Engineering quality evaluation archive management method based on digital twinning

Abstract

The invention relates to an engineering quality evaluation archive management method based on digital twinning, which belongs to the technical field of engineering quality management, and comprises the following steps of: integrating a BIM (Building Information Modeling) (modeling each component and coding to construct a digital twinning mapping relation), an AR (Augmented Reality) technology (realizing double superposition of the model and display) and cloud storage; according to the invention, intelligentization of constructional engineering quality evaluation and digitalization and convenience of quality acceptance data management are realized, the efficiency and level of engineering quality management are effectively improved, engineering problems can be early warned in time in the construction process, detection and repair can be carried out in time, and the reliability and traceability of engineering quality are guaranteed; in addition, according to the scheme, potential hazards can be warned in advance according to data in the project quality verification and evaluation file management method.

Description

Technical Field

[0001] The present invention belongs to the technical field of engineering quality management, and particularly relates to a method for managing engineering quality inspection and evaluation archives based on digital twin. Background Art

[0002] In modern engineering construction, with the continuous expansion of project scale and increasing complexity, the traditional methods of engineering quality inspection, evaluation, and data management are difficult to meet the requirements of efficient and accurate management. On the one hand, during the on-site quality inspection and evaluation process, construction workers, supervisors, and managers often need to frequently check against a large number of paper drawings or two-dimensional electronic drawings, making it difficult to intuitively and quickly discover the differences between the actual construction and the design model, resulting in low inspection and evaluation efficiency and prone to omissions. On the other hand, the processes of filling, reviewing, and archiving quality inspection data rely mostly on manual offline operations, which not only consume a large amount of time and manpower but also are prone to problems such as data loss, inconsistent information, and difficult searching, seriously affecting the efficiency and quality of project management.

[0003] In recent years, the development of Building Information Modeling (BIM) technology, Augmented Reality (AR) technology, and cloud computing technology has provided new ideas and means to solve these problems. BIM technology can create a digital three-dimensional building model and integrate various types of project information, but it still has deficiencies in combining with the actual site for quality inspection and evaluation. AR technology can integrate the virtual BIM model with the real scene, enhancing the intuitiveness and accuracy of on-site inspection and evaluation, but lacks effective integration with a complete data management and intelligent analysis system. The filling and archiving of quality inspection data also rely on manual sorting, with low efficiency and difficult data searching and tracing. With the development of digital technology, although some electronic data management systems have emerged, they still have deficiencies in real-time acquisition and enhanced display of on-site information, remote collaborative filling and review, and secure and efficient data archiving, making it difficult to meet the requirements of modern building engineering quality management.

[0004] Therefore, it is of great practical significance to develop a system that integrates BIM technology, AR quality inspection and evaluation technology, online filling of quality inspection data, cloud archiving, management, and intelligent analysis, which can effectively improve the level of engineering quality control and promote the digital and intelligent development of the engineering construction industry. Summary of the Invention

[0005] In view of the deficiencies of the prior art, the technical problem to be solved by the present invention is to provide a method for managing engineering quality inspection and evaluation archives based on digital twin, so as to solve the problems of low efficiency, poor accuracy, and inconvenient data management existing in the current engineering quality inspection, evaluation, and acceptance data management, and realize the efficient control of engineering quality and the digital and intelligent management of data.

[0006] To solve the above technical problems, the technical solution adopted by the present invention is: a method for managing engineering quality inspection and evaluation files based on digital twin, including the following steps:

[0007] S1. Create a parametric BIM model, construct a digital twin mapping relationship, import the model into AR software, and perform scene configuration and settings on the BIM model in the AR software;

[0008] S2. After each construction process is completed or each inspection lot is completed, use a mobile terminal device installed with an AR program to scan the construction target, realize the double superposition of the AR model and the BIM model, record on-site quality problems or areas where the model comparison is inconsistent, and generate a quality inspection and evaluation report and a rectification notice;

[0009] S3. After each construction process or each inspection lot is completed or each sub-project construction is completed, perform scanning and comparison through step 2, record the quality acceptance materials, and upload them to the cloud. The cloud system generates a unique identification code and an electronic visa according to the differences in the construction of each process, inspection lot or sub-project;

[0010] S4. Construct a unified data platform based on the above data;

[0011] The digital twin mapping relationship is obtained by clarifying the size, material and steel bar configuration information of the components, and assigning a unique code to each component. The corresponding code corresponds to the attribute information of the component. The attribute information includes the material, specification, manufacturer and installation date of the component;

[0012] The content of the data platform in step S4 includes statistical analysis and calculation of the mean value, standard deviation and frequency of quality indicators at each stage;

[0013] The data platform in step S4 can integrate and analyze the BIM model data, AR inspection and evaluation data, and quality acceptance material data according to a predetermined cycle, so as to generate a quality trend chart, a problem distribution bar chart and a rectification efficiency analysis table;

[0014] The data platform in step S4 can give an early warning when the statistical quantity of any quality problem exceeds a preset threshold. The set thresholds for early warning include a primary threshold and a secondary threshold. The primary threshold and the secondary threshold are triggered independently of each other. The primary threshold includes the lower limit of concrete strength and the upper limit of steel bar spacing deviation. When the actual data breaks through the threshold, an immediate primary warning is triggered; The secondary threshold includes the incidence rate of quality problems and the rectification delay rate. When the incidence rate of quality problems and the rectification delay rate exceed the set value, a secondary warning is immediately given.

