Engineering supervision system and method based on BIM

Abstract

The invention discloses a BIM-based engineering supervision system and supervision method, and relates to the field of engineering supervision, and the method comprises the steps: building an engineering building three-dimensional model based on a BIM technology, and obtaining the initial report information of an engineering; dividing a complete project into a plurality of subunits, and generating a preliminary project detailed rule standard; progress information of each subunit is obtained, and an actual project progress report is generated; according to a preliminary project detailed rule standard and a project actual progress report, establishing a fitting deviation degree of each subunit; based on the neural network model, establishing an identification model of the project progress, and judging whether the comprehensive progress of each current subunit can complete the project according to requirements or not; and establishing a project supervision visual platform, and sorting, storing and analyzing the progress information of each subunit in the project period. The method has the advantages that the relation among the subunits of the project is effectively improved, the phenomenon of data islands is reduced, and the reliability and the accuracy of project progress comprehensive evaluation are greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of engineering supervision, and specifically to an engineering supervision system and method based on BIM. Background Art

[0002] The construction period of a construction project is long, and each link is intricate. The connection between departments is not strong, and the phenomenon of data islands is likely to occur, resulting in the stagnation and delay of the construction project. Therefore, through BIM technology and digital technology, the project progress of each link and department is digitally processed, and through data analysis, the effective connection between each link and department is achieved, thereby effectively improving the construction efficiency of the construction project and improving refined management.

[0003] The existing engineering supervision technologies based on BIM technology focus on fitting the actual situation of the engineering building with the model, or establishing a BIM information platform for displaying and viewing the progress of each sub-unit of the project. The former, such as the "Reverse Monitoring Method and System Based on Laser Scanner and BIM" (Patent No.: CN107702662B) disclosed by the Chinese Patent Network, obtains the data of the construction site by using a laser scanner to form a point cloud model, matches the point cloud model with the corresponding 3D model, calculates the position difference and component difference of the corresponding points in the point cloud model and the 3D model, and outputs the results to solve the problem of manual supervision of the construction quality and progress in the current engineering construction field. However, such technologies only focus on the supervision of the construction aspect, lack the overall consideration of the project progress, and it is difficult to ensure the reliability and accuracy of the project progress. The latter, such as the "An Information Management System for Engineering Supervision Based on BIM" (Patent No.: CN113065756A) disclosed by the Chinese Patent Network, divides a complete engineering project into multiple sub-units through a BIM information model establishment module, and each sub-unit establishes an information model according to its own work tasks. The project progress data can be communicated, viewed, and compared through this model. However, such technologies require human participation and evaluation during the implementation process, and cannot be judged and evaluated independently by the system. Due to subjective experience, the phenomenon of inaccurate project progress judgment results and information delay is likely to occur. Summary of the Invention

[0004] To solve the above technical problems, an engineering supervision system and method based on BIM are provided. This technical solution solves the problems raised in the above background art, namely, only focusing on the supervision of the construction aspect, lacking the overall consideration of the project progress, being difficult to ensure the reliability and accuracy of the project progress, and requiring human participation and evaluation, being unable to be judged and evaluated independently by the system, and the phenomenon of inaccurate project progress judgment results and information delay due to subjective experience.

[0005] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] A BIM-based engineering supervision method, characterized by comprising:

[0007] Based on engineering requirements and design schemes, establish a three-dimensional model of the engineering building using BIM technology, and obtain the initial report information of the project through the model;

[0008] According to the initial report information of the project, divide the complete project into several sub-units and generate preliminary engineering rule standards;

[0009] Based on the Internet of Things technology and on-site acquisition devices, obtain the progress information of each sub-unit and generate an actual project progress report;

[0010] Based on big data, establish the fitting deviation degree of each sub-unit according to the preliminary engineering rule standards and the actual project progress report;

[0011] According to the fitting deviation degree of each sub-unit, establish an identification model of the project progress based on a neural network model, and judge whether the current comprehensive progress of each sub-unit can complete the project as required;

[0012] Establish an engineering supervision visualization platform to organize, store and analyze the progress information of each sub-unit during the project.

