Intelligent construction site management method and system based on Internet of Things

Abstract

The invention relates to the technical field of the Internet of Things, in particular to an intelligent construction site management method and system based on the Internet of Things. The method comprises the following steps: acquiring construction site environment monitoring data in real time based on multiple Internet of Things sensor nodes; performing real-time environment state change mining on the construction site environment monitoring data, performing global environment perception modeling, and constructing a construction site digital environment perception map; obtaining a construction site omnibearing monitoring video and construction site construction equipment state monitoring data; carrying out target dynamic optical flow tracking on the omnibearing monitoring video of the construction site one by one, carrying out dynamic operation state evolution, and constructing a real-time behavior portrait model of each worker; and performing operation state equipment temperature change trend analysis and dynamic load change evolution according to the state monitoring data of the construction site construction equipment, and generating comprehensive state characteristics of each piece of equipment. The construction site management efficiency and the construction site construction safety are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and in particular to a smart construction site management method and system based on the Internet of Things. Background Art

[0002] With the rapid development of the construction industry and the continuous advancement of information technology, the traditional construction site management model can no longer meet the requirements of modern engineering construction for safety, efficiency and precision. Construction site management involves the coordination and scheduling of personnel, equipment, materials, environment and other aspects. Traditional manual management methods often have problems such as delayed information transmission, low management accuracy, slow response speed, and weak ability to respond to emergencies. Especially in the construction process, there are a large number of complex work tasks, dynamic working environments and special requirements that are highly dependent on safety assurance, which makes the complexity and challenges of construction site management continue to increase.

[0003] With the development of Internet of Things (IoT) technology, smart construction site management has gradually become an important direction for improving construction efficiency and safety. IoT technology connects various sensors, smart devices and monitoring systems to the Internet to achieve real-time data collection, transmission and analysis, providing a new idea for construction site management. By deploying various sensors and monitoring equipment on the construction site, it is possible to monitor the environmental conditions of the construction site, personnel work behavior, construction equipment operation status and other information in real time, greatly improving the level of intelligent construction site management.

[0004] However, despite the huge potential for the application of IoT technology in smart construction sites, how to efficiently and accurately integrate various sensor data, monitor the dynamic changes of the construction site in real time, and make intelligent decisions through data analysis is still a major technical challenge currently faced. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a smart construction site management method and system based on the Internet of Things to solve at least one of the above technical problems.

[0006] To achieve the above object, the present invention provides a smart construction site management method based on the Internet of Things, comprising the following steps: Step S1: Acquire construction site environmental monitoring data in real time based on multiple IoT sensor nodes; perform real-time environmental state change mining on the construction site environmental monitoring data, perform global environmental perception modeling, and construct a digital environmental perception map of the construction site; Step S2: Obtain the all-round monitoring video of the construction site and the monitoring data of the status of the construction equipment on the construction site; perform dynamic optical flow tracking on each target in the all-round monitoring video of the construction site, and perform dynamic operation state evolution to build a real-time behavior portrait model for each worker; Step S3: Analyze the temperature change trend of the equipment in operation and the dynamic load change evolution according to the status monitoring data of the construction equipment on the construction site, and generate the comprehensive status characteristics of each equipment; Step S4: Perform panoramic visualization rendering and dynamic simulation interactive modeling on the digital environment perception map of the construction site according to the real-time behavior portrait model of each worker and the comprehensive status characteristics of each device, and build a dynamic twin model of the construction site; Step S5: Analyze the worker's work behavior at multiple time points on the construction site dynamic twin model, monitor the work safety risks, and build an abnormal work safety warning strategy; Step S6: Dynamically regulate the process resources of the dynamic twin model of the construction site, and optimize intelligent collaborative management based on the abnormal operation safety warning strategy to build a smart construction site collaborative management engine.

