Digital engineering management method and device based on BIM model and multi-data fusion

Abstract

The invention relates to a digital engineering management method and device based on a BIM model and multi-data fusion, and the method comprises the steps: obtaining an engineering BIM model and construction operation data, and carrying out the data extraction, so as to obtain basic BIM data; constructing and obtaining a digital engineering system based on the construction operation data and the basic BIM data; the digital engineering system carries out data fusion on the construction operation data and the basic BIM data to obtain engineering management data; engineering management data are monitored in real time based on a digital engineering system and are mined and analyzed to obtain risk early warning data; and analyzing the risk early warning data to obtain an early warning signal, and transmitting the early warning signal to a corresponding manager through a digital engineering system. Through the BIM data and multi-data fusion digital engineering management technology, digital, intelligent and informationized engineering management is realized, the scientificity and accuracy of engineering management are improved, and the technical effects of improving the engineering management quality are achieved.

Description

Technical Field

[0001] The present application relates to the technical field of engineering management, and in particular to a digital engineering management method and device based on BIM model and multi-data fusion. Background Art

[0002] BIM technology has the characteristics of visibility, simulation, optimization and coordination. Applying BIM technology to engineering safety management can effectively reduce the occurrence of safety accidents. The existing BIM-based technology in engineering safety management mainly builds a building information model (BIM) and a smart construction site management platform. It combines various engineering data to achieve construction simulation, scheme optimization, identification, control, early warning and other multi-party intelligent linkage management, safety education, etc. However, due to the differences between BIM data and other engineering data in data standards, formats, and data temporal information, the traditional engineering management system not only requires manual standardization of data processing and information extraction and conversion, but also requires a lot of manual processing for BIM applied to engineering management to achieve seamless integration of schemes and status. Therefore, the engineering management system does not associate the data information of all elements of the project, resulting in decentralized management, which greatly affects the engineering management analysis ability and collaborative control ability, and thus the engineering early warning ability is insufficient. Summary of the invention

[0003] In order to overcome the shortcomings of the above-mentioned prior art, the present application provides a digital engineering management method and device based on BIM model and multi-data fusion, which specifically adopts the following technical solutions:

[0004] A digital engineering management method based on BIM model and multi-data fusion, comprising the following steps:

[0005] Acquire the engineering BIM model and construction operation data respectively, and extract data from the engineering BIM model to obtain basic BIM data of the corresponding model; the basic BIM data includes at least model attribute information and model rendering information;

[0006] Input construction operation data and basic BIM data into the computer platform to build a digital engineering system;

[0007] The digital engineering system integrates construction operation data with basic BIM data to obtain multi-data integrated engineering management data;

[0008] Real-time monitoring of engineering management data based on digital engineering systems, mining and analysis of engineering management data, and obtaining risk warning data;

[0009] By analyzing risk warning data, warning signals are obtained and transmitted to corresponding managers through the digital engineering system.

[0010] Optionally: the step of extracting data from the engineering BIM model includes:

[0011] Obtain three-dimensional engineering models through modeling software;

[0012] Extracting a corresponding attribute set for each model element of the three-dimensional engineering model, wherein the attribute set includes at least geometric topological data, material information data, construction attribute information, engineering time information, engineering cost information, and design-construction conflict information of the corresponding model element;

[0013] The data in the attribute set is sorted and de-redundant by indexing, and each model element is mapped one-to-one with the corresponding attribute set by rule mapping to obtain the attribute data packet of each model element.

[0014] Optionally: the step of extracting data from the engineering BIM model includes:

[0015] Obtain three-dimensional engineering models through modeling software;

[0016] Extracting corresponding texture data for each model element of the three-dimensional engineering model, wherein the texture data at least includes model mesh data, model material data, and model mapping data;

[0017] The texture data is extended by adding offset, rotation and scaling information through texture transformation operations to form a texture extension file in glTF format.

[0018] Each model element is mapped one-to-one with the corresponding texture extension file by a rule mapping method to obtain the texture data packet of each model element.

[0019] Optional: After obtaining the basic BIM data of the corresponding model, it is necessary to optimize the basic BIM data. The steps of data optimization include:

[0020] Compress the texture extension file data using the draco compression algorithm;

[0021] By using the webp format to replace the image format in the texture extension file, and storing it in the KTX texture format as the container format;

[0022] By changing the granularity of the 3D engineering model, the size of the 3D engineering model is reduced;

[0023] The surface model elements in the three-dimensional engineering model are obtained, and the surface model elements are triangulated by using a triangulated polygon algorithm and a triangulation algorithm.

[0024] Optionally, the construction operation data is obtained through multiple data sources, and the data sources include:

[0025] Input the basic engineering data provided by the customer into the digital engineering system, and the digital engineering system automatically identifies, associates and re-enters the basic data of each unit project in accordance with the three-level project management method to obtain construction operation data;

[0026] Supervision personnel submit engineering supervision data, safety accident data, and quality accident data to the digital engineering system through the supervision-side PAD application; or construction personnel submit engineering implementation progress data, construction quality process data, construction safety data, project settlement data, and project completion settlement data to the digital engineering system through the construction-side PAD application to obtain construction operation data;

[0027] By installing a smart construction site system at the construction site, the hardware collection device at the construction site reports monitoring data to the digital engineering system to obtain construction operation data;

[0028] The acquired construction operation data is corrected and modified through the digital engineering system to obtain new construction operation data.

[0029] Optional: When the digital engineering system performs data fusion on the construction operation data and the basic BIM data, big data cleaning technology or data fusion technology based on machine learning is used to clean the construction operation data and the basic BIM data.

[0030] Optional: The specific steps of using big data cleaning technology to clean the construction data and basic BIM data include:

[0031] Obtain the original data of construction operation data and basic BIM data through multiple data sources;

[0032] Remove duplication, noise and missing values ​​from the original data from different data sources to obtain processed construction data and basic BIM data;

[0033] The processed construction operation data and basic BIM data are sequentially subjected to data normalization, data standardization and normalization conversion to obtain cleaned construction operation data and basic BIM data;

[0034] Integrate the cleaned data from different data sources to obtain the integrated engineering BIM model and construction operation data.

[0035] Optional: The digital engineering system uses AI video surveillance analysis or machine learning algorithms to perform data mining and data analysis on engineering management data.

[0036] Optional: the risk warning data at least includes progress risk warning and budget risk warning;

[0037] The progress risk warning is obtained by analyzing the engineering management data using a decision tree algorithm;

[0038] The budget risk warning is obtained by evaluating the over-budget risk through machine learning technology, time series analysis or causal analysis.

[0039] Furthermore, the present application also discloses a digital engineering management device based on BIM model and multi-data fusion, the device comprising:

[0040] A data acquisition module is used to respectively acquire the engineering BIM model and the construction operation data, and to extract data from the engineering BIM model to acquire basic BIM data of the corresponding model; the basic BIM data at least includes model attribute information and model rendering information;

[0041] A digital system building module is used to input construction operation data and basic BIM data into a computer platform to build a digital engineering system;

[0042] The data fusion module is used in the digital engineering system to fuse the construction operation data with the basic BIM data to obtain multi-data fused engineering management data:

[0043] Data mining and analysis module, which is used to monitor engineering management data in real time based on the digital engineering system, and to mine and analyze engineering management data to obtain risk warning data;

[0044] The early warning signal module is used to obtain early warning signals by analyzing risk early warning data, and transmit the early warning signals to corresponding managers through the digital engineering system.

[0045] Beneficial Effects

[0046] The technical solution of this application has the following beneficial effects:

[0047] The digital engineering management method of the present application realizes digital, intelligent and information-based engineering management through the digital engineering management technology of integrating BIM data with multiple data. It solves the problem of insufficient correlation of data fusion for engineering management in the prior art, improves the scientificity and accuracy of engineering management, and enhances the technical effect of engineering management quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is a flow chart of the digital engineering management method in an embodiment of the present application.

[0049] Figure 2 Schematic diagram of the signal-level data fusion technology used in data fusion in the embodiments of the present application.

[0050] Figure 3 Schematic diagram of the feature-level data fusion technology used in data fusion in the embodiments of the present application.

[0051] Figure 4 Schematic diagram of the decision-level data fusion technology used in data fusion in the embodiments of the present application.

[0052] Figure 5 1 is a structural diagram of an escalator passenger accident monitoring and early warning device in an embodiment of the present application. DETAILED DESCRIPTION

[0053] The present application is further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present application, and cannot be used to limit the scope of protection of the present application. It should be noted that the following detailed descriptions are all exemplary and are intended to provide further explanation of the present application.

[0054] Combination Figure 1 As shown, the embodiment of the present application specifically discloses a digital engineering management method based on BIM model and multi-data fusion, which includes the following steps:

[0055] Step 1: Obtain the engineering BIM model and construction operation data respectively, and extract data from the engineering BIM model to obtain basic BIM data of the corresponding model; the basic BIM data includes at least model attribute information and model rendering information;

[0056] Step 2: Input the construction operation data and basic BIM data into the computer platform to build a digital engineering system;

[0057] Step 3: The digital engineering system integrates the construction operation data with the basic BIM data to obtain multi-data integrated engineering management data;

[0058] Step 4: Monitor engineering management data in real time based on the digital engineering system, and mine and analyze the engineering management data to obtain risk warning data;

[0059] Step 5: Obtain warning signals by analyzing risk warning data, and transmit the warning signals to corresponding managers through the digital engineering system.

[0060] Specifically, the data extraction of the engineering BIM model in step one of the embodiment of the present application at least includes the steps of extracting model attribute information and extracting model rendering information.

[0061] (1) The data extraction method for extracting model attribute information is as follows:

[0062] The three-dimensional engineering model is obtained by creating a modeling software; in the embodiment of the present application, the three-dimensional engineering model is an rvt format model created by AutoDesk Revit modeling software, which can only be viewed and edited on the Windows platform through AutoDesk Revit modeling software, and related operations cannot be performed on the domestic platform.

[0063] The corresponding attribute set is extracted for each model element of the three-dimensional engineering model in rvt format, and the attribute set at least includes geometric topological data, material information data, construction attribute information, engineering time information, engineering cost information and design and construction conflict information of the corresponding model element.

[0064] The data in the attribute set is sorted and de-redundant by indexing, and each model element is mapped to the corresponding attribute set one by one by rule mapping to obtain the attribute data package of each model element. This attribute data contains the complete information of each model element in the 3D engineering model, and is universal and can be viewed and edited on different platforms.

[0065] (2) The data extraction method for model rendering information extraction is:

[0066] First, a three-dimensional engineering model is created by modeling software; in the embodiment of the present application, the three-dimensional engineering model is an rvt format model created by AutoDesk Revit modeling software.

[0067] In order to achieve the rendering of three-dimensional engineering models on different platforms, the embodiment of the present application extracts the corresponding texture data for each model element of the three-dimensional engineering model, and the texture data includes at least model mesh data, model material data and model map data; preferably, the embodiment of the present application extracts, parses and post-processes the mesh, material and map of the three-dimensional rvt model through plug-in development, and finally generates texture data in glTF format. Among them, glTF (GLTransmission Format) is an open source and free format for efficient transmission and loading of three-dimensional scenes and models. It supports fast decompression and use while occupying a small space. At the same time, glTF files support multiple extensions, and the three-dimensional engineering model information is integrated into the glTF file by using glTF extensions.

[0068] Then, the texture data is extended by adding offset, rotation and scaling information through texture transformation operation to form a texture extension file in glTF format; the texture data of a general three-dimensional engineering model has offset, rotation, and scale (scaling) attributes, while the material information derived by conventional methods can only be realized by modifying the model mesh texture or modifying the map, resulting in the mesh information or texture map cannot be reused, and the final model occupies a larger space and the rendering efficiency is reduced. The embodiment of the present application adopts texture transformation operation, such as KHR_texture_trans form extension, and then adds offset, rotation, scale and other information to the texture data, so that the texture data is more consistent with the texture material of the three-dimensional engineering model.

[0069] Finally, each model element is mapped to the corresponding texture extension file one by one through rule mapping, and the texture data package of each model element is obtained. The unique ID of each model element in the 3D engineering model can be used to correspond to the attribute data package and the texture data package respectively, so as to fully reproduce the 3D engineering model on different platforms.

[0070] It should be noted that after obtaining the engineering BIM model, it can also be processed to make the model lightweight. For example, the three-dimensional model can be subjected to a variety of lightweight methods such as topological structure reduction, discrete point elimination, geometric component replacement, model size compression, and material compression, so that it can run smoothly in different modeling software.

[0071] Furthermore, after obtaining the basic BIM data of the corresponding model, the basic BIM data needs to be optimized to reduce the data volume and improve the transmission efficiency. The data optimization methods described in the embodiments of the present application include:

[0072] (1) Data compression of texture extension files is performed using the draco compression algorithm; in the embodiment of the present application, the texture extension files in the glTF format can support draco compression through the KHR_draco_mesh_compression extension. Draco compression can generally reduce the size of the mesh file to 20% or even less of the original file. Using draco compressed meshes reduces the size of the texture extension files in the glTF format, occupies less space, downloads less data, and speeds up transmission.

[0073] (2) By using the webp format to replace the image format in the texture extension file, and storing it in the KTX texture format as the container format; general three-dimensional engineering model materials usually use texture maps in jpg and png formats. However, by using the webp format to replace jpg and png images, this application can have a smaller file size without losing image quality, making web page loading time shorter and requiring less storage space. Furthermore, based on the KTX (Khronos Texture, a universal file format) file format that supports APIs such as OpenGL, OpenGL ES and WebGL, it can be used as a container format to store texture extension files that can be directly supported by the GPU, avoiding re-calculation processing at runtime, and improving rendering performance by supporting textures in the KTX format.

[0074] (3) Reduce the size of the 3D engineering model by changing the granularity of the 3D engineering model; the data of the 3D engineering model created by AutoDeskRevit modeling software is generally very large. For large amounts of data, most hardware devices on different platforms cannot smoothly perform rendering-related processing. In the embodiment of the present application, by supporting users to freely change the granularity of the model, the model size can be reduced to achieve the purpose of smooth rendering and processing. The organizational structure of the RVT 3D model is generated according to the way the model is built. By supporting custom grouping, the model can be reorganized and grouped according to needs to generate different organizational structures to meet different model view functions.

[0075] (4) Obtain the surface model elements in the three-dimensional engineering model, and triangulate the surface model elements through the triangulation polygon algorithm and the triangulation algorithm. Generally, when the triangulation function provided by the AutoDesk Revit software is used to triangulate the surface objects, a large number of unnecessary small triangles will be generated on the sides or edges of the surface objects. These small triangles will not have a visual improvement on the rendering, but will cause a lot of additional overhead on the rendering performance and take up more space. The embodiment of the present application finds the surface elements in the rvt model, uses the triangulation polygon algorithm (such as the Ear Clipping algorithm) and the triangulation algorithm (Delaunay triangulation algorithm) to perform customized triangulation on these surface elements, thereby reducing the triangles of the surface mesh by more than 60%, reducing the occupied space, and improving the rendering performance.

[0076] In addition, in step 1, the construction operation data can be obtained through multiple data sources, and the data sources at least include:

[0077] (1) Input the basic engineering data provided by the customer into the digital engineering system, such as project contracts, engineering plans, engineering costs, engineering budgets, engineering quantity calculations, engineering BIM models, contractor information, etc. The digital engineering system automatically identifies, associates, and re-enters the basic data of each unit project in accordance with the three-level project management method to obtain construction operation data;

[0078] (2) Supervision personnel submit engineering supervision data, safety accident data, and quality accident data to the digital engineering system through the PAD application on the supervision side; or construction personnel submit engineering implementation progress data, construction quality process data, construction safety data, project settlement data, and project completion settlement data to the digital engineering system through the PAD application on the construction side to obtain construction operation data;

[0079] (3) By installing a smart construction site system at the construction site, the hardware collection device at the construction site is used to report monitoring data to the digital engineering system to obtain construction operation data; for example, cameras, gates, smart safety helmets, mechanical equipment monitoring instruments, drones, IoT sensors and other equipment are installed or equipped at the construction site.

[0080] (4) Correct and modify the acquired construction work data through the digital engineering system, such as project planning steps, project payment plan, project budget data, etc., to obtain new construction work data.

[0081] Furthermore, when the digital engineering system described in the embodiment of the present application fuses the construction operation data with the basic BIM data, it usually uses big data cleaning technology or data fusion technology based on machine learning to clean the construction operation data and the basic BIM data. By fusing the construction operation data with the basic BIM data, data such as project progress WES data, section subcontracting data, project component budget data, project quality process data, project supervision process data, project safety management process data, and smart construction site real-time data can be formed to form a complete project management data package. The above data is visualized and monitored in the digital engineering system, and provides data basis for project completion settlement.

[0082] (1) Use big data cleaning technology to clean data:

[0083] Specifically, when the big data cleaning technology is used to clean the construction data and basic BIM data, the general process of data cleaning includes data collection, data preprocessing, data conversion and data integration. The specific steps of the embodiment of the present application include:

[0084] Obtain the original data of construction operation data and basic BIM data through multiple data sources;

[0085] Remove duplication, noise and missing values ​​from the original data from different data sources to obtain processed construction data and basic BIM data;

[0086] The processed construction operation data and basic BIM data are sequentially subjected to data normalization, data standardization and normalization conversion to obtain cleaned construction operation data and basic BIM data;

[0087] Integrate the cleaned data from different data sources to obtain the integrated engineering BIM model and construction operation data.

[0088] It should be noted that the embodiment of the present application preferably uses kafka (an open source stream processing platform) as a distributed message queue, uses flink (an open source stream processing framework) to calculate and process real-time data streams, and uses Kettle (an open source ELT tool) to extract, convert and load the collected engineering data to achieve rapid data cleaning and warehousing.

[0089] (2) Data fusion technology using machine learning:

[0090] Since the information in the network world is diverse and complex, it is crucial to mine valuable information from the huge amount of network information. Data fusion technology is committed to integrating large-scale, heterogeneous, imprecise, inconsistent and even conflicting data into accurate, reliable, consistent and valuable data sets. At present, data fusion technology based on machine learning is mainly divided into three categories: signal-level data fusion, feature-level data fusion and decision-level data fusion; in addition, according to whether there is a supervisory signal, it can be divided into supervised and unsupervised data fusion. Combined with Figure 2 As shown, it discloses a signal level data fusion technology: multi-sensor data fusion; combined Figure 3 As shown, it discloses a feature-level data fusion technology: alarm fusion; combined Figure 4 As shown, it discloses a decision-level data fusion technology: classifier fusion.

[0091] Specifically, the digital engineering system described in the embodiment of the present application uses AI video surveillance analysis or machine learning algorithms to perform data mining and data analysis on engineering management data.

[0092] For example, YOLOv5 (a target detection algorithm program) is used for image target recognition and monitoring. YOLOv5 uses an end-to-end deep learning method to complete target detection tasks in a single forward propagation. YOLOv5 is designed to provide high accuracy, fast processing speed, and a small model, making it suitable for use on embedded devices and low-power platforms.

[0093] For another example, the digital engineering system integrates mature algorithms for five major types of machine learning, including classification, regression, clustering, prediction, and association. It also includes a variety of trainable models: logistic regression, decision tree, random forest, naive Bayes, support vector machine, linear regression, K-means, DBSCAN, and Gaussian mixture model. In addition to providing the main algorithms and modeling functions, the digital engineering system also has data preprocessing functions, including field splitting, row filtering and mapping, column selection, random sampling, filtering null values, merging columns, merging rows, JOIN, row selection, removing duplicate values, sorting, adding sequence numbers, adding calculated fields, etc.

[0094] More specifically, in the embodiment of the present application, the risk warning data at least includes a progress risk warning and a budget risk warning;

[0095] The progress risk warning is obtained by analyzing the engineering management data using a decision tree algorithm; for example, data cleaning and feature selection are performed based on the personnel statistics, vehicle statistics, and material supply conditions reported by the smart construction site system, combined with information such as engineering quantity calculation and weather information, and the decision tree algorithm is used to predict the construction progress to avoid resource conflicts and delays.

[0096] The budget risk warning is obtained by evaluating the over-budget risk through machine learning technology, time series analysis or causal analysis; for example, based on the project scale, construction period, resource input and progress data, machine learning, time series analysis, causal analysis and other budget data prediction methods are used to evaluate and predict the over-budget risk.

[0097] It should be further explained that different types of warning signals can be obtained by analyzing the risk warning data in step five.

[0098] For example, project risk analysis can be performed by analyzing risk warning data:

[0099] (1) Risk prediction and analysis: By collecting project information and past project data, the prediction model is trained through online learning using FlinkML, and the model is used for online streaming. The model is continuously fed into the link in the form of model streaming, making the risk prediction model more real-time. At the same time, risk assessment and identification are carried out to predict possible risks and uncertainties in the future.

[0100] (2) Risk quantification and assessment: The fundamental goal of risk analysis is to quantify the possibility and scope of risk in a project, and to classify risks according to the possibility and scope of impact. Based on data analysis, after the model inference is completed, the inference samples will be recommended to some front-line customers. The customers will give feedback on the results, and then some samples will be spliced. Finally, the data flow returned to the training will form a closed loop, and a risk assessment model will be constructed to quantify and assess the risks to help project managers make decisions.

[0101] (3) Risk management and control: Determine risk management strategies and measures based on risk assessment results, and monitor risk changes in real time through big data. In risk management, big data can help project managers dynamically monitor and warn of project risks, and quickly take effective measures to control risks within an acceptable range.

[0102] Alternatively, project cost analysis can be performed by analyzing risk warning data:

[0103] (1) Cost forecasting and analysis: Based on historical data, project scale, project complexity and other factors, using big data analysis technology and machine learning technology, using linear regression algorithm, establish a cost forecasting model to predict the overall cost of the project and the cost of each stage of the project.

[0104] (2) Cost control and optimization: Collect and analyze actual project cost data in real time, compare it with the project budget, find cost deviations and make timely adjustments. Based on data cleaning and analysis, identify high cost risks and inefficient cost elements, and implement cost control measures to optimize cost performance.

[0105] In addition, it should be noted that the digital engineering management method of the embodiment of the present application also has a 4D visual engineering management function, such as multi-dimensional management based on the BIM model and multi-terminal global control of the overall status of the project.

[0106] The entire digital engineering management method is based on the BIM model, which is associated with project engineering data and presented on multiple terminals. The BIM model contains information about the entire construction cycle and has a high degree of visualization, making progress management, project management, and construction design more intuitive. Webgl technology is used for rendering on the digital management platform, and a wealth of functions are provided. For example, perspective switching, floor viewing, single component information acquisition and viewing, 3D measurement and other functions.

[0107] Furthermore, the system built based on the entire digital engineering management method can integrate revit plug-in tools, digital management platform, smart construction site platform, construction party mobile terminal, supervision party mobile terminal, etc., and manage all aspects of the project management cycle through multiple terminals.

[0108] In addition, combined Figure 5 As shown, the present application also discloses a digital engineering management device based on BIM model and multi-data fusion, the device comprising:

[0109] A data acquisition module is used to respectively acquire the engineering BIM model and the construction operation data, and to extract data from the engineering BIM model to acquire basic BIM data of the corresponding model; the basic BIM data at least includes model attribute information and model rendering information;

[0110] A digital system building module is used to input construction operation data and basic BIM data into a computer platform to build a digital engineering system;

[0111] The data fusion module is used in the digital engineering system to fuse the construction operation data with the basic BIM data to obtain multi-data fused engineering management data:

[0112] Data mining and analysis module, which is used to monitor engineering management data in real time based on the digital engineering system, and to mine and analyze engineering management data to obtain risk warning data;

[0113] The early warning signal module is used to obtain early warning signals by analyzing risk early warning data, and transmit the early warning signals to corresponding managers through the digital engineering system.

[0114] The device provided in the embodiment of the present application can achieve Figure 1 To avoid repetition, the various processes implemented by the method embodiment will not be described again here.

[0115] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A digital engineering management method based on BIM model and multi-data fusion, characterized in that: The steps include: Acquire the engineering BIM model and construction operation data respectively, and extract data from the engineering BIM model to obtain basic BIM data of the corresponding model; the basic BIM data includes at least model attribute information and model rendering information; Input construction operation data and basic BIM data into the computer platform to build a digital engineering system; The digital engineering system integrates construction operation data with basic BIM data to obtain multi-data integrated engineering management data; Real-time monitoring of engineering management data based on digital engineering systems, mining and analysis of engineering management data, and obtaining risk warning data; By analyzing risk warning data, warning signals are obtained and transmitted to corresponding managers through the digital engineering system.

2. The digital engineering management method according to claim 1, characterized in that: The step of extracting data from the engineering BIM model comprises: Obtain three-dimensional engineering models through modeling software; Extracting a corresponding attribute set for each model element of the three-dimensional engineering model, wherein the attribute set includes at least geometric topological data, material information data, construction attribute information, engineering time information, engineering cost information, and design-construction conflict information of the corresponding model element; The data in the attribute set is sorted and de-redundant by indexing, and each model element is mapped one-to-one with the corresponding attribute set by rule mapping to obtain the attribute data packet of each model element.

3. The digital engineering management method according to claim 2 is characterized in that: The step of extracting data from the engineering BIM model includes: Obtain three-dimensional engineering models through modeling software; Extracting corresponding texture data for each model element of the three-dimensional engineering model, wherein the texture data at least includes model mesh data, model material data, and model mapping data; The texture data is extended by adding offset, rotation and scaling information through texture transformation operations to form a texture extension file in glTF format. Each model element is mapped one-to-one with the corresponding texture extension file by a rule mapping method to obtain the texture data packet of each model element.

4. The digital engineering management method according to claim 3 is characterized in that: After obtaining the basic BIM data of the corresponding model, the basic BIM data needs to be optimized. The steps of data optimization include: Compress the texture extension file data using the draco compression algorithm; By using the webp format to replace the image format in the texture extension file, and storing it in the KTX texture format as the container format; By changing the granularity of the 3D engineering model, the size of the 3D engineering model is reduced; The surface model elements in the three-dimensional engineering model are obtained, and the surface model elements are triangulated by using a triangulated polygon algorithm and a triangulation algorithm.

5. The digital engineering management method according to claim 1, characterized in that: When obtaining construction operation data, it is obtained through multiple data sources, and the data sources include: Input the basic engineering data provided by the customer into the digital engineering system, and the digital engineering system automatically identifies, associates and re-enters the basic data of each unit project in accordance with the three-level project management method to obtain construction operation data; Supervision personnel submit engineering supervision data, safety accident data, and quality accident data to the digital engineering system through the supervision-side PAD application; or construction personnel submit engineering implementation progress data, construction quality process data, construction safety data, project settlement data, and project completion settlement data to the digital engineering system through the construction-side PAD application to obtain construction operation data; By installing a smart construction site system at the construction site, the hardware collection device at the construction site reports monitoring data to the digital engineering system to obtain construction operation data; The acquired construction operation data is corrected and modified through the digital engineering system to obtain new construction operation data.

6. The digital engineering management method according to claim 5, characterized in that: When the digital engineering system performs data fusion on the construction operation data and the basic BIM data, it uses big data cleaning technology or data fusion technology based on machine learning to clean the construction operation data and the basic BIM data.

7. The digital engineering management method according to claim 6, characterized in that: The specific steps of using big data cleaning technology to clean construction data and basic BIM data include: Obtain the original data of construction operation data and basic BIM data through multiple data sources; Remove duplication, noise and missing values ​​from the original data from different data sources to obtain processed construction data and basic BIM data; The processed construction operation data and basic BIM data are sequentially subjected to data normalization, data standardization and normalization conversion to obtain cleaned construction operation data and basic BIM data; Integrate the cleaned data from different data sources to obtain the integrated engineering BIM model and construction operation data.

8. The digital engineering management method according to claim 1, characterized in that: The digital engineering system uses AI video surveillance analysis or machine learning algorithms to perform data mining and data analysis on engineering management data.

9. The digital engineering management method according to claim 1, characterized in that: The risk warning data at least includes progress risk warning and budget risk warning; The progress risk warning is obtained by analyzing the engineering management data using a decision tree algorithm; The budget risk warning is obtained by evaluating the over-budget risk through machine learning technology, time series analysis or causal analysis.

10. A digital engineering management device based on BIM model and multi-data fusion, characterized in that: The device comprises: A data acquisition module is used to respectively acquire the engineering BIM model and the construction operation data, and to extract data from the engineering BIM model to acquire basic BIM data of the corresponding model; the basic BIM data at least includes model attribute information and model rendering information; A digital system building module is used to input construction operation data and basic BIM data into a computer platform to build a digital engineering system; The data fusion module is used in the digital engineering system to fuse the construction operation data with the basic BIM data to obtain multi-data fused engineering management data: Data mining and analysis module, which is used to monitor engineering management data in real time based on the digital engineering system, and to mine and analyze engineering management data to obtain risk warning data; The early warning signal module is used to obtain early warning signals by analyzing risk early warning data, and transmit the early warning signals to corresponding managers through the digital engineering system.

Images