[0013] Preferably, the step of establishing a three-dimensional model of the engineering building using BIM technology and obtaining the initial report information of the project through the model according to engineering requirements and design schemes specifically includes:

[0014] Based on engineering actual requirements and design schemes, use BIM technology to build the construction environment of the engineering building and the composition of each part of the building, construct the mapping relationship between the actual building and the model, and establish a three-dimensional model of the engineering building;

[0015] Divide the project into several sub-regions according to the construction method of the engineering building;

[0016] According to engineering actual requirements and design schemes, determine the components, material information and construction sequence in the three-dimensional model of the engineering building;

[0017] Set the information acquisition period, establish the project progress time series, and obtain the project progress situation in the project progress time series through the three-dimensional model of the engineering building;

[0018] Obtain the initial report information of the complete project according to the simulated complete project results in the three-dimensional model of the engineering building.

[0019] Preferably, the process of dividing the complete project into several sub-units according to the initial report information of the project and generating preliminary project rule standards specifically includes:

[0020] Dividing the complete project into several sub-units, where the sub-units include: material reserve and loss units, cost fund units, construction units, and personnel allocation units in each project sub-region;

[0021] Based on big data and historical data, obtain the acceptable change range for each sub-unit. Among them, when a certain sub-unit exceeds the acceptable change range, it indicates that the project progress is abnormal;

[0022] According to the initial report information of the project and combining with the acceptable change range of each sub-unit, refine the progress data of each sub-unit in the project progress time series;

[0023] Generate preliminary project rule standards according to the progress data of each sub-unit in the project progress time series.

[0024] Preferably, the process of obtaining the progress information of each sub-unit based on Internet of Things technology and on-site acquisition devices and generating an actual project progress report specifically includes:

[0025] According to the nature of each sub-unit, based on Internet of Things technology and on-site acquisition devices, collect the actual progress data of each sub-unit in the project progress time series;

[0026] Among them, Internet of Things technology includes: electronic information submission forms, progress cycle electronic reports, and on-site acquisition devices include: laser scanners, image acquisition devices;

[0027] Generate an actual project progress report according to the actual progress data of each sub-unit in the project progress time series.

[0028] Preferably, the process of establishing the fitting deviation degree of each sub-unit based on big data according to the preliminary project rule standards and the actual project progress report specifically includes:

[0029] According to the preliminary project rule standards and the actual project progress report, obtain the digitalized project progress simulation list and the actual list;

[0030] Perform Kalman filtering and normalization processing on the digitalized project progress simulation list and actual list data to improve the reliability of the data;

[0031] Establish an expression for the fitting deviation degree of each sub-unit according to the processed project progress simulation list and actual list data;

[0032] Based on the calculation results of the fitting deviation degrees of each subunit, an identification matrix of the project progress is established as the input matrix of the neural network;

[0033] The expression for the fitting deviation degree of each subunit is:

[0034]

[0035] In the formula, θ i is the fitting deviation degree of the i-th subunit in the current project progress time series, α is the influence factor of the i-th subunit in the whole project, x i is the simulated data of the i-th subunit in the current project progress time series, is the actual data of the i-th subunit in the current project progress time series, and n is the number of sequences that have been carried out in the project progress time series.

[0036] Preferably, based on the neural network model according to the fitting deviation degrees of each subunit, an identification model of the project progress is established. Judging whether the current comprehensive progress of each subunit can complete the project as required specifically includes:

[0037] Based on the fitting deviation degrees of each subunit, an identification model of the project progress is established based on the neural network model, and based on big data, a training sample set and a target sample set are obtained to learn and train the model;

[0038] The prediction results of the model for the sample data are obtained, and a judgment index for the accuracy of the identification model of the project progress is established to judge the accuracy of the model prediction;

[0039] According to the progress data of the subunits in the received project progress time series, through the identification model of the project progress, comprehensively judge whether the current comprehensive progress of each subunit can complete the project as required;

[0040] When it is judged that the project is difficult to complete, the progress data of the current subunits are fed back, a warning is made, the subunit with the largest fitting deviation degree is marked, and the data is saved. When it is judged that the project can be completed, the data is saved;

[0041] The judgment index for the accuracy of the established identification model of the project progress specifically includes:

[0042] The prediction results of the identification model of the project progress for the sample data are obtained, and the deviation between the prediction results and the sample data is sorted out and statistically analyzed;

[0043] Based on the precision rate algorithm, the precision of the number of non-false positive samples in the prediction results of the identification model of the project progress is determined;