[0007] The present invention acquires the construction site environmental data in real time through multiple IoT sensor nodes, ensuring that environmental changes such as temperature, humidity, air quality, etc. can be sensed in the shortest time, and the safety hazards and construction conditions of the construction site can be reflected in time. Through data mining and analysis, abnormal changes in the environment (such as weather changes, pollution, etc.) can be discovered in time, and a comprehensive environmental perception model can be established. This real-time environmental perception effectively supports the safety decision-making of construction personnel and ensures that the construction site operations comply with safety regulations. By establishing a digital environmental perception map of the construction site, the overall environment of the construction site can be displayed more intuitively and accurately, providing data support for subsequent process scheduling and safety monitoring. By tracking the dynamic optical flow of the all-round video of the construction site, each construction target (such as personnel, equipment, etc.) can be realized. In this way, the activities of each worker can be monitored in real time, potential safety hazards can be discovered in time, and the real-time behavior of each worker can be portraited and modeled to ensure that the working status of the workers is fully understood. This not only helps to improve the refinement of construction site management, but also provides basic data for subsequent safety risk analysis. By obtaining the status monitoring data of the equipment, the operation status of various types of equipment can be mastered in real time, including key parameters such as temperature changes and load changes. This helps to perform predictive maintenance before the equipment fails or becomes abnormal, and reduce construction delays and safety hazards caused by equipment failures. Through dynamic load change evolution analysis, any abnormal fluctuations in the operation of the equipment can be discovered in real time to ensure that the construction equipment is in a healthy state. Healthy working status, thereby improving construction efficiency and equipment life. By combining the environmental perception map of the construction site with the worker behavior portrait and equipment status characteristics, panoramic visualization rendering is achieved. This not only helps managers monitor the construction site situation in all directions, but also simulates various scenarios that occur during the construction process through dynamic simulation modeling, and makes plans in advance. This modeling method can accurately reflect the real-time status of the construction site and provide an accurate basis for real-time decision-making. During the construction process, the dynamic changes of the construction site will receive timely feedback to ensure that the project proceeds smoothly according to the scheduled plan. By conducting multi-point analysis of the work behavior of each worker, irregular behaviors or potential risks in their work, such as workers not wearing protective equipment and improper working postures, are discovered and identified in advance. Correction, based on real-time monitoring of workers' operating behaviors, provides real-time safety risk warnings for managers to prevent accidents. When a worker enters a dangerous area, the system will automatically issue a warning and remind other staff to take preventive measures. By dynamically adjusting process resources, the construction progress and personnel allocation can be effectively adjusted to ensure that resources in each link are optimally configured. Based on real-time data, site managers can make quick decisions to avoid waste of resources. The intelligent collaborative management engine can achieve efficient coordination and information sharing between multiple links through comprehensive analysis of various types of data and safety warnings, thereby improving the overall efficiency of site construction. In addition, with the help of big data analysis, it can provide more optimization suggestions for construction management and improve management level and work efficiency.

[0008] In this specification, a smart construction site management system based on the Internet of Things is provided, which is used to execute the smart construction site management method based on the Internet of Things as described above, including: The environmental perception module is used to obtain the construction site environmental monitoring data in real time based on multiple IoT sensor nodes; to mine the real-time environmental state changes of the construction site environmental monitoring data, to perform global environmental perception modeling, and to construct a digital environmental perception map of the construction site; The behavior profiling module is used to obtain all-round monitoring videos of the construction site and monitoring data on the status of construction equipment on the construction site; it tracks the dynamic optical flow of each target in the all-round monitoring videos of the construction site, and performs dynamic operation status evolution to build a real-time behavior profiling model for each worker; The equipment status module is used to analyze the temperature change trend of the operating equipment and the dynamic load change evolution based on the status monitoring data of the construction equipment on the construction site, and generate the comprehensive status characteristics of each equipment; The dynamic twin model module is used to perform panoramic visualization rendering and dynamic simulation interactive modeling on the digital environment perception map of the construction site based on the real-time behavior portrait model of each worker and the comprehensive status characteristics of each device, and to build a dynamic twin model of the construction site; The safety warning module is used to analyze the worker's operating behavior at multiple time points on the construction site dynamic twin model, monitor the operating safety risks, and build a safety warning strategy for abnormal operations; The intelligent collaborative management module is used to dynamically control the process resources of the dynamic twin model of the construction site, and to optimize the intelligent collaborative management based on the abnormal operation safety warning strategy to build a smart construction site collaborative management engine.

[0009] The present invention monitors environmental parameters, and managers can adjust the construction methods of the construction site in time or take necessary protective measures to ensure that the workers' working environment meets safety standards, thereby improving work efficiency. Through real-time mining of environmental changes, potential environmental problems (such as air pollution, excessive noise, etc.) can be quickly identified, and data support can be provided for subsequent safety warnings and adjustments. Through dynamic optical flow tracking, the movements and trajectories of each worker are analyzed in real time, and managers have a clear understanding of the work situation of each worker on the construction site, thereby optimizing human resource allocation. The real-time behavior portrait model can effectively identify unsafe work behaviors or abