Visual management system and method for project supervision information

Abstract

The invention discloses a project supervision information visualization management system and management method, and the system comprises the steps: obtaining multi-source construction data in a construction project through employing a sensor and a BIM technology, carrying out the data preprocessing of the multi-source construction data, extracting the feature data of the construction progress, environment parameters and material utilization rate in an initial multi-source data set, and carrying out the data preprocessing of the initial multi-source data set; obtaining a feature multi-source data set; a CatBoost engineering construction period prediction model is established based on the combination of a CatBoost category type feature gradient lifting algorithm and an NSGA-3 third-generation non-dominated sorting genetic algorithm; the feature multi-source data set is input into the model for training, and the predicted completion time of the construction project is obtained; and generating a real-time progress chart of the construction project according to the predicted completion time, and sending the real-time progress chart to the visual interaction management platform. The project supervision information can be efficiently processed and analyzed at the cloud, and real-time monitoring and state prediction of the supervision information are realized.

Description

Technical Field

[0001] The invention relates to the field of engineering management, and in particular to an engineering supervision information visualization management system and management method. Background Art

[0002] In the field of modern engineering construction, the management and analysis of engineering supervision information occupies a pivotal position. This information is not only related to the overall quality of the project, but also a key factor in engineering safety and progress control. However, the traditional supervision information management method is often limited by manual operation and has a series of obvious drawbacks. The traditional engineering supervision information management method mainly relies on manual recording and analysis, which is incapable of processing large amounts of data. First, due to the limited speed and accuracy of manual recording, information updates are often delayed and the latest status of the project cannot be reflected in a timely manner. Secondly, manual analysis of data requires a lot of time and manpower, and the data processing efficiency is low, which cannot meet the high requirements of modern engineering construction for information real-time. In addition, due to the limitations of manual operation, it is difficult to achieve real-time monitoring of the project with traditional methods, and it is impossible to discover and solve problems in a timely manner. Therefore, how to conduct real-time progress monitoring for construction projects and visualize them, predict the construction progress in advance, and thus discover deficiencies in construction projects in a timely manner is a technical problem that needs to be solved at this stage. Summary of the invention

[0003] The purpose of the present invention is to solve the above problems and to design a project supervision information visualization management system and management method.

[0004] To achieve the above object, the technical solution of the present invention is that, further, in the above-mentioned engineering supervision information visualization management system, the engineering supervision information visualization management system comprises the following steps:

[0005] A multi-source data acquisition module is used to acquire multi-source construction data in a construction project by using sensors and BIM technology, and to perform data preprocessing on the multi-source construction data to obtain an initial multi-source data set;

[0006] A data feature extraction module is used to extract feature data of construction progress, environmental parameters and material utilization rate from the initial multi-source data set to obtain a feature multi-source data set;

[0007] The prediction model building module is used to build a CatBoost project construction deadline prediction model based on the combination of the CatBoost category feature gradient boosting algorithm and the NSGA-3 third-generation non-dominated sorting genetic algorithm;

[0008] A predicted time acquisition module, used for inputting the characteristic multi-source data set into the CatBoost project construction deadline prediction model for training to obtain the predicted completion time of the construction project;

[0009] The project visualization module is used to generate a real-time progress chart of the construction project according to the predicted completion time, and send the real-time progress chart to the visualization interaction management platform.

[0010] Furthermore, in the above-mentioned project supervision information visualization management system, the multi-source data acquisition module includes the following sub-modules:

[0011] A deployment submodule is used to deploy various sensors at the construction site, wherein the sensors include at least temperature and humidity sensors, noise monitors, air quality detectors, video monitoring equipment, and GPS positioning devices;

[0012] The data submodule is used to dynamically capture and update the three-dimensional geometric information, time schedule information and cost information of the project from the design stage to the construction process by combining BIM building information modeling technology;

[0013] The processing submodule is used to clean the collected raw data, wherein the cleaning at least includes removing outliers, filling missing values, unifying data formats and data standardization to obtain an initial multi-source data set.