[0044] Based on the recall rate algorithm, the precision of the number of non-false negative samples in the prediction results of the identification model of the project progress is determined;

[0045] Based on the F1-Score algorithm, comprehensively evaluate the comprehensive performance of the prediction results of the recognition model for the project progress;

[0046] According to the comprehensive evaluation results of the F1-Score algorithm, based on big data, set a threshold, and judge whether the comprehensive evaluation result is greater than the threshold. If so, it means that the prediction results of the recognition model for the project progress are not credible, and it is necessary to re-evaluate or adjust the parameters. If not, it means that the prediction results of the recognition model for the project progress are credible, and trust the prediction results;

[0047] The arithmetic expression of the precision rate is as follows:

[0048]

[0049] In the formula, Y i is the precision rate of the i-th group of data of the prediction results of the recognition model for the project progress, X i is the number of positive samples in the i-th group of data of the prediction results of the recognition model for the project progress, Z i is the number of false positive samples in the i-th group of data of the prediction results of the recognition model for the project progress, where the number of false positive samples refers to the number of samples that are actually negative but predicted as positive;

[0050] The arithmetic expression of the recall rate is as follows:

[0051]

[0052] In the formula, R i is the recall rate of the i-th group of data of the prediction results of the recognition model for the project progress, G i is the number of false negative samples in the i-th group of data of the prediction results of the recognition model for the project progress, where the number of false negative samples refers to the number of samples that are actually positive but predicted as negative;

[0053] The arithmetic expression of the F1-Score is as follows:

[0054]

[0055] In the formula, F 1-i is the harmonic mean of the precision rate and recall rate of the i-th group of data of the prediction results of the recognition model for the project progress, reflecting the comprehensive performance of the prediction results of the recognition model for the project progress.

[0056] Preferably, the establishment of the project supervision visualization platform to organize, store and analyze the progress information of each sub-unit during the project specifically includes:

[0057] Establish a project supervision visualization platform, receive and obtain the progress information of each sub-unit during the project, and organize, store and analyze it;

[0058] Based on the engineering supervision visualization platform, it is used to build the operating environment of the recognition model for the project progress and display the prediction results of the model;

[0059] Based on the engineering supervision visualization platform, build the operating environment of the 3D model of the engineering building, which is used to display and control the progress of the 3D model of the engineering building;

[0060] Based on the engineering supervision visualization platform, set up an engineering progress anomaly prompt window and a query window, which are used to query the current progress data and anomaly information fed back by each sub-unit.

[0061] Furthermore, this solution proposes a BIM-based engineering supervision system, which is used to implement the BIM-based engineering supervision method as described above, including:

[0062] A data processing module, which is used to establish a 3D model of the engineering building based on BIM technology according to the engineering requirements and design scheme, and obtain the initial report information of the project through the model; according to the initial report information of the project, divide the complete project into several sub-units, and generate preliminary engineering rules and standards; based on the Internet of Things technology and on-site acquisition devices, obtain the progress information of each sub-unit, and generate an actual project progress report;

[0063] A progress prediction module, which is used to establish the fitting deviation degree of each sub-unit based on big data according to the preliminary engineering rules and standards and the actual project progress report; according to the fitting deviation degree of each sub-unit, establish an identification model of the project progress based on the neural network model, and judge whether the comprehensive progress of each sub-unit can complete the project as required;

[0064] A visualization platform module, which is used to establish an engineering supervision visualization platform to sort out, store and analyze the progress information of each sub-unit during the project.

[0065] Preferably, the data processing module includes:

[0066] A 3D modeling unit, which is used to establish a 3D model of the engineering building based on BIM technology according to the engineering requirements and design scheme, and obtain the initial report information of the project through the model;

[0067] An engineering rules unit, which is used to divide the complete project into several sub-units according to the initial report information of the project, and generate preliminary engineering rules and standards;

[0068] A progress report unit, which is used to obtain the progress information of each sub-unit based on the Internet of Things technology and on-site acquisition devices, and generate an actual project progress report.

[0069] Preferably, the progress prediction module includes:

[0070] A fitting deviation unit, which is used to establish the fitting deviation degree of each subunit based on big data, according to the preliminary engineering rule standards and the actual project progress report.

