strict and suitable for high-demand scenarios. If any audit item is "not in compliance with requirements", the final audit result is directly "failed". It is applicable to key areas such as security audit, legal compliance audit, and quality management. The policy rule is that if any audit item is "not in compliance with requirements", the final result is "failed".

[0089] Percentage pass strategy: This strategy allows a certain degree of flexibility and is suitable for looser audit scenarios. If the proportion of passed items to the total items reaches the preset threshold, the final audit result is "passed", otherwise it is "failed". It is suitable for non-critical business audits, comprehensive evaluation systems, etc.

[0090] The weighted scoring strategy assigns different weights to each audit item, calculates the weighted total score, and compares the total score with the set threshold to obtain the final audit result. This strategy is suitable for situations where the importance of different items is inconsistent. It is suitable for scenarios such as multi-dimensional comprehensive evaluation, performance appraisal, and complex system evaluation. The strategy rules include: assigning a weight to each item (such as 0.1, 0.2, 0.3, etc.) and distinguishing them according to their importance; calculating the weighted total score of all items (entries that meet the requirements get full marks, and those that do not meet the requirements get 0 points); setting a passing threshold (such as a total score of 70%), if the weighted total score is greater than or equal to the threshold, the final result is "passed"; otherwise, it is "failed".

[0091] Classification and stratification strategy: categorize items into different levels or layers (such as "critical", "important", "general") according to their importance. Items at different levels have different passing standards. This strategy is suitable for multi-level risk control. It is suitable for risk management of complex systems, project acceptance and other scenarios.

[0092] Custom combination strategies combine some or all of the above strategies according to the needs of specific scenarios. For example, the key items use the "single item veto strategy" and the general items use the "percentage pass strategy", and the two are combined to obtain the final audit result. It is suitable for customized audit scenarios and audit system scenarios that take multiple conditions into consideration.

[0093] In some embodiments, the method further includes step S107: calculating the entity length and entity offset of the building information model to be reviewed based on the geometric information, and comparing them with the attribute length and attribute position of the standard building information model, generating a position review result and sending it to the target review port according to a preset link.

[0094] In some embodiments, the method further includes step S108: when there is a problem data field in the target audit result that does not comply with the field constraint rules and the field relationship constraint rules, searching for other data field contents that generate the problem data field, and regenerating and updating the problem data field.

[0095] In some embodiments, the method is further used to decompose the target engineering project into multiple model files, wherein the model files include multiple sub-building information models, and calculate the comprehensive accuracy rate as the review result, including steps S201 to S204:

[0096] Step S201: Assign weights to each item in the target audit result according to the audit difficulty and workload. The calculation formula is:

[0097]

[0098] Among them, ω i represents the weight of the i-th entry, x i represents the building difficulty of the i-th entry, y i represents the workload of the i-th item; n represents the number of model files.

[0099] Step S202: Calculate the precision rate for the target project, the calculation formula is:

[0100]

[0101] Among them, P is the precision rate of the target engineering project, P ij represents the precision of the ith entry in the jth model file; m represents the total number of entries in the jth model file.

[0102] Step S203: Calculate the recall rate of the target project, and the calculation formula is:

[0103]

[0104] Among them, R is the recall rate of the target project, Pij represents the recall rate of the i-th entry in the j-th model file; m represents the total number of entries in the j-th model file;

[0105] Step S204: Calculate the comprehensive accuracy F of the target project, the calculation formula is:

[0106]

[0107] On the other hand, the present invention also provides a BIM model information review device, including a processor, a memory, and a computer program / instructions stored in the memory, wherein the processor is used to execute the computer program / instructions, and when the computer program / instructions are executed, the device implements the steps of the above method.

[0108] On the other hand, the present invention further provides a computer-readable storage medium having a computer program / instruction stored thereon, wherein the computer program / instruction implements the steps of the above method when executed by a processor.

[0109] On the other hand, the present invention also provides a computer program product, comprising a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0110] The present invention is described below in conjunction with a specific embodiment:

[0111] This embodiment provides a BIM model information review method, which aims to strengthen the model information standardization review and add a reminder function for non-standard BIM model attribute values ​​and logical errors. Compared with traditional BIM model review tools, a model attribute value library and an attribute value rule library are newly added. After completing the attribute field check, the attribute value of the correct attribute field can be checked according to the attribute value rule library.

[0112] Reference Figure 2 The BIM model automatic audit tool contains four modules: standard attribute field library, model attribute field library, attribute value library, and attribute value rule library. Each module works in sequence. The audit software first reads the standard file and creates a standard attribute field library. Then it reads the building information model file to be audited, creates a model attribute field library, and compares the model attributes with the standard fields. Then, the model attributes with correct fields are extracted and saved as the attribute value library. Finally, according to the pre-created attribute value rule library, the information in the attribute value library is checked one by one.