[0014] Furthermore, in the above-mentioned engineering supervision information visualization management system, the data feature extraction module includes the following submodules:

[0015] The extraction submodule is used to extract the deviation between the actual construction progress and the planned progress in the BIM building information model, wherein the deviation at least includes the actual completion percentage of each sub-project and the activity duration on the critical path;

[0016] The analysis submodule is used to analyze the sensor data and extract the environmental factors that affect the construction progress. The environmental factors include at least temperature, humidity, wind speed and the change trend of environmental factors over time;

[0017] The collection submodule is used to collect material in-and-out records and on-site consumption reports, and calculate the use efficiency and remaining amount of building materials;

[0018] The submodule is used to obtain a feature multi-source data set including construction progress, environmental parameter impact, and material utilization efficiency.

[0019] Furthermore, in the above-mentioned project supervision information visualization management system, the prediction model building module includes the following submodules:

[0020] Establish a submodule for establishing a CatBoost project construction deadline prediction model based on the CatBoost category feature gradient boosting algorithm;

[0021] A parameter submodule, for optimizing the parameters of the CatBoost project construction deadline prediction model using a Bayesian optimization algorithm, wherein the parameters include at least an L2 regularization parameter, a learning rate, and a maximum depth of the number;

[0022] The submodule is used to set the initial values ​​of the L2 regularization parameter, learning rate, and maximum depth to [3,9], [0,0.5], and [4,8] respectively;

[0023] The mapping submodule is used to establish a nonlinear mapping function relationship between input parameters and feature data using an algorithm, and uses the RF random forest algorithm to evaluate the importance of input parameters;

[0024] The submodule is obtained to construct a CatBoost project construction deadline prediction model with the goal of minimizing the absolute value of feature data.

[0025] Furthermore, in the above-mentioned project supervision information visualization management system, the predicted time acquisition module includes the following submodules:

[0026] A division submodule is used to divide the feature multi-source data set into five data sets, of which four data sets are used for model training and the remaining one is used for evaluating model performance;

[0027] Setting a submodule for using a mean square error loss function as an objective function of the CatBoost project construction deadline prediction model;

[0028] A submodule is obtained, which is used to input the characteristic multi-source data set into the CatBoost project construction deadline prediction model for training to obtain the predicted completion time of the construction project.

[0029] Furthermore, in the above-mentioned engineering supervision information visualization management system, the engineering visualization module includes the following submodules:

[0030] Creating a submodule for creating a dynamic real-time progress bar chart for the predicted completion time, showing the actual progress of each construction stage compared with the predicted progress, with color coding indicating progress ahead of schedule and progress behind schedule;

[0031] A submodule for embedding the real-time progress chart into a visual interactive management platform, which supports multi-user login and allows project managers, supervision engineers and on-site workers to view and interact in real time, such as marking problems, submitting change applications, and commenting and communicating;

[0032] The early warning submodule is used to automatically trigger an early warning notification when the prediction results show that certain tasks are about to be delayed, and remind the relevant responsible persons through email, SMS or platform messages.

[0033] Furthermore, in a method of a project supervision information visualization management system, the project supervision information visualization management method comprises the following steps:

[0034] Using sensors and BIM technology to acquire multi-source construction data in a construction project, and performing data preprocessing on the multi-source construction data to obtain an initial multi-source data set;

[0035] Extracting characteristic data of construction progress, environmental parameters and material utilization rate from the initial multi-source data set to obtain a characteristic multi-source data set;

[0036] The CatBoost project construction deadline prediction model is established based on the combination of CatBoost category feature gradient boosting algorithm and NSGA-3 third-generation non-dominated sorting genetic algorithm;

[0037] Inputting the characteristic multi-source data set into the CatBoost project construction deadline prediction model for training to obtain the predicted completion time of the construction project;

[0038] A real-time progress chart of the construction project is generated according to the predicted completion time, and the real-time progress chart is sent to the visual interactive management platform.

[0039] Furthermore, in a method for implementing a project supervision information visualization management system, the multi-source construction data in the construction project is acquired by using sensors and BIM technology, and the multi-source construction data is preprocessed to obtain an initial multi-source data set, including:

[0040] Deploy various sensors at the construction site, including at least temperature and humidity sensors, noise monitors, air quality detectors, video surveillance equipment, and GPS positioning devices;

[0041] Combined with BIM building information modeling technology, dynamically capture and update the project's 3D geometry information, time schedule information, and cost information from the design stage to the construction process;

[0042] The collected raw data are cleaned, and the cleaning at least includes removing outliers, filling missing values, unifying data formats and data standardization to obtain an initial multi-source data set.

[0043] Furthermore, in a method for implementing a project supervision information visualization management system, extracting characteristic data of construction progress, environmental parameters and material utilization rate from the initial multi-source data set to obtain a characteristic multi-source data set includes:

[0044] Extract the deviation between the actual construction progress and the planned progress in the BIM building information model, where the deviation at least includes the actual completion percentage of each sub-project and the activity duration on the critical path;

[0045] Analyze sensor data and extract environmental factors that affect construction progress. Environmental factors include at least temperature, humidity, wind speed, and their changing trends over time.

[0046] Collect material in-and-out records and on-site consumption reports, and calculate the use efficiency and remaining amount of building materials;

[0047] A characteristic multi-source data set including construction progress, environmental parameter impact, and material utilization efficiency is obtained.

[0048] Its beneficial effect is that by using sensors and BIM technology to obtain multi-source construction data in the construction project, the multi-source construction data is preprocessed to obtain an initial multi-source data set; the characteristic data of construction progress, environmental parameters and material utilization rate in the initial multi-source data set are extracted to obtain a characteristic multi-source data set; based on the CatBoost category-type feature gradient boosting algorithm and the NSGA-3 third-generation non-dominated sorting genetic algorithm, a CatBoost engineering construction deadline prediction model is established; the characteristic multi-source data set is input into the CatBoost engineering construction deadline prediction model for training to obtain the predicted completion time of the construction project; a real-time progress chart of the construction project is generated according to the predicted completion time, and the real-time progress chart is sent to the visual interactive management platform. It can enable the engineering supervision information to be efficiently processed and analyzed in the cloud, and realize the real-time monitoring and status prediction of the supervision information. The data analysis results are displayed in the form of graphics, charts, etc. through visualization technology, making the information more intuitive and easy to understand, which is convenient for managers to quickly understand the engineering status and make decisions. In addition, the prediction model can be used to predict the progress and status of the project construction in advance, and further project progress decisions can be made based on the progress and status, such as material usage, personnel control, etc., to improve the management efficiency and quality of project supervision. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the following detailed description of the preferred embodiment.The drawings are only for the purpose of illustrating the preferred embodiments and are not to be construed as limiting the invention.

[0050] Figure 1 This is a schematic diagram of a first embodiment of a project supervision information visualization management system in an embodiment of the present invention;

[0051] Figure 2 It is a schematic diagram of a second embodiment of a project supervision information visualization management system in an embodiment of the present invention;

[0052] Figure 3It is a schematic diagram of a first embodiment of a method for visual management of engineering supervision information in an embodiment of the present invention. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0054] Those skilled in the art will appreciate that, unless otherwise stated, the singular forms "a", "an", "said" and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0055] The present invention will be described in detail below in conjunction with the accompanying drawings. Figure 1 As shown, a project supervision information visualization management system includes the following modules:

[0056] A multi-source data acquisition module is used to obtain multi-source construction data in the construction project using sensors and BIM technology, and to pre-process the multi-source construction data to obtain an initial multi-source data set;

[0057] Specifically, the present embodiment also includes a deployment submodule for deploying various sensors at the construction site, the sensors including at least a temperature and humidity sensor, a noise monitor, an air quality detector, a video monitoring device, and a GPS positioning device;

[0058] The data submodule is used to dynamically capture and update the three-dimensional geometric information, time schedule information and cost information of the project from the design stage to the construction process by combining BIM building information modeling technology;

[0059] The processing submodule is used to clean the collected raw data. Cleaning at least includes removing outliers, filling missing values, unifying data formats and data standardization to obtain an initial multi-source data set.