[0071] A progress prediction unit, which is used to establish an identification model of the project progress based on the neural network model according to the fitting deviation degree of each subunit, and judge whether the current comprehensive progress of each subunit can complete the project as required.

[0072] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0073] The construction period of construction projects is long, and each link is intricate. The connection between departments is not strong, and the phenomenon of data islands is likely to occur. According to the BIM technology, this solution establishes a three-dimensional model of the engineering building, generates the initial report information of the project, and according to the initial report information of the project, combined with the acceptable change range of each subunit, refines the progress data of each subunit in the project progress time series, generates the preliminary engineering rule standards. Secondly, according to the nature of each subunit, based on the Internet of Things technology and on-site acquisition devices, the actual progress data of each subunit in the project progress time series is collected, and an actual project progress report is generated. Through the preliminary engineering rule standards and the actual project progress report, a digitalized project progress simulation list and an actual list are obtained, and the fitting deviation degree of each subunit is determined. Finally, through the fitting deviation degree of each subunit, based on the neural network model, an identification model of the project progress is established to judge whether the current comprehensive progress of each subunit can complete the project as required, and according to the training results of the model, a judgment index for the accuracy of the identification model of the project progress is established, and the comprehensive performance of the prediction results of the identification model of the project progress is comprehensively evaluated, effectively improving the connection between each subunit of the project, reducing the phenomenon of data islands, and greatly improving the reliability and accuracy of the comprehensive evaluation of the project progress. Brief Description of the Drawings

[0074] Figure 1 It is a flowchart of a BIM-based project supervision method of the present invention;

[0075] Figure 2 It is a flowchart of establishing the fitting deviation degree of each subunit based on big data according to the preliminary engineering rule standards and the actual project progress report of the present invention;

[0076] Figure 3 It is a flowchart of establishing an identification model of the project progress based on the neural network model according to the fitting deviation degree of each subunit and judging whether the current comprehensive progress of each subunit can complete the project as required of the present invention. Detailed Embodiments

[0077] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments in the following description are only examples, and those skilled in the art can think of other obvious variations.

[0078] Referring to Figure 1 As shown, a project supervision method based on BIM includes:

[0079] According to the project requirements and design scheme, based on BIM technology, establish a three-dimensional model of the project building, and obtain the initial report information of the project through the model;

[0080] According to the initial report information of the project, divide the complete project into several sub-units, and generate preliminary project rules and standards;

[0081] Based on the Internet of Things technology and on-site acquisition devices, obtain the progress information of each sub-unit, and generate an actual project progress report;

[0082] Based on big data, according to the preliminary project rules and standards and the actual project progress report, establish the fitting deviation degree of each sub-unit;

[0083] According to the fitting deviation degree of each sub-unit, based on the neural network model, establish an identification model of the project progress, and judge whether the current comprehensive progress of each sub-unit can complete the project as required;

[0084] Establish a project supervision visualization platform to organize, store and analyze the progress information of each sub-unit during the project.

[0085] It can be explained that the construction period of construction projects is long, all aspects are intricate, and the connection between departments is weak, which easily leads to the phenomenon of data islands. According to BIM technology, this solution establishes a three-dimensional model of the engineering building, generates the initial report information of the project, and based on the initial report information of the project, combines the acceptable change range of each subunit to refine the progress data of each subunit in the engineering progress time series, generating preliminary engineering rule standards. Secondly, according to the nature of each subunit, based on the Internet of Things technology and on-site acquisition devices, the actual progress data of each subunit in the engineering progress time series is collected to generate an actual project progress report. Through the preliminary engineering rule standards and the actual project progress report, a digital engineering progress simulation list and an actual list are obtained to determine the fitting deviation degree of each subunit. Finally, based on the fitting deviation degree of each subunit, a neural network model is used to establish an engineering progress recognition model to judge whether the current comprehensive progress of each subunit can complete the project as required, and according to the training results of the model, a judgment index for the accuracy of the engineering progress recognition model is established to comprehensively evaluate the comprehensive performance of the prediction results of the engineering progress recognition model, effectively improving the connection between each subunit of the project, reducing the phenomenon of data islands, and greatly improving the reliability and accuracy of the comprehensive evaluation of project progress.

[0086] Refer to Figure 2 As shown, the establishment of the fitting deviation degree of each subunit based on big data according to the preliminary engineering rule standards and the actual project progress report specifically includes:

[0087] According to the preliminary engineering rule standards and the actual project progress report, obtain a digital engineering progress simulation list and an actual list;

[0088] Perform Kalman filtering and normalization processing on the digitalized engineering progress simulation list and actual list data to improve the reliability of the data;

[0089] According to the processed engineering progress simulation list and actual list data, establish an expression for the fitting deviation degree of each subunit;

[0090] According to the calculation results of the fitting deviation degree of each subunit, establish an engineering progress recognition matrix as the input matrix of the neural network;

[0091] The expression for the fitting deviation degree of each subunit is:

[0092]

[0093] In the formula, θ i is the fitting deviation degree of the i-th subunit in the current engineering progress time series, α is the influence factor of the i-th subunit in the whole project, and x i is the simulation data of the i-th subunit in the current engineering progress time series. is the actual data of the i-th subunit in the current project progress time series, and n is the number of sequences that have been carried out in the project progress time series.

[0094] It can be explained that during the data collection process, it is impossible to guarantee the absolute accuracy of the data. Especially for the comprehensive collection of the construction model at the construction site, the existing technology can effectively reduce the data error between the actual and simulated data under the current project progress time series, but it cannot avoid the data accumulation error in the long-term construction, resulting in inaccurate final evaluation results. This solution establishes an expression for the fitting deviation degree of each subunit, accumulates the fitting deviation data of each previous subunit, and thus reflects the fitting deviation degree of each subunit through the data accumulation of each subunit under different project progress time series, thereby improving the comprehensive evaluation ability of the project progress.

[0095] Refer to Figure 3 As shown, based on the neural network model according to the fitting deviation degree of each subunit, establishing an identification model of project progress, and judging whether the current comprehensive progress of each subunit can complete the project as required specifically includes:

[0096] Based on the fitting deviation degree of each subunit, establish an identification model of project progress based on the neural network model, and based on big data, obtain a training sample set and a target sample set to learn and train the model;

[0097] Obtain the prediction results of the model for the sample data, establish a judgment index for the accuracy of the identification model of project progress, and judge the accuracy of the model prediction;

[0098] According to the progress data of the subunits in the received project progress time series, through the identification model of project progress, comprehensively judge whether the current comprehensive progress of each subunit can complete the project as required;

[0099] When it is judged that it is difficult to complete the project, feedback the progress data of the current subunits, give a warning, mark the subunit with the largest fitting deviation degree, and save the data. When it is judged that the project can be completed, save the data;

[0100] The judgment index for the accuracy of the established identification model of project progress specifically includes:

[0101] Obtain the prediction results of the identification model of project progress for the sample data, and sort out and count the deviation between the prediction results and the sample data;

[0102] Based on the precision algorithm, determine the accuracy of the number of non-false positive samples in the prediction results of the identification model of project progress;

[0103] Based on the recall algorithm, determine the accuracy of the number of non-false negative samples in the prediction results of the identification model of project progress;

[0104] Based on the F1-Score algorithm, comprehensively evaluate the comprehensive performance of the prediction results of the recognition model for project progress;

[0105] According to the comprehensive evaluation result of the F1-Score algorithm, based on big data, set a threshold, and judge whether the comprehensive evaluation result is greater than the threshold. If so, it means that the prediction result of the recognition model for project progress is not credible, and it is necessary to re-evaluate or adjust the parameters. If not, it means that the prediction result of the recognition model for project progress is credible, and trust the prediction result;

[0106] The arithmetic expression of the precision rate is as follows:

[0107]

[0108] In the formula, Y i is the precision rate of the i-th group of data in the prediction result of the recognition model for project progress, X i is the number of positive samples in the i-th group of data in the prediction result of the recognition model for project progress, Z i is the number of false positive samples in the i-th group of data in the prediction result of the recognition model for project progress. Among them, the number of false positive samples refers to the number of samples that are actually negative samples but are predicted as positive samples;

[0109] The arithmetic expression of the recall rate is as follows:

[0110]

[0111] In the formula, R i is the recall rate of the i-th group of data in the prediction result of the recognition model for project progress, G i is the number of false negative samples in the i-th group of data in the prediction result of the recognition model for project progress. Among them, the number of false negative samples refers to the number of samples that are actually positive samples but are predicted as negative samples;

[0112] The arithmetic expression of the F1-Score is as follows:

[0113]

[0114] In the formula, F 1-i is the harmonic mean of the precision rate and the recall rate of the i-th group of data in the prediction result of the recognition model for project progress, reflecting the comprehensive performance of the prediction result of the recognition model for project progress.

[0115] It can be explained that the neural network model can perform end-to-end learning directly from the original data to the target output. This method does not require manual intervention or the definition of intermediate features. Through the feedback iteration of neural network model training, the internal relationship between the progress data of each subunit in the engineering progress time series and the complete project can be effectively reflected, which is used to comprehensively judge whether the current comprehensive progress of each subunit can complete the project as required. However, the neural network model cannot determine the final accuracy of the model through a single prediction result during the prediction process. Therefore, in this solution, through the evaluation index of the recognition model accuracy of the engineering progress, the prediction results of the recognition model of the engineering progress are further judged, and the comprehensive performance of the prediction results of the recognition model of the engineering progress is comprehensively evaluated. By judging whether the comprehensive evaluation result is greater than the threshold, if so, it means that the prediction result of the recognition model of the engineering progress is not credible, and it is necessary to re-evaluate or adjust the parameters. If not, it means that the prediction result of the recognition model of the engineering progress is credible, and the prediction result is trusted, thereby effectively improving the reliability and accuracy of the prediction results of the recognition model of the engineering progress.

[0116] Furthermore, based on the same inventive concept as the above BIM-based engineering supervision method, this solution proposes a BIM-based engineering supervision system, including:

[0117] A data processing module, which is used to establish a three-dimensional model of the engineering building based on BIM technology according to the engineering requirements and design scheme, and obtain the initial report information of the project through the model; divide the complete project into several subunits according to the initial report information of the project, and generate preliminary engineering rules and standards; based on the Internet of Things technology and on-site collection devices, obtain the progress information of each subunit, and generate an actual project progress report;

[0118] A progress prediction module, which is used to establish the fitting deviation degree of each subunit based on big data according to the preliminary engineering rules and standards and the actual project progress report; based on the fitting deviation degree of each subunit, establish an identification model of the engineering progress based on the neural network model, and judge whether the current comprehensive progress of each subunit can complete the project as required;

[0119] A visualization platform module, which is used to establish an engineering supervision visualization platform to organize, store and analyze the progress information of each subunit during the project;

[0120] The data processing module includes:

[0121] A three-dimensional modeling unit, which is used to establish a three-dimensional model of the engineering building based on BIM technology according to the engineering requirements and design scheme, and obtain the initial report information of the project through the model;

[0122] A project detail unit, which is used to divide the complete project into several sub-units according to the initial report information of the project and generate preliminary project detail standards;

[0123] A progress report unit, which is used to obtain the progress information of each subunit based on the Internet of Things technology and the on-site collection device, and generate an actual progress report of the project;

[0124] The progress forecast module includes:

[0125] A fitting deviation unit, which is used to establish the fitting deviation degree of each subunit based on big data, according to preliminary engineering specification standards and actual engineering progress reports;

[0126] The progress prediction unit is used to establish an identification model of the project progress based on the neural network model according to the degree of fitting deviation of each sub-unit, and judge whether the current comprehensive progress of each sub-unit can complete the project as required.

[0127] In summary, the advantages of the present invention are: effectively improving the connection between various sub-units of the project, reducing the data island phenomenon, and greatly improving the reliability and accuracy of the comprehensive evaluation of the project progress.

[0128] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A BIM-based engineering supervision method, characterized in that: include: According to the project requirements and design plan, based on BIM technology, establish a three-dimensional model of the project building, and obtain the initial report information of the project through the model; Based on the initial report information of the project, the complete project is divided into several sub-units, and preliminary project details and standards are generated; Based on the Internet of Things technology and on-site collection devices, the progress information of each sub-unit is obtained and a report on the actual progress of the project is generated; Based on big data, the degree of fitting deviation of each subunit is established according to the preliminary engineering specification standards and the actual progress report of the project; According to the fitting deviation degree of each sub-unit, based on the neural network model, the identification model of the project progress is established to judge whether the current comprehensive progress of each sub-unit can complete the project as required; A visual platform for project supervision is established to organize, store and analyze the progress information of each sub-unit during the project.

2. The BIM-based engineering supervision method according to claim 1, characterized in that: The above-mentioned construction of a three-dimensional model of the project building based on the BIM technology according to the project requirements and design plan, and obtaining the initial report information of the project through the model specifically includes: According to the actual needs and design plan of the project, based on BIM technology, build the construction environment of the project building and the composition of each part of the building, and build a mapping relationship between the actual building and the model to establish a three-dimensional model of the project building; Divide the project into several sub-areas according to the construction method of the project building; Determine the components, material information and construction sequence of the three-dimensional model of the project building according to the actual needs of the project and the design plan; Set up information collection cycles, establish a time series of project progress, and obtain the project progress in the time series of project progress through the three-dimensional model of the project building; Based on the complete engineering results simulated in the engineering building 3D model, obtain the initial report information of the complete project.

3. The BIM-based engineering supervision method according to claim 2 is characterized in that: According to the initial report information of the project, the complete project is divided into several sub-units, and the preliminary project details and standards are generated, including: Divide the complete project into several sub-units, where the sub-units include: material reserve and loss unit, cost capital unit, construction unit and staffing unit for each project sub-area; Based on big data and historical data, the acceptable variation range of each sub-unit is obtained. When a sub-unit exceeds the acceptable variation range, it indicates that the project progress is abnormal. Based on the initial report information of the project and the acceptable change range of each sub-unit, refine the progress data of each sub-unit in the project progress time series; Generate preliminary engineering detail standards based on the progress data of each sub-unit in the engineering progress time series.

4. The BIM-based engineering supervision method according to claim 3 is characterized in that: The method of obtaining the progress information of each subunit based on the Internet of Things technology and the on-site collection device and generating the actual progress report of the project specifically includes: According to the nature of each sub-unit, based on the Internet of Things technology and on-site collection devices, the actual progress data of each sub-unit in the project progress time series is collected; Among them, the Internet of Things technology includes: electronic information submission form, progress cycle electronic report, and the on-site collection device includes: laser scanner, image collection device; Generate a project actual progress report based on the actual progress data of each sub-unit in the project progress time series.

5. The BIM-based engineering supervision method according to claim 4 is characterized in that: Based on big data, according to the preliminary engineering specifications and actual engineering progress report, the fitting deviation degree of each sub-unit is established, which specifically includes: According to the preliminary engineering specifications and standards and the actual progress report of the project, obtain the digitized engineering progress simulation list and actual list; Perform Kalman filtering and normalization on the digitized engineering progress simulation list and actual list data to improve data reliability; According to the processed engineering progress simulation list and actual list data, the fitting deviation degree expression of each sub-unit is established; According to the calculation results of the fitting deviation degree of each sub-unit, the identification matrix of the engineering progress is established as the input matrix of the neural network; The fitting deviation degree expression of each subunit is: In the formula, θ i is the fitting deviation degree of the ith subunit in the current project progress time series, α is the influence factor of the ith subunit in the entire project, x i is the simulation data of the ith subunit in the current project progress time series, is the actual data of the ith subunit in the current project progress time series, and n is the number of sequences that have been carried out in the project progress time series.