[0113] The overall software architecture is as follows Figure 4As shown in the figure, it is mainly composed of rule base construction sub-tool, semantic model conversion module, BIM model checking engine, quick modification and other modules. The rule base construction sub-tool is used to convert BIM standards into domain rule base. For example, the semantic model conversion module converts BIM model files into semantic models. The BIM model checking engine is used to compare and analyze the domain rule base with the semantic model. The analysis and review results are exported as a model review report. Finally, the quick modification module provides tools to assist in batch improvement of models.

[0114] The functions of BIM model automatic review software mainly include:

[0115] (1) Basic settings for model review:

[0116] The model import module is used to import BIM model files into the software. It provides model data preprocessing functions and can perform operations such as attribute extraction on the imported BIM model data to ensure the accuracy and efficiency of subsequent audits.

[0117] The basic settings of model checking can be adjusted according to project requirements, and the functions implemented include reviewing model selection, selecting and updating standards and specifications, automatic matching of model files and platform structures, result display and manual adjustment, and model filtering.

[0118] The software has a built-in rule base for the project. The setting of the rule base includes three aspects:

[0119] 1.1 Set the field type:

[0120] Field types include: string, numeric, enumeration, Boolean, and time. Each type corresponds to certain constraints.

[0121] String: When the string has special format specifications, fill in the constraint rule number in 'Format Specification' to correspond to the specific string check requirements.

[0122] Numeric type: The value should satisfy the constraints of the 'Numeric Range' column.

[0123] Enumeration type: The property should be within the range of values ​​contained in the 'Enumeration Value' column.

[0124] Boolean: The value should be yes or no.

[0125] Date type: The attribute should be in 8-digit format, in the order of year, month, and day. The format of the grader such as "2024-1-1" and the format of month, day, and year "12012023" are both incorrect.

[0126] 1.2 Set field associations:

[0127] When data fields are directly associated, fill in the relevant fields and relationships in "Data Association Relationship" and "Data Association Constraint". For example, the "Top Forward Dimension" field in Table A is consistent with the 8th field in Table B.

[0128] 1.3 Setting constraints on the same field between different models

[0129] In "Data constraints between different models", fill in the constraints on the same fields between different models, for example, positioning information cannot be repeated.

[0130] User-defined rule base: Projects or customers may have different special requirements. Users can modify the rules based on their own needs on the exported rule base template. Re-import the modified template into the software and review according to the newly defined rules.

[0131] (2) Model review:

[0132] The model review module is one of the core modules for realizing automatic review of BIM models. Users can check the model information completeness, standardization, logic, geometric correctness and other aspects according to the built-in model checking algorithm to ensure that the model meets the requirements of relevant standards and specifications.

[0133] Taking the positioning information check algorithm as an example, the steps are as follows:

[0134] 2.1.1 Filter the auxiliary modeling elements in the model file to exclude the influence of irrelevant elements such as two-dimensional lines and text.

[0135] 2.1.2 Extract 3D model attribute information.

[0136] 2.1.3 According to the grouping specification requirements, filter the construction attributes in the attributes and save the attributes as an external library file for use in geometric calculations.

[0137] 2.1.4 Extract the positioning information from the construction attributes, conduct checks on standardization, consistency, repeatability, symbol rules, etc., save the error results in a report and feed it back to the detailer.

[0138] 2.1.5 Extract the mileage information from the approved positioning information and calculate the information length of the model based on the mileage information

[0139] 2.1.6 Calculate the actual length of the model using the entity length calculation method.

[0140] 2.1.7 Screen the types and lengths, summarize and compile statistics on the models that meet the requirements, and provide feedback to the detailer.

[0141] 2.1.8 At the same time, the geometric length calculated by the algorithm is compared with the information length calculated by the attributes, and the erroneous model with inconsistent lengths is fed back to the deepener.

[0142] Taking the calculation of model mileage and offset accuracy as an example, the algorithm is as follows:

[0143] 2.2.1 Create a three-dimensional space line at the center of the project and export the coordinates of the corresponding pile points at a certain interval.

[0144] 2.2.2 The program reads the coordinates in the excel table to generate points and connects them in series to form curve Q1.

[0145] 2.2.3 Project the three-dimensional curve Q1 onto the horizontal plane (Z value is 0) to generate curve Q2.

[0146] 2.2.4 Extract the center point of the calculation model and record it as P1, and determine the point P2 where P1 is projected onto the horizontal plane (Z value is 0).

[0147] 2.2.5 Determine the point P3 on the Q2 curve that is closest to P2.

[0148] 2.2.6 The mileage post number of the starting point P4 of the Q2 curve is ZH1, and the chord length of P3 from P4 is S, so the converted mileage post number ZH=ZH1+S.

[0149] 2.2.7 The distance between P2 and P3 is L, which is the absolute value of the left and right offset.

[0150] 2.2.8 Determine the tangent vector V1 of the curve Q2 at P3, the vector V2 from P3 to P2, and determine the angle from V1 to V2 in the counterclockwise direction: when the angle is 90 degrees, the point is left-biased, and the offset value is L; when the angle is 270 degrees, the point is right-biased, and the offset value is -L. Perform an audit based on the final offset value.

[0151] (3) Results display:

[0152] According to the built-in BIM standard specifications of the project, the model is fully and automatically checked. The results of each audit can be displayed by category or saved as an external Excel file so that users can trace and review the inspection process, understand the details of each step and the reasons for the audit results.

[0153] (4) Model batch modification:

[0154] The software currently includes a batch modification module that provides automatic repair suggestions for detected errors, helping users quickly correct problems in the model. When there are errors in the positioning information, the correct positioning information can be automatically calculated and reported through the relationship with relevant attribute information such as IFD code and pier number.

[0155] (5) Audit and evaluation:

[0156] The existing technology lacks an objective evaluation method, the audit quality is difficult to assess, there is a perfunctory approach, and the audit results are poor.

[0157] In order to improve the scientificity and effectiveness of the review, this embodiment innovatively introduces a method for calculating the accuracy of the comprehensive evaluation model review. This method comprehensively considers the three key indicators of weight distribution, precision rate and recall rate, and provides an objective and quantifiable feedback mechanism for the review effect through quantitative analysis.

[0158] In this embodiment, the target engineering project is decomposed into multiple model files, each of which contains multiple sub-building information models, and the comprehensive accuracy rate is calculated as the review result, including:

[0159] For each item in the target audit result, a weight is assigned according to the audit difficulty and workload. The calculation formula is:

[0160]

[0161] Among them, ω i represents the weight of the i-th entry, x i represents the building difficulty of the i-th entry, y i represents the workload of the ith entry; n represents the number of model files;

[0162] Calculate the precision rate for the target project, the calculation formula is:

[0163]

[0164] Among them, P is the precision rate of the target engineering project, P ij represents the precision of the ith entry in the jth model file; m represents the total number of audited entries;

[0165] Calculate the recall rate for the target project using the following formula:

[0166]

[0167] Among them, R is the recall rate of the target project, P ij represents the recall rate of the ith entry in the jth model file; m represents the total number of audited entries;

[0168] Calculate the comprehensive accuracy F of the target project, the calculation formula is:

[0169]

[0170] Corresponding to the above method, the present invention also provides an apparatus / system, which includes a computer device, the computer device includes a processor and a memory, the memory stores computer instructions, the processor is used to execute the computer instructions stored in the memory, and when the computer instructions are executed by the processor, the apparatus / system implements the steps of the method described above.

[0171] The embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the aforementioned edge computing server deployment method are implemented. The computer-readable storage medium can be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the technical field.

[0172] In summary, the BIM model information audit method, device and computer program of the present invention, for the building information model to be audited after data cleaning, uses a pre-trained semantic recognition model to perform semantic extraction on the name, function description and specification description of each data field in the building information model to be audited, the semantics of the data field name is used to compare with the semantics of the standard field name summarized in the standard attribute field library, and extracts the data fields contained in the standard attribute field library in the building information model to be audited, and constructs an attribute value library of the model to be audited, thereby achieving accurate alignment of the data fields at the semantic layer; further based on the constraints recorded in the attribute value rule library, the numerical values ​​or description texts recorded in each data field in the attribute value library of the model to be audited are checked, the audit items are enriched, and the constraint mode of the numerical values ​​or texts in the data fields during the audit process is deepened.

[0173] Furthermore, corresponding constraint rules are configured for different forms of numerical values ​​in data fields, which can adapt to the audit requirements of diverse data forms in BIM model audit scenarios and improve efficiency and audit accuracy.

[0174] It should be understood by those skilled in the art that the exemplary components, systems and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software or a combination of the two. Whether it is performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier.

[0175] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.

[0176] In the present invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with features of other embodiments or replace features of other embodiments.

[0177] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the embodiments of the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A BIM model information review method, characterized in that: The method comprises the following steps: Acquire a building information model to be reviewed, wherein the building information model to be reviewed includes geometric information and non-geometric information recorded based on data fields; the geometric information includes location, shape and size information of the building object; the non-geometric information includes material, specification, cost and construction time information of the building object Performing data preprocessing on the building information model to be reviewed, wherein the data preprocessing includes data cleaning and format standardization; Obtain a pre-trained semantic recognition model, and perform semantic extraction on the data field names in the pre-processed building information model to be reviewed to obtain corresponding semantics of the field names to be reviewed; Acquire a standard attribute field library for a building information model, compare the semantics of the field name to be reviewed with the semantics of the standard field name of each data field in the standard attribute field library, so as to extract the data fields of the building information model to be reviewed that are included in the standard attribute field library, and construct a model attribute value library to be reviewed, wherein the structure of the model attribute value library to be reviewed is the same as that of the standard attribute field library; Acquire an attribute value rule library related to the standard attribute field library, and audit the values ​​of each data field in the audit of the attribute value library of the model to be audited based on the intra-field constraint rules and inter-field relationship constraint rules recorded in the attribute value rule library to obtain a target audit result; A standard compliance summary report is generated based on the target audit results and sent to the target audit port according to the preset link.

2. The BIM model information review method according to claim 1, characterized in that: The non-geometric information also includes the text description information, and the text description information includes a functional description and a specification description of the building; Performing semantic extraction on the functional description to obtain the functional semantics to be reviewed, and performing semantic extraction on the specification description to obtain the specification semantics to be reviewed; The functional semantics to be audited corresponding to the textual description information in the model attribute value library to be audited are compared for consistency with the standard functional semantics recorded in the standard attribute field library; the standard semantics to be audited corresponding to the textual description information in the model attribute value library to be audited are compared for consistency with the standard standard semantics recorded in the standard attribute field library, and the target audit result is expanded based on the comparison result.

3. The BIM model information review method according to claim 1, characterized in that: The pre-training steps of the semantic recognition model include: Acquire a training sample set, the training sample set comprising a plurality of positive samples and negative samples, the positive samples comprising semantically consistent text pairs in the field related to the building information model, the negative samples comprising semantically inconsistent text pairs in the field related to the building information model, and labels of whether the semantics are consistent are added to the positive samples and the negative samples; Acquire an initial neural network model, wherein the initial neural network model adopts a skip-garm model and a similarity calculation module, wherein the skip-garm model performs semantic extraction on the text pairs in the positive sample or the negative sample, and adopts the similarity calculation module to calculate the similarity of the text pairs and obtain a prediction result of the consistency comparison; The initial neural network model is trained using the training sample set, a cross entropy loss function is calculated according to the prediction result and the label, parameters of the initial neural network model are updated according to the cross entropy loss function, and the updated skip-garm model is used as a semantic recognition model.

4. The BIM model information review method according to claim 1, characterized in that: The field constraints include: Constraint rules within string fields to check the length, format, value range, uniqueness, mandatory field, and presence of prohibited characters of the string; Numeric field constraints to check the range, precision, scale and unit of the numerical value in the data field; Constraint rules within enumeration fields to check whether the enumeration value in the data field is included in a preset range; Boolean field constraint rules to check the value format, non-emptiness and dependency on the target field of the Boolean value in the data field; The constraint rules in the date type field are used to check the format and sequence of the date type data in the data field.

5. The BIM model information review method according to claim 1, characterized in that: The method further includes: calculating the entity length and entity position of the building information model to be reviewed according to the geometric information, and comparing them with the attribute length and attribute position of the standard building information model, generating a position review result and sending it to the target review port according to the preset link.

6. The BIM model information review method according to claim 1, characterized in that: The method also includes: when the target audit result has a problem data field that does not comply with the field constraint rules and the inter-field relationship constraint rules, searching for other data field contents that generated the problem data field, and regenerating and updating the problem data field.

7. The BIM model information review method according to claim 6, characterized in that: The method is also used to calculate the overall accuracy as the review result, including: For each item in the target audit result, a weight is assigned according to the audit difficulty and workload, and the calculation formula is: Among them, ω i represents the weight of the i-th entry, x i Indicates the review difficulty of the i-th item, y i represents the workload of the i-th item; n represents the number of model files; The precision rate is calculated for the target engineering project, and the calculation formula is: Where P is the precision of the target project, P ij represents the precision of the ith entry in the jth model file; m represents the total number of audited entries; The recall rate is calculated for the target engineering project, and the calculation formula is: Among them, R is the recall rate of the target project, P ij represents the recall rate of the ith entry in the jth model file; m represents the total number of audited entries; Calculate the comprehensive accuracy F of the target engineering project, the calculation formula is:

8. A BIM model information review device, comprising a processor, a memory, and a computer program / instruction stored in the memory, characterized in that: The processor is used to execute the computer program / instructions. When the computer program / instructions are executed, the device implements the steps of the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method as claimed in any one of claims 1 to 7 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Citation Information

Patent Citations

  • Constructional engineering design semantic structured description and compliance review method and system

    CN116186831A

  • Method and apparatus for automatically determining compliance with building regulations

    US20090125283A1

Cited By

  • Form file generation method and device, electronic equipment and medium

    CN119761326A

  • Water engineering construction drawing review method and related equipment

    CN120952717A

  • Digital delivery management platform and method based on BIM (Building Information Modeling) technology

    CN121094816A

  • Method and system for automatically marking and checking attribute size of 3D (three-dimensional) model

    CN121615196A