[0060] A data feature extraction module is used to extract feature data of construction progress, environmental parameters and material utilization rate from the initial multi-source data set to obtain a feature multi-source data set;

[0061] Specifically, the present embodiment also includes an extraction submodule for extracting the deviation between the actual construction progress and the planned progress in the BIM building information model, wherein the deviation at least includes the actual completion percentage of each sub-project and the activity duration on the critical path;

[0062] The analysis submodule is used to analyze the sensor data and extract the environmental factors that affect the construction progress. The environmental factors include at least temperature, humidity, wind speed and the change trend of environmental factors over time;

[0063] The collection submodule is used to collect material in-and-out records and on-site consumption reports, and calculate the use efficiency and remaining amount of building materials;

[0064] The submodule is used to obtain a feature multi-source data set including construction progress, environmental parameter impact, and material utilization efficiency.

[0065] The prediction model building module is used to build a CatBoost project construction deadline prediction model based on the combination of the CatBoost category feature gradient boosting algorithm and the NSGA-3 third-generation non-dominated sorting genetic algorithm;

[0066] Specifically, this embodiment also includes establishing a submodule for establishing a CatBoost project construction deadline prediction model based on a CatBoost category feature gradient boosting algorithm;

[0067] The parameter submodule is used to optimize the parameters of the CatBoost project construction deadline prediction model using the Bayesian optimization algorithm. The parameters include at least the L2 regularization parameter, the learning rate, and the maximum depth of the number.

[0068] The submodule is used to set the initial values ​​of the L2 regularization parameter, learning rate, and maximum depth to [3,9], [0,0.5], and [4,8] respectively;

[0069] The mapping submodule is used to establish a nonlinear mapping function relationship between input parameters and feature data using an algorithm, and uses the RF random forest algorithm to evaluate the importance of input parameters;

[0070] The submodule is obtained to construct a CatBoost project construction deadline prediction model with the goal of minimizing the absolute value of feature data.

[0071] Specifically, the CatBoost categorical feature gradient boosting algorithm in this embodiment is a gradient boosting tree (GBDT) algorithm developed by the Yandex team, which is particularly good at processing categorical features and shows good performance on multiple data sets.

[0072] Algorithm Introduction:

[0073] Basics: CatBoost algorithm is based on the ensemble learning method of gradient boosted decision tree (GBDT), which forms a powerful prediction model by combining multiple weak learners (decision trees). Features: CatBoost has two significant features: adaptive learning rate and category feature processing. Adaptive learning rate helps the algorithm better control the contribution of weak learners in each round of iteration, thereby improving the accuracy of the entire model. Category feature processing uses a category feature encoding technology to convert category features into digital features to better handle the impact of category features.

[0074] Category feature processing

[0075] Hashing: CatBoost hashes the value of each categorical feature and then maps the hash value to a number. The advantage of this is that the information of categorical features can be converted into digital features, so that the impact of categorical features can be better handled. Ordered Target Statistics: CatBoost introduces an innovative algorithm for processing categorical features, Ordered Target Statistics, to deal with the problem of prediction offset. This method uses the statistical value of the target variable to replace the categorical feature value to alleviate the "target leakage" problem.

[0076] Algorithm Advantages

[0077] Support for categorical features: CatBoost naturally supports categorical features, without the need for one-hot encoding or label encoding, making model training more concise and efficient. Handling missing values: CatBoost can automatically handle missing values ​​without manual filling or removal, saving time for data preprocessing. Strong robustness: CatBoost has strong robustness in the face of noisy data and outliers, and the model has strong generalization ability. Efficient performance: CatBoost has high performance when processing large-scale data, and supports GPU acceleration, which improves modeling efficiency.

[0078] Algorithm Principle

[0079] Objective function: CatBoost's objective function is a squared loss function, which is used to optimize the model's parameters. Iterative process: In each iteration, CatBoost trains a new decision tree to correct the errors of the previous model. This process continues until the model converges or the predetermined number of iterations is reached. Feature weight adjustment: CatBoost dynamically calculates the influence of each feature on the model and adjusts the feature weights based on their influence to better handle the dependencies between features.