6. The BIM-based engineering supervision method according to claim 5 is characterized in that: According to the fitting deviation degree of each sub-unit, based on the neural network model, the identification model of the project progress is established to judge whether the comprehensive progress of each sub-unit can complete the project as required, which specifically includes: According to the fitting deviation degree of each sub-unit, a recognition model of engineering progress is established based on the neural network model, and based on big data, training sample sets and target sample sets are obtained to learn and train the model; Obtain the prediction results of the model for sample data, establish the evaluation index of the accuracy of the identification model of the project progress, and judge the accuracy of the model prediction; According to the progress data of the sub-units in the project progress time series, through the identification model of the project progress, comprehensively judge whether the current comprehensive progress of each sub-unit can complete the project as required; When it is judged that the project is difficult to complete, the progress data of each sub-unit is fed back, a warning is issued, and the sub-unit with the largest fitting deviation is marked and the data is saved. When it is judged that the project can be completed, the data is saved; The evaluation indexes for the accuracy of the identification model for establishing the project progress specifically include: Obtain the prediction results of the identification model of the project progress for the sample data, and organize and compile statistics on the deviations between the prediction results and the sample data; Based on the precision algorithm, determine the accuracy of the number of non-false positive samples of the prediction results of the recognition model of the project progress; Based on the recall algorithm, determine the accuracy of the number of non-false negative samples of the prediction results of the recognition model of the project progress; Based on the F1-Score algorithm, the comprehensive performance of the prediction results of the identification model of the project progress is comprehensively evaluated; According to the comprehensive evaluation results of the F1-Score algorithm, based on big data, a threshold is set to determine whether the comprehensive evaluation result is greater than the threshold. If so, it means that the prediction results of the identification model of the project progress are not credible and need to be re-evaluated or the parameters adjusted. If not, it means that the prediction results of the identification model of the project progress are credible and the prediction results are trusted; The arithmetic expression based on the precision rate is: Where Y i is the precision rate of the i-th group of data predicted by the identification model of the project progress, X i is the number of positive samples of the i-th group of data predicted by the recognition model of the project progress, Z i is the number of false positive samples of the i-th group of data in the prediction result of the recognition model of the project progress, where the number of false positive samples refers to the number of samples that are actually negative but predicted to be positive; The recall arithmetic expression is: In the formula, R i is the recall rate of the i-th group of data predicted by the identification model of the project progress, G i is the number of false negative samples of the i-th group of data in the prediction result of the recognition model of the project progress, where the number of false negative samples refers to the number of samples that are actually positive but predicted to be negative; The F1-Score arithmetic expression is: In the formula, F 1-i It is the harmonic mean of the precision and recall of the i-th group of data of the prediction results of the recognition model of the project progress, reflecting the comprehensive performance of the prediction results of the recognition model of the project progress.

7. The BIM-based engineering supervision method according to claim 6 is characterized in that: The establishment of a project supervision visualization platform to organize, store and analyze the progress information of each sub-unit during the project specifically includes: Establish a visual platform for project supervision to receive and obtain the progress information of each sub-unit during the project, and organize, store and analyze it; Based on the engineering supervision visualization platform, it is used to build the operating environment of the engineering progress recognition model and display the prediction results of the model; Based on the engineering supervision visualization platform, build an engineering construction 3D model operating environment to display and control the progress of the engineering construction 3D model; Based on the engineering supervision visualization platform, a project progress exception prompt window is set up, and a query window is set up to query the current progress data and exception information fed back by each sub-unit.

8. A BIM-based engineering supervision system, characterized in that: The method for implementing the BIM-based engineering supervision method according to any one of claims 1 to 7 comprises: A data processing module, which is used to establish a three-dimensional model of the engineering building based on BIM technology according to the engineering requirements and design plan, and obtain the initial report information of the engineering through the model; divide the complete project into several sub-units according to the initial report information of the engineering, and generate preliminary engineering detailed standards; obtain the progress information of each sub-unit based on the Internet of Things technology and on-site collection devices, and generate an actual progress report of the engineering; The progress prediction module is used to establish the fitting deviation degree of each sub-unit based on big data, preliminary engineering specification standards and actual engineering progress reports; according to the fitting deviation degree of each sub-unit, a recognition model of engineering progress is established based on a neural network model to determine whether the current comprehensive progress of each sub-unit can complete the project as required; The visualization platform module is used to establish a visualization platform for project supervision and to organize, store and analyze the progress information of each sub-unit during the project.

9. The BIM-based engineering supervision system according to claim 8, characterized in that: The data processing module comprises: A three-dimensional modeling unit, which is used to establish a three-dimensional model of the engineering building based on the BIM technology according to the engineering requirements and design plan, and obtain the initial report information of the engineering through the model; A project detail unit, which is used to divide the complete project into several sub-units according to the initial report information of the project and generate preliminary project detail standards; The progress report unit is used to obtain the progress information of each subunit based on the Internet of Things technology and the on-site collection device, and generate an actual progress report of the project.

10. The BIM-based engineering supervision system according to claim 9, characterized in that: The progress prediction module includes: A fitting deviation unit, which is used to establish the fitting deviation degree of each subunit based on big data, according to preliminary engineering specification standards and actual engineering progress reports; The progress prediction unit is used to establish an identification model of the project progress based on the neural network model according to the degree of fitting deviation of each sub-unit, and to judge whether the current comprehensive progress of each sub-unit can complete the project as required.

Images