[0015] Furthermore, the coding rules for assigning unique codes to each component in the digital twin mapping relationship are as follows: table code, major category code, middle category code, minor category code, detailed category code, primary expansion category code and secondary expansion category code.

[0016] 3. The engineering quality inspection and evaluation file management method based on digital twin according to claim 1 is characterized in that: in the step S2, the mobile terminal device installed with the AR program captures the phenomenon graph by means of shooting, identifies the components and construction parts on site by using the methods of image recognition and spatial positioning, and then displays them on the display screen of the mobile terminal device. Moreover, the scanned model is superimposed and displayed with the BIM model, and the two models are distinguished by colors.

[0017] Furthermore, the scene configuration in step S1 includes identification and positioning configuration;

[0018] The identification and positioning configuration is realized by means of QR code assignment positioning or two-point marking positioning;

[0019] For QR code assignment positioning, after the bim model is imported into the AR software, the bim model is partitioned into components. Each area is composed of a sectional frame. In each sectional frame, a component with high recognition is selected, and a QR code is set beside it. Moreover, the distance between the QR code and the component and the distance between the QR code and the ground are recorded, and the QR code is printed. Wait for the construction of the target area to be completed, and paste the QR code to the specified position according to the recorded distance between the QR code and the component and the distance between the QR code and the ground. The positioning is realized according to the actual QR code scanned by the mobile terminal device;

[0020] For two-point marking positioning, after the bim model is imported into the AR software, the bim model is partitioned into components. Each area is composed of a sectional frame. Two points are selected and marked in the bim model within the sectional frame and recorded. Moreover, the actual positions of the two points are found in reality and selected and recorded by using the mobile terminal device, so as to realize positioning.

[0021] Furthermore, in step S4, entering the data platform performs data extraction, cleaning and transformation through the Extract tool, Transform tool and Load tool, eliminates data noise, eliminates the inconsistency of the same data, and is organized and stored according to the engineering structure, time series and professional categories to construct an engineering quality database.

[0022] Furthermore, in step S4, statistical analysis and data mining are performed on the engineering quality database;

[0023] Statistical analysis is used to calculate the mean value, standard deviation, frequency, average strength of concrete components and steel bar installation deviation rate of quality indicators for each specialty and each construction stage;

[0024] Data mining refers to using clustering analysis to identify areas or processes where quality problems frequently occur, and time series analysis to monitor the fluctuation trend of quality indicators over time, and to find the hidden relationships between quality problems and construction techniques or material suppliers;

[0025] The data mining in step S4 is analyzed and processed by SPSS data analysis software or Tableau data analysis software.

[0026] Further, the quality trend chart in step S4 is a line chart or a bar chart with time as the horizontal axis and quality indicators as the vertical axis; the quality indicators include the qualification rate of each inspection lot and the scores of key processes;

[0027] The problem distribution bar chart is divided according to profession, floor or construction area. The professions include building fabrication, electromechanical systems, and decoration systems. The occurrence frequencies of various quality problems are counted and displayed in the form of bars;

[0028] The rectification efficiency analysis form includes the problem discovery time, the rectification notice issuance time, the actual rectification completion time, the rectification responsible person, and the rectification effect evaluation.

[0029] Further, after the set threshold of the early warning is triggered, the staff conducts on-site verification, scans the target component again through a mobile terminal device installed with an AR program, compares it with the BIM model in the mobile terminal device, re-checks and verifies the location where the problem appears, and conducts rectification and information recording according to the BIM model.

[0030] Further, an AI machine learning model is introduced into the data platform in step S4 to analyze historical quality data, predict potential risk processes for unfinished construction processes and ongoing construction processes, and generate optimized construction plans.

[0031] Further, the specific analysis method for the AI machine learning model to analyze historical quality data is as follows: Before construction, temperature sensors, humidity sensors, and vibration sensors are installed in the construction area. After construction, the data of the defective areas are integrated, combined with the detection values of the temperature sensors, humidity sensors, and vibration sensors, to construct a time series database of defective areas, and analyze whether the defects are caused by human factors or environmental factors.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] The present invention integrates the BIM model, AR technology, and cloud storage, realizes the intelligentization of building engineering quality inspection and evaluation, the digitization and convenience of quality inspection data management, effectively improves the efficiency and level of engineering quality management, and ensures the reliability and traceability of engineering quality. Description of the Drawings

[0034] Figure 1 This is a flowchart of a method for engineering quality assessment and acceptance data management based on digital twin in the present invention. Detailed implementation manners

[0035] To make the above features and advantages of the present invention more obvious and understandable, specific embodiments are hereinafter given and described in detail in conjunction with the accompanying drawings as follows.

[0036] As Figure 1 shown, this embodiment provides a system for engineering quality assessment and acceptance data management based on digital twin, including a BIM model management module, an AR assessment module, a cloud management module for quality acceptance data, and a data integration and management module.

[0037] The BIM model is established using SketchUp / Autodesk Revit / Rhino / 3dMax / Maya. These software can all establish accurate BIM models according to real sizes. The BIM model management module is responsible for establishing and maintaining the BIM model of the project, and deepening the model according to different stages and professional requirements of the project. For example, in the architectural structure specialty, the model is split and refined according to requirements such as inspection lot division plans, bill of quantities, and measurement specifications to ensure that the model can accurately reflect the actual situation of the project and provide a basis for subsequent quality assessment and measurement. During the model deepening process, relevant standards and specifications are strictly followed, and components are accurately modeled and property defined. For example, for components such as columns, beams, and slabs of concrete structures, their size, material, strength and other property information are clarified.

[0038] The AR software of the AR assessment module uses ARkit / ARcore / vuforia / unity3d / easyar, and the AR device of the AR assessment module uses a head-mounted augmented device / vision glasses / handheld monitoring device. In this embodiment, a handheld monitoring device (that is, a mobile terminal device) is adopted. The AR assessment module realizes the rapid inspection of the consistency between the real model and the virtual model by integrating the BIM model with the construction site with the help of AR technology. The on-site management personnel load the AR assessment software through the mobile terminal and scan the actual components at the construction site. The software can automatically overlay and display the corresponding component information in the BIM model on the terminal screen, and intuitively compare the deviations of key information such as the size, position, and reserved holes between the model and the actual components. Once it is found that the deviation exceeds the preset threshold, the system automatically generates a rectification order and pushes it to the mobile terminal of the relevant responsible person through the network, and at the same time records the detailed information of the problem, including the position, type, discovery time, etc., for subsequent traceability and statistical analysis.

[0039] The cloud management module for quality acceptance data includes building a cloud system to realize the functions of online creation, editing, review, archiving, and retrieval of quality acceptance data. The platform provides a rich template library for data, covering standard forms and document formats such as construction quality acceptance codes, inspection lot acceptance records, sub-project acceptance records, division project acceptance records, and unit project completion acceptance reports for various projects. Users can select appropriate templates according to the actual situation of the project to fill in the data.

[0040] During the process of filling in the data, the system automatically associates relevant information in the BIM model, such as the component number, name, attributes, and the affiliated part, as well as the acceptance evaluation results and problem rectification situations in the quality acceptance and evaluation process, to ensure the accuracy and integrity of the data. For example, when the user fills in the acceptance data of a certain inspection lot, the system automatically extracts the component information involved in this inspection lot from the BIM model and fills in the corresponding acceptance data according to the acceptance evaluation results, such as the number of qualified items, the number of unqualified items, and the rectification situation. At the same time, it supports users to upload various attachments, such as construction drawings, material inspection reports, test and detection data, and concealed project acceptance records, to enrich the content of the data.

[0041] In the data review link, the system supports a multi-level review process, and different levels of review personnel (such as quality inspectors, technical responsible persons of the construction unit, supervision engineers, project leaders of the construction unit, etc.) and review authorities can be set according to the project management requirements. The review personnel conduct online reviews of the submitted data on the cloud platform, can view the detailed content of the data, the associated BIM model information and acceptance evaluation records, make annotations and return the data with problems for modification, and track the review progress in real time.

[0042] The data integration and management module is responsible for integrating BIM model data, AR acceptance evaluation data, and quality acceptance data, and establishing an engineering quality database. The quality acceptance data that has passed the review is automatically stamped with an electronic signature and archived and stored, and is classified and managed according to dimensions such as project classification, construction stage, and data type, which is convenient for subsequent query and statistical analysis. Through in-depth analysis of these data, potential quality problems and risk trends are mined to provide decision-making support for engineering quality management. For example, count the incidence of quality problems in different specialties and different construction stages, analyze the correlation between quality problems and construction techniques and material suppliers, generate quality reports and warning information, and remind management personnel to take timely measures to prevent and solve quality problems.

[0043] As the core data repository of the system, the cloud data storage and management center has powerful data storage and management capabilities. It receives and stores data from the AR on-site auxiliary inspection and evaluation module and the mobile terminal data filling and review module, classifies, organizes, and backs up the data. Using cloud computing technology, it can achieve rapid data retrieval and invocation, providing efficient data services for all parties involved in the project. At the same time, the cloud center is also responsible for managing user permissions and data access control to ensure the security and confidentiality of the data.

[0044] The data security encryption and verification module uses advanced encryption algorithms (such as AES, RSA, etc.) to encrypt the quality acceptance data uploaded to the cloud to prevent the data from being stolen or tampered with during transmission and storage. During the data archiving and invocation process, the integrity of the data is verified through hash algorithms (such as MD5, SHA, etc.) to ensure the authenticity and reliability of the data. At the same time, this module also records the operation logs of the data, including the time, personnel, and content of operations such as uploading, modifying, reviewing, and downloading, for easy traceability and auditing.

[0045] In this embodiment, a method for engineering quality inspection and evaluation based on digital twin is proposed for a digital twin-based engineering quality inspection and acceptance data management system, including the following steps:

[0046] S1. Create a parametric BIM model, construct a digital twin mapping relationship, and import the model into the AR software, and perform scene configuration and settings on the BIM model in the AR software;

[0047] S2. After each construction process is completed or after each inspection lot, use a mobile terminal device with an installed AR program to scan the construction target to achieve the double superposition of the AR model and the BIM model, record on-site quality problems or areas where the model comparison is inconsistent, and generate a quality inspection and evaluation report and a rectification notice.

[0048] S3. After each construction process, each inspection lot is completed, or after each sub-project construction is completed, perform scanning and comparison through step 2, record the quality acceptance data, and upload it to the cloud. The cloud system generates a unique identification code and an electronic visa according to the differences in the construction of each process, inspection lot, or sub-project.

[0049] S4. Construct a unified data platform (i.e., the engineering quality database) based on the above data.

[0050] Furthermore, the digital twin mapping relationship in step S1 is formed by specifying the dimensions, materials, and steel bar configuration information of components and assigning a unique code to each component. The corresponding code corresponds to the attribute information of the component. The attribute information includes the material, specification, manufacturer, and installation date of the component. The coding rules for assigning unique codes to each component in the digital twin mapping relationship are as follows: table code, major category code, medium category code, minor category code, detailed category code, primary expansion category code, and secondary expansion category code.

[0051] A specific coding example is as follows:

[0052]

[0053] Describe a single category. If the classification rules are complex, expansion can be carried out after the detailed category code, and the expansion category code should not exceed two levels.

[0054] The specific implementation method of step S1 is as follows: Designers use professional BIM modeling software (SketchUp / Autodesk Revit / Rhino / 3dMax / Maya) to create an initial BIM model of the project according to the engineering design drawings and specification requirements. During the modeling process, fully consider the needs of subsequent quality inspection and evaluation and construction management, and reasonably refine and optimize the model to ensure that the model can accurately reflect the actual situation of the project.

[0055] For example, conduct detailed modeling of the key parts of the building structure (such as beam-column joints, connection parts between the foundation and the main body, etc.), and clarify the information such as the dimensions, materials, and steel bar configuration of the components; accurately design the pipeline routing and equipment installation positions of the mechanical and electrical systems, and reserve sufficient space and interface information. At the same time, assign a unique code and rich attribute information to each component in the model, such as the material, specification, manufacturer, installation date, etc. of the component, for subsequent data management and query.

[0056] Import the created BIM model into the AR development platform or plug-in to configure and set the AR scene. This includes defining parameters such as the display method, color, and transparency of the model in the AR view, as well as setting the reference points and algorithms for image recognition and spatial positioning to ensure that the BIM model can accurately match and overlay with the actual scene at the construction site. At the same time, conduct functional testing and optimization on the AR application program to ensure its running stability and fluency on different mobile terminal devices.

[0057] Furthermore, the mobile terminal device installed with the AR program in step S2 captures the scene graphics through photography, uses image recognition and spatial positioning methods to identify the components and construction parts on site, and then displays them on the display screen of the mobile terminal device. Moreover, the scanned model is overlaid and displayed with the BIM model, and the two models are distinguished by color.

[0058] The specific steps of step S2 are as follows: After the construction personnel complete a certain construction process or inspection lot at the construction site, they use a mobile terminal device installed with an AR application to scan the construction area. The AR application captures on-site images through the camera, uses image recognition and spatial positioning technologies to identify the components and construction parts on-site, and superimposes and displays the corresponding BIM model information on the mobile terminal screen in real time. The on-site personnel check the quality and conduct inspection and evaluation by comparing the model in the AR view with the actual construction situation.

[0059] During the inspection and evaluation process, if quality problems or situations inconsistent with the model are found, the on-site personnel directly mark and record the problem parts in the AR view through the mobile terminal, describe in detail information such as the type, location, and severity of the problems, and take photos or videos as attachments. At the same time, the system automatically associates the inspection and evaluation information with the corresponding components and parts in the BIM model through the unique component code to ensure the traceability of the problems. For example, when it is found that the masonry size of a certain wall does not conform to the BIM model, the system automatically records the number and location information of the wall in the model and associates the inspection and evaluation problem with the BIM model data of the wall for convenient subsequent query and rectification.

[0060] According to the inspection and evaluation results, the system automatically generates a quality inspection and evaluation report and a rectification notice, specifying the rectification requirements, rectification period, and responsible person. The rectification notice is synchronized to the cloud platform and pushed to the mobile terminal or computer terminal of the relevant responsible person through the system to remind them to carry out rectification in a timely manner. After the on-site personnel complete the rectification, they use the AR application again for re-inspection, confirm that the problem has been solved, and upload the re-inspection results to the cloud system to update the inspection and evaluation records and BIM model information.

[0061] Preferably, the scene configuration in step S1 includes identification and positioning configuration; the identification and positioning configuration is implemented by using the method of QR code assignment positioning or two-point marking positioning.

[0062] Among them, for the first type, in the QR code assignment positioning, after the bim model is imported into the AR software, the bim model is partitioned into components. Each area is composed of a section box. In each section box, a component with high recognition is selected, a QR code is set beside it, and the distance between the QR code and the component and the distance between the QR code and the ground are recorded. Then the QR code is printed. After the construction of the target area is completed, the QR code is posted to the specified position according to the recorded distance between the QR code and the component and the distance between the QR code and the ground. The positioning is achieved by scanning the actual QR code with the mobile terminal device.

[0063] Example of QR code assignment positioning application: In an AR software with a BIM model, select a corner of a wall in the selected section box. Set a QR code on any side wall, record the distance between the center of the QR code and the perpendicular line of the corner, record the distance between the center of the QR code and the ground, and then print the QR code. After the construction of the section box is completed, paste the QR code at this position, and the pasting position of the QR code is the same as the recorded distance above; during subsequent inspections, when the staff detects the components in the section box, they need to locate their positions. They only need to find the QR code and scan it with a mobile terminal device. The AR software with a BIM model can obtain the position information of the personnel, and through the information transmitted by the accelerator sensor and gyroscope sensor in the mobile terminal device during subsequent movements, the person in the AR software with a BIM model can move following the actual movement.

[0064] For the second method, for two-point marking positioning, after the BIM model is imported into the AR software, the BIM model is partitioned into components. Each area is composed of a section box. Select two points in the BIM model within the section box for marking and recording, and then find the actual positions of these two points in reality and use a mobile terminal device to select and record them, thereby achieving positioning.

[0065] Example of two-point marking positioning application: In an app with a BIM model, select two adjacent corners of a wall or two adjacent columns, then find this place in reality, and then scan the two corners of the wall or the two columns with a mobile terminal device respectively and correspond them to the corresponding areas in the BIM model. The AR software with a BIM model can obtain the position information of the personnel, and through the information transmitted by the accelerator sensor and gyroscope sensor in the mobile terminal device during subsequent movements, the person in the AR software with a BIM model can move following the actual movement.

[0066] The content of the data platform in step S4 includes statistical analysis and calculation of the mean, standard deviation, and frequency of quality indicators at each stage.

[0067] The data platform in step S4 can integrate and analyze BIM model data, AR inspection and evaluation data, and quality acceptance data according to a predetermined cycle, thereby generating a quality trend chart, a problem distribution bar chart, and a rectification efficiency analysis table.

[0068] The data platform in step S4 can give an early warning when the statistical quantity of any quality problem exceeds a preset threshold.

[0069] The data platform in step S4 has functions of entity measurement and space measurement within the visible range, and this data platform can realize virtual reality visual management. The BIM model can hide components such as floor slabs when comparing with the entity, which is convenient for viewing component models such as mechanical and electrical pipelines passing through several floors.

[0070] Furthermore, the specific steps of Step S3 are as follows: After the construction personnel complete the construction of each inspection lot or sub-project and pass the AR-assisted quality inspection and evaluation, they log in to the cloud-based quality acceptance data management platform, select the corresponding project location, construction stage, and data type, and start filling in the quality acceptance data. The platform automatically loads the corresponding data templates (such as the "GD-C5-71162 Quality Acceptance Record for Steel Bar Installation Inspection Lot" form, the "GD-C5-71165 Quality Acceptance Record for Concrete Construction Inspection Lot" form, etc.) according to the sub-categories of the inspected components (such as: "exterior basement wall", "rectangular column", "curtain wall - channel steel", etc.), and associates the BIM model components and AR inspection and evaluation record data to provide pre-filled information and data reference for users. During the filling process, users only need to supplement key information such as the actual construction data and inspection results, such as the actual consumption of materials, the implementation of construction techniques, the qualified quantity of inspection lots, etc. During the filling process, the relevant component information in the BIM model can be viewed by clicking on the link at any time to ensure that the data is consistent with the actual construction situation. After the data filling is completed, the user can submit an online review and approval application, and the system automatically pushes the data to the relevant reviewers for review and approval according to the preset review process. The reviewers view the data content on the cloud platform, compare the BIM model information and inspection and evaluation records, and review the accuracy, integrity, and standardization of the data. If any problems are found, the reviewers make annotations on the platform and return it for modification, and at the same time the system automatically notifies the data fillers to make rectifications. The approved data is automatically archived and stored, and the system generates a unique identification code and electronic signature for each piece of data to ensure the authenticity and legality of the data. And it is classified and managed according to methods such as project classification and time sequence for convenient subsequent query and statistical analysis.

[0071] Further, the specific steps of step S4 are as follows: 1. Data integration analysis starts with building a unified data platform, aggregating multi-source data including the geometric, attribute, and spatial relationship information of building components from the BIM model, the on-site actual and model deviation data obtained from AR inspection and evaluation, and the construction process records, inspection results, and rectification details covered in the quality acceptance materials. Use ETL (Extract, Transform, Load) tools for data extraction, cleaning, and transformation, eliminate data noise and inconsistencies, organize and store according to dimensions such as engineering structure, time series, and professional category, and build an engineering quality data warehouse. In the analysis stage, a combination of statistical analysis and data mining algorithms is adopted. Statistical analysis calculates the means, standard deviations, frequencies, etc. of quality indicators for each specialty and construction stage, such as the average strength of concrete components and the statistics of steel bar installation deviation rates; data mining uses association rule mining to explore the hidden connections between quality problems and construction processes and material suppliers, clustering analysis to identify areas or processes with frequent quality problems, and time series analysis to monitor the fluctuation trends of quality indicators over time. With the help of professional data analysis software (such as SPSS, Tableau) and custom-developed analysis modules, in-depth data analysis and visualization (such as PowerBI) display are achieved. Among them, for quantitative quality problems, such as the number and spacing of steel bars and structural dimensions, the differences between the AR actual measurement data and the corresponding specifications are used for determination; for qualitative quality problems, such as the surface appearance of concrete, technical means such as 3D scanning are relied on for auxiliary judgment. When dealing with these quality problems, the keywords in their descriptions are extracted and classified and statistically analyzed according to the problem types.

[0072] 2. The system will integrate and deeply analyze the BIM model data, AR inspection and evaluation data, and quality acceptance data at regular intervals. During this process, the above data integration and analysis methods are used to generate various types of quality reports and charts, including quality trend charts, problem distribution bar charts, and rectification efficiency analysis tables. Quality trend chart: A line chart or bar chart is drawn with time on the horizontal axis and quality indicators (such as the passing rate of each inspection lot and the score of key processes) on the vertical axis. It intuitively presents the dynamic changes in the quality of the overall project or specific parts over time, helping to grasp the quality development trend and judge whether it is steadily improving or there are abnormal fluctuations. For example, if the passing rate of the wall plastering inspection lot continues to decline during a certain period, it indicates that there may be systematic problems in this process that need to be further investigated. Problem distribution bar chart: It is divided according to specialties (structure, electrical and mechanical, decoration, etc.), floors, or construction areas, and the occurrence frequencies of various quality problems are counted and presented in the form of bars. It clearly shows the distribution pattern of quality problems in the project, clarifying the key control areas or specialties. For example, if column concrete defects occur concentratedly in the low floors during the structural construction stage of a high-rise building, this area can be focused on to analyze whether there are defects in the formwork and pouring process. Rectification efficiency analysis table: It includes fields such as the problem discovery time, rectification notice issuance time, actual rectification completion time, rectification responsible person, and rectification effect evaluation. Calculate indicators such as the average rectification duration and on-time rectification rate, and present them in tabular form. It accurately evaluates the efficiency and effect of the rectification work, reflecting the construction team's response and solution capabilities to quality problems. If the average rectification duration of a certain construction team far exceeds the expectation, it is necessary to strengthen supervision and management or provide technical support.

[0073] 3. Once the statistical number of a certain type of quality problems exceeds the pre-set threshold, the system will automatically issue an early warning to the relevant responsible persons and managers with corresponding authority (such as the construction party, the supervision party, etc.). For example, when the incidence of quality problems in a certain profession shows a significant upward trend within a certain period of time, the system will immediately issue an early warning notification. This can prompt managers to quickly increase the supervision and inspection of the construction process of the profession, adjust the quality management strategy in a timely manner, and ensure that the quality of the project is always in an effective and controllable state. Through the study of various quality reports and charts, managers can fully and deeply understand the actual situation of project quality, accurately identify the weak links and potential risk points in quality control work, and provide a strong decision-making support basis for project quality management. Early warning triggering is based on set threshold rules and dynamic data monitoring. A threshold is set for key quality indicators (such as the lower limit of concrete strength and the upper limit of steel bar spacing deviation). When the actual data exceeds the threshold, the early warning is triggered immediately; a secondary threshold is set for statistical indicators such as the incidence of quality problems and the rectification delay rate, and an alarm is triggered when it exceeds the threshold. Real-time data monitoring is combined with regular analysis. New data is continuously compared with the threshold during the construction process, or data is analyzed on a daily / weekly / monthly basis to determine whether an early warning is triggered. Early warning information is pushed through multiple channels, such as system pop-ups in project management software or mobile applications, push notifications to the mobile phones of relevant responsible persons, and emails to the project management team's email address to explain in detail the content of the early warning, the project parts involved and the data details, to ensure that the responsible persons obtain information in a timely manner and take action.

[0074] 4. After receiving the early warning, a multi-disciplinary joint investigation team is formed, including construction technicians, quality management personnel, design representatives and expert consultants. On-site investigation of the problem site, scan the target component again through a mobile terminal device equipped with an AR program, compare it with the BIM model in the mobile terminal device, re-check the location of the problem, and combine the BIM model with AR real-life scene to trace the construction process, use tools such as fishbone diagrams to analyze the causes, and accurately locate the root causes such as illegal construction process operations, unqualified material quality, design defects or management loopholes. Formulate rectification plans based on the causes, and clarify rectification measures, responsible persons, deadlines and resource allocation plans. Organize special training and on-site guidance for construction process problems; trace supplier claims and replace materials for material problems; coordinate design units to issue change plans for design defects; optimize management processes, strengthen supervision and performance evaluation. Use BIM and AR technology assistance during rectification implementation, such as optimizing construction process simulation based on BIM models, and AR-assisted on-site rectification positioning and review to ensure accurate and efficient rectification. After the rectification is completed, a review and acceptance will be conducted, data records and analysis results will be updated, the warning status will be lifted, and experience will be reviewed and summarized to optimize the quality management system to prevent the recurrence of similar problems.

[0075] In step S4, an AI machine learning model is introduced into the data platform to analyze historical quality data, predict potential risk processes for unfinished and ongoing construction processes, and generate optimized construction plans. The specific analysis method of the AI machine learning model for historical quality data is as follows: Before construction, temperature sensors, humidity sensors, and vibration sensors are installed in the construction area. After construction, the data of defective areas are integrated, and combined with the detection values of the temperature sensors, humidity sensors, and vibration sensors, a time series database of defective areas is constructed to analyze whether the defects are caused by human factors or environmental factors.

[0076] The specific analysis process is as follows: The AI machine learning model analyzes multiple defect reports, extracts the data of the temperature sensors, humidity sensors, and vibration sensors at the specified positions in the process, forms a time series database of defective areas, and cleans the data to identify whether the temperature sensors, humidity sensors, and vibration sensors generate abnormal values during construction. If so, combined with the defect report, analyze whether the defect is caused by environmental factors. If so, record the abnormal values of the above sensors and the duration of the abnormal values. If not, according to the defect report, analyze what kind of human factor operation causes the change in the sensor values and record it. Apply the obtained data to the unfinished and ongoing construction processes, and predict potential risk processes and generate optimized construction plans according to the data changes of the temperature sensors, humidity sensors, and vibration sensors.

[0077] Application Example 1

[0078] Taking a large commercial complex construction project as an example, at the initial stage of the project, the design team uses Autodesk Revit software to create a BIM model of the project. During the modeling process, the model is split and coded according to the functional zones, construction sections, floors, and component types of the building to ensure that each component has a unique identifier. For example, for the building structure part, the frame columns are numbered according to the floors and areas, such as "F3 - A01 - C01" representing the first frame column in Area A on the 3rd floor; for the mechanical and electrical systems, the pipelines are classified and coded according to the system type, floor, and direction, such as "PL - F2 - W01" representing the first branch of the water supply pipeline system on the 2nd floor. At the same time, the attribute information of the components, including dimensions, materials, strength grades, installation positions, etc., is defined in detail in the model to provide rich data support for subsequent quality inspection and data management.

[0079] During the construction process, construction workers and supervisors use tablets installed with customized AR applications for on-site quality inspection and evaluation. When inspecting and evaluating the steel bar binding process during the main structure construction, the construction workers open the AR application and scan the steel bar skeletons at the construction site. Through image recognition and spatial positioning technologies, the AR application quickly identifies the type, specification, quantity, and binding position of the steel bars, and superimposes the steel bar information in the corresponding BIM model on the tablet screen. The on-site personnel check whether the spacing, anchorage length, cover thickness, etc. of the steel bars meet the design requirements and specification standards by referring to the AR view. If it is found that the spacing of the steel bars in a certain area is too large, the on-site personnel directly mark the problem area in the AR view, take photos, and record the problem description as "The steel bar spacing of the frame beam in area F4-B03 exceeds the design requirements, the measured value is XX mm, and the design value is XX mm". The system automatically associates this problem with the corresponding beam component in the BIM model, generates a quality inspection and evaluation report and a rectification notice, and pushes them to the mobile terminals of the steel bar foreman and the construction supervisor. After the rectification is completed, the construction workers use the AR application again for re-inspection. After confirming that the problem has been solved, they upload the re-inspection results to the system to update the inspection and evaluation records and the BIM model information.

[0080] At the project startup stage, the system administrator imports basic materials such as the design drawings of the construction project, quality standard specifications, and construction process documents into the cloud data storage and management center, and establishes an engineering structure tree and a data directory system in the system to lay a foundation for subsequent data management and inspection and evaluation work. At the same time, user accounts are created for all parties involved in the project (construction workers, quality inspectors, supervisors, owner representatives, etc.) and corresponding permissions are assigned to ensure that each user can only access and operate the functions and data within their scope of responsibility. For example, construction workers can only fill in and modify the data of the construction parts they are responsible for, while supervisors and owner representatives have the permissions to review and access the data.

[0081] After the construction and AR inspection and evaluation are completed, the online filling process of quality acceptance materials is carried out on the cloud. The construction data clerk logs in to the cloud quality acceptance material management platform, selects the material template of "main structure - steel bar sub - project - inspection lot of a certain floor", and the system automatically associates the BIM model and inspection and evaluation data, and fills in pre - filled information such as component number, steel bar specification, inspection lot quantity, qualified quantity, etc. The data clerk supplements the actual construction data and inspection results, such as the raw material inspection report number of the steel bar, the test detection data of the welded joint, etc., and uploads relevant attachments. After the data filling is completed, a review application is submitted. According to the preset review process, the system sequentially pushes the materials to the construction unit's quality inspector, technical person in charge and supervision engineer for review. The review personnel view the material content on the cloud platform, compare the BIM model information and inspection and evaluation records. If it is found that the date filled in the raw material inspection report of the steel bar in the materials is incorrect, the review personnel make annotations on the platform and return it for modification. At the same time, the system automatically notifies the data clerk to make rectifications. The materials that pass the review are automatically stamped with an electronic signature and automatically classified and archived according to the preset archiving rules. The archiving rules can be set according to factors such as the structural level of the project, construction stage, material type, etc., to ensure the orderliness and standardization of material storage, and facilitate subsequent query and statistical analysis.

[0082] During the project completion acceptance or when it is necessary to consult the quality acceptance materials, authorized personnel can search for materials through the material retrieval function of the cloud platform. The retrieval interface provides a variety of retrieval conditions, such as project name, construction location, material type, time range, etc. Users can combine these conditions according to actual needs for precise retrieval. For example, users can enter "Project name: XX Building, Material type: Hidden project acceptance record, Time range: May 20XX - June 20XX" for retrieval, and the system will quickly return a list of materials that meet the conditions.

[0083] When the user clicks to view a certain material, the system first performs an integrity check on the material, and judges whether the material has been tampered with by comparing the stored hash value. If the check passes, the system decrypts the material and displays it to the user in the original format or a browsable web format. At the same time, the system displays the operation log of the material on the interface, including information such as the creation time, modification time, review personnel, download times of the material, etc., which is convenient for users to trace the historical record of the material and understand the flow process of the material.

[0084] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art of this industry should understand that the present invention is not limited by the above - mentioned embodiments. What is described in the above - mentioned embodiments and the specification only illustrates the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of the present invention claimed is defined by the appended claims and their equivalents.

Claims

1. A method for managing engineering quality inspection and evaluation archives based on digital twins, characterized by: The following steps are involved: S1. Create a parametric BIM model and build a digital twin mapping relationship, import the model into the AR software, and configure and set the scene of the BIM model in the AR software; S2. After each construction process is completed or inspected, use a mobile terminal device with an AR program installed to scan the construction target, realize the dual superposition of the AR model and the BIM model, record quality problems or areas where the model comparison is inconsistent on the spot, and generate a quality inspection report and rectification notice; S3. After each construction process or each inspection batch is completed or each sub-project is completed, it will be scanned and compared through step 2, and the quality acceptance data will be recorded and uploaded to the cloud. The cloud system will generate a unique identification code and electronic visa according to the differences in each process, inspection batch or sub-project construction; S4. Build a unified data platform based on the above data; The digital twin mapping relationship is formed by clarifying the size, material and reinforcement configuration information of the component and assigning a unique code to each component, and the corresponding code corresponds to the attribute information of the component, and the attribute information includes the material, specification, manufacturer and installation date of the component; The data platform content of step S4 includes statistical analysis and calculation of the mean, standard deviation and frequency of quality indicators at each stage; The data platform in step S4 can integrate and analyze BIM model data, AR evaluation data, and quality acceptance data according to a predetermined cycle, thereby generating a quality trend chart, a problem distribution bar chart, and a rectification efficiency analysis table; The data platform of step S4 can issue an early warning based on the statistical number of any quality problem exceeding a preset threshold; the set threshold for the early warning includes a primary threshold and a secondary threshold, and the primary threshold and the secondary threshold are triggered independently of each other. The primary threshold includes the lower limit of concrete strength and the upper limit of steel bar spacing deviation. When the actual data exceeds the threshold, a primary early warning is immediately triggered; the secondary threshold includes the incidence rate of quality problems and the rectification delay rate. When the incidence rate of quality problems and the rectification delay rate exceed the set value, a secondary early warning is immediately issued.

2. According to a method for managing engineering quality inspection and evaluation archives based on digital twins according to claim 1, it is characterized by: The coding rules for assigning unique codes to each component in the digital twin mapping relationship are as follows: table code, major category code, medium category code, minor category code, detailed category code, first-level extension category code and second-level extension category code.

3. According to a method for managing engineering quality inspection and evaluation archives based on digital twins according to claim 1, it is characterized by: In step S2, the mobile terminal device installed with the AR program captures the phenomenon graphics by means of camera, identifies the components and construction sites on site by means of image recognition and spatial positioning methods, and then displays them on the display screen of the mobile terminal device. The scanned model is superimposed on the BIM model and the two models are distinguished by color.

4. According to a method for managing engineering quality inspection and evaluation archives based on digital twins according to claim 1, it is characterized by: The scene configuration in step S1 includes identification and positioning configuration; The identification and positioning configuration is implemented by means of two-dimensional code assignment positioning or two-point marking positioning; The QR code assignment positioning adopts the method of partitioning the BIM model into components after the BIM model is imported into the AR software. Each area is composed of a section frame. A component with high recognition is selected in each section frame, and a QR code is set next to it. The distance between the QR code and the component and the distance between the QR code and the ground are recorded, and the QR code is printed. After the construction of the target area is completed, the QR code is posted to the designated location according to the recorded distance between the QR code and the component and the distance between the QR code and the ground. The positioning is realized by scanning the actual QR code with a mobile terminal device; Two-point marking positioning is to divide the BIM model into partition components after the BIM model is imported into the AR software. Each area is composed of a section frame. Two points are selected in the BIM model within the section frame for marking and recording. The actual positions of the two points are found in reality and selected and recorded using a mobile terminal device to achieve positioning.

5. According to a method for managing engineering quality inspection and evaluation archives based on digital twins according to claim 1, it is characterized by: In step S4, the data platform is entered to extract, clean and transform data through Extract tools, Transform tools and Load tools to eliminate data noise, eliminate inconsistencies in the same data, organize and store data according to project structure, time series and professional categories, and build a project quality database.

6. According to the digital twin-based engineering quality inspection and evaluation archive management method of claim 5, it is characterized by: In the step S4, statistical analysis and data mining are performed on the engineering quality database; Statistical analysis was used to calculate the mean, standard deviation, frequency, mean value of concrete component strength and steel bar installation deviation rate for each specialty and construction stage; Data mining refers to cluster analysis to identify areas or processes where quality problems frequently occur, and time series analysis to monitor the fluctuation trend of quality indicators over time, and to find out the hidden relationship between quality problems and construction processes or material suppliers; The data mining in step S4 is analyzed and processed by SPSS data analysis software or Tableau data analysis software.

7. According to a method for managing engineering quality inspection and evaluation archives based on digital twins according to claim 1, it is characterized by: The quality trend graph in step S4 is a line graph or a bar graph with time as the horizontal axis and quality indicators as the vertical axis; the quality indicators include the qualified rate of each inspection batch and the key process score; The problem distribution bar chart is divided according to the profession, floor or construction area. The profession includes construction production, electromechanical system and decoration system. The frequency of occurrence of various quality problems is counted and displayed in a bar chart; The rectification efficiency analysis table includes the time when the problem was discovered, the time when the rectification notice was issued, the actual time when the rectification was completed, the person responsible for the rectification and the evaluation of the rectification effect.

8. According to a method for managing engineering quality inspection and evaluation archives based on digital twins according to claim 1, it is characterized by: After the set threshold of the early warning is triggered, the staff will conduct on-site verification, scan the target component again through a mobile terminal device installed with an AR program, compare it with the BIM model on the mobile terminal device, re-check the location of the problem, and make rectifications and record information according to the BIM model.

9. The method for managing engineering quality inspection and evaluation archives based on digital twins according to claim 1 is characterized by: In step S4, an AI machine learning model is introduced into the data platform to analyze historical quality data, predict potential risk processes for unfinished construction processes and ongoing construction processes, and generate an optimized construction plan.

10. The method for managing engineering quality inspection and evaluation archives based on digital twins according to claim 9 is characterized in that: The specific analysis method of the AI ​​machine learning model for analyzing historical quality data is as follows: before construction, temperature sensors, humidity sensors and vibration sensors are installed in the construction area. After construction, the data of the defective areas are integrated, and the detection values ​​of the temperature sensors, humidity sensors and vibration sensors are combined to construct a time series database of defective areas to analyze whether the defects are caused by human factors or environmental factors.

Images