[0080] Specifically, the NSGA-3 algorithm in this embodiment is an improved version of the NSGA-2 algorithm, which is an evolutionary algorithm for super-multidimensional objective optimization problems. The structure of the NSGA-3 algorithm is basically the same as that of the NSGA-2 algorithm. The biggest difference is that it helps maintain the diversity among population members by providing and adaptively updating some well-distributed reference points. Specifically, DEB et al. designed an elite retention strategy based on reference points to retain those non-dominated population individuals close to the reference points, improve the search rate of excellent individuals and maintain the diversity of the population, so that the algorithm has a good ability to search the Pareto frontier when performing super-multidimensional optimization.

[0081] The predicted time acquisition module is used to input the characteristic multi-source data set into the CatBoost project construction deadline prediction model for training to obtain the predicted completion time of the construction project;

[0082] Specifically, the present embodiment further includes a division submodule for dividing the feature multi-source data set into five data sets, of which four data sets are used for model training and the remaining one is used for evaluating model performance;

[0083] Set up a submodule for using the mean square error loss function as the objective function of the CatBoost project construction deadline prediction model;

[0084] A submodule is obtained, which is used to input the feature multi-source data set into the CatBoost project construction deadline prediction model for training to obtain the predicted completion time of the construction project.

[0085] The project visualization module is used to generate a real-time progress chart of the construction project based on the predicted completion time, and send the real-time progress chart to the visualization interactive management platform.

[0086] Specifically, the present embodiment also includes a creation submodule for predicting the completion time and creating a dynamic real-time progress bar chart, showing the actual progress of each construction stage compared with the predicted progress, with color coding indicating progress ahead of schedule and progress behind schedule;

[0087] Submodule, used to embed real-time progress charts into the visual interactive management platform. The platform supports multi-user login, allowing project managers, supervision engineers and on-site workers to view and interact in real time, such as marking problems, submitting change requests, and commenting and communicating;

[0088] The early warning submodule is used to automatically trigger an early warning notification when the prediction results show that certain tasks are about to be delayed, and remind the relevant responsible persons through email, SMS or platform messages.

[0089] Its beneficial effects are that by using sensors and BIM technology to obtain multi-source construction data in construction projects, data preprocessing is performed on multi-source construction data to obtain an initial multi-source data set; feature data of construction progress, environmental parameters and material utilization rate in the initial multi-source data set are extracted to obtain a feature multi-source data set; a CatBoost project construction deadline prediction model is established based on the combination of the CatBoost category feature gradient boosting algorithm and the NSGA-3 third-generation non-dominated sorting genetic algorithm; the feature multi-source data set is input into the CatBoost project construction deadline prediction model for training to obtain the predicted completion time of the construction project; a real-time progress chart of the construction project is generated according to the predicted completion time, and the real-time progress chart is sent to the visual interactive management platform. It can enable engineering supervision information to be efficiently processed and analyzed in the cloud, and realize real-time monitoring and status prediction of supervision information. The data analysis results are displayed in the form of graphics, charts, etc. through visualization technology, making the information more intuitive and easy to understand, which is convenient for managers to quickly understand the project status and make decisions. In addition, the prediction model can be used to predict the progress and status of the project construction in advance, and further project progress decisions can be made based on the progress and status, such as material usage, personnel control, etc., to improve the management efficiency and quality of project supervision.

[0090] See also Figure 2 , a schematic diagram of a second embodiment of a project supervision information visualization management system in an embodiment of the present invention, the recognition prediction model establishment module includes the following submodules:

[0091] Establish a submodule for establishing a CatBoost project construction deadline prediction model based on the CatBoost category feature gradient boosting algorithm;

[0092] The parameter submodule is used to optimize the parameters of the CatBoost project construction deadline prediction model using the Bayesian optimization algorithm. The parameters include at least the L2 regularization parameter, the learning rate, and the maximum depth of the number.

[0093] The submodule is used to set the initial values ​​of the L2 regularization parameter, learning rate, and maximum depth to [3,9], [0,0.5], and [4,8] respectively;

[0094] The mapping submodule is used to establish a nonlinear mapping function relationship between input parameters and feature data using an algorithm, and uses the RF random forest algorithm to evaluate the importance of input parameters;

[0095] The submodule is obtained to construct a CatBoost project construction deadline prediction model with the goal of minimizing the absolute value of feature data.

[0096] See also Figure 3, a third embodiment schematic diagram of a method for visual management of engineering supervision information in an embodiment of the present invention, the method for visual management of engineering supervision information comprises the following steps:

[0097] Step 101: Use sensors and BIM technology to obtain multi-source construction data in the construction project, perform data preprocessing on the multi-source construction data, and obtain an initial multi-source data set;

[0098] Step 102, extracting characteristic data of construction progress, environmental parameters and material utilization rate from the initial multi-source data set to obtain a characteristic multi-source data set;

[0099] Step 103, establishing a CatBoost project construction deadline prediction model based on a combination of a CatBoost category feature gradient boosting algorithm and a NSGA-3 third generation non-dominated sorting genetic algorithm;

[0100] Step 104, input the feature multi-source data set into the CatBoost project construction deadline prediction model for training to obtain the predicted completion time of the construction project;

[0101] Step 105: Generate a real-time progress chart of the construction project according to the predicted completion time, and send the real-time progress chart to the visual interactive management platform.

[0102] 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 by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A visual management system for engineering supervision information, It is characterized in that The engineering supervision information visualization management system includes the following modules: A multi-source data acquisition module is used to acquire multi-source construction data in a construction project by using sensors and BIM technology, and to perform data preprocessing on the multi-source construction data to obtain an initial multi-source data set; A data feature extraction module is used to extract feature data of construction progress, environmental parameters and material utilization rate from the initial multi-source data set to obtain a feature multi-source data set; The prediction model building module is used to build a CatBoost project construction deadline prediction model based on the combination of the CatBoost category feature gradient boosting algorithm and the NSGA-3 third-generation non-dominated sorting genetic algorithm; A predicted time acquisition module, used for inputting the characteristic multi-source data set into the CatBoost project construction deadline prediction model for training to obtain the predicted completion time of the construction project; The project visualization module is used to generate a real-time progress chart of the construction project according to the predicted completion time, and send the real-time progress chart to the visualization interaction management platform.

2. A project supervision information visualization management system as claimed in claim 1, It is characterized in that The multi-source data acquisition module includes the following sub-modules: A deployment submodule is used to deploy various sensors at the construction site, wherein the sensors include at least temperature and humidity sensors, noise monitors, air quality detectors, video monitoring equipment, and GPS positioning devices; The data submodule is used to dynamically capture and update the three-dimensional geometric information, time schedule information and cost information of the project from the design stage to the construction process by combining BIM building information modeling technology; The processing submodule is used to clean the collected raw data, wherein the cleaning at least includes removing outliers, filling missing values, unifying data formats and data standardization to obtain an initial multi-source data set.

3. A project supervision information visualization management system as claimed in claim 1, It is characterized in that The data feature extraction module includes the following submodules: The extraction submodule is used to extract the deviation between the actual construction progress and the planned progress in the BIM building information model, wherein the deviation at least includes the actual completion percentage of each sub-project and the activity duration on the critical path; The analysis submodule is used to analyze the sensor data and extract the environmental factors that affect the construction progress. The environmental factors include at least temperature, humidity, wind speed and the change trend of environmental factors over time; The collection submodule is used to collect material in-and-out records and on-site consumption reports, and calculate the use efficiency and remaining amount of building materials; The submodule is used to obtain a feature multi-source data set including construction progress, environmental parameter impact, and material utilization efficiency.

4. A project supervision information visualization management system as claimed in claim 1, It is characterized in that The prediction model building module includes the following submodules: Establish a submodule for establishing a CatBoost project construction deadline prediction model based on the CatBoost category feature gradient boosting algorithm; A parameter submodule, for optimizing the parameters of the CatBoost project construction deadline prediction model using a Bayesian optimization algorithm, wherein the parameters include at least an L2 regularization parameter, a learning rate, and a maximum depth of the number; The submodule is used to set the initial values ​​of the L2 regularization parameter, learning rate, and maximum depth to [3,9], [0,0.5], and [4,8] respectively; The mapping submodule is used to establish a nonlinear mapping function relationship between input parameters and feature data using an algorithm, and uses the RF random forest algorithm to evaluate the importance of input parameters; The submodule is obtained to construct a CatBoost project construction deadline prediction model with the goal of minimizing the absolute value of feature data.

5. A project supervision information visualization management system as claimed in claim 1, It is characterized in that The predicted time acquisition module includes the following submodules: A division submodule is used to divide the feature multi-source data set into five data sets, of which four data sets are used for model training and the remaining one is used for evaluating model performance; Setting a submodule for using a mean square error loss function as an objective function of the CatBoost project construction deadline prediction model; A submodule is obtained, which is used to input the characteristic multi-source data set into the CatBoost project construction deadline prediction model for training to obtain the predicted completion time of the construction project.

6. A project supervision information visualization management system as claimed in claim 1, It is characterized in that The engineering visualization module includes the following submodules: Creating a submodule for creating a dynamic real-time progress bar chart for the predicted completion time, showing the actual progress of each construction stage compared with the predicted progress, with color coding indicating progress ahead of schedule and progress behind schedule; A submodule for embedding the real-time progress chart into a visual interactive management platform, which supports multi-user login and allows project managers, supervision engineers and on-site workers to view and interact in real time, such as marking problems, submitting change applications, and commenting and communicating; The early warning submodule is used to automatically trigger an early warning notification when the prediction results show that certain tasks are about to be delayed, and remind the relevant responsible persons through email, SMS or platform messages.

7. A method for visual management of engineering supervision information, It is characterized in that The project supervision information visualization management method comprises the following steps: Using sensors and BIM technology to acquire multi-source construction data in a construction project, and performing data preprocessing on the multi-source construction data to obtain an initial multi-source data set; Extracting characteristic data of construction progress, environmental parameters and material utilization rate from the initial multi-source data set to obtain a characteristic multi-source data set; The CatBoost project construction deadline prediction model is established based on the combination of CatBoost category feature gradient boosting algorithm and NSGA-3 third-generation non-dominated sorting genetic algorithm; Inputting the characteristic multi-source data set into the CatBoost project construction deadline prediction model for training to obtain the predicted completion time of the construction project; A real-time progress chart of the construction project is generated according to the predicted completion time, and the real-time progress chart is sent to the visual interactive management platform.

8. A method for visual management of engineering supervision information as claimed in claim 7, It is characterized in that The method of using sensors and BIM technology to obtain multi-source construction data in a construction project and preprocessing the multi-source construction data to obtain an initial multi-source data set includes: Deploy various sensors at the construction site, including at least temperature and humidity sensors, noise monitors, air quality detectors, video surveillance equipment, and GPS positioning devices; Combined with BIM building information modeling technology, dynamically capture and update the project's 3D geometry information, time schedule information, and cost information from the design stage to the construction process; The collected raw data are cleaned, and the cleaning at least includes removing outliers, filling missing values, unifying data formats and data standardization to obtain an initial multi-source data set.

9. A method for visual management of engineering supervision information according to claim 7, It is characterized in that The extracting characteristic data of construction progress, environmental parameters and material utilization rate from the initial multi-source data set to obtain a characteristic multi-source data set includes: Extract the deviation between the actual construction progress and the planned progress in the BIM building information model, where the deviation at least includes the actual completion percentage of each sub-project and the activity duration on the critical path; Analyze sensor data and extract environmental factors that affect construction progress. Environmental factors include at least temperature, humidity, wind speed, and their changing trends over time. Collect material in-and-out records and on-site consumption reports, and calculate the use efficiency and remaining amount of building materials; A characteristic multi-source data set including construction progress, environmental parameter impact, and material utilization efficiency is obtained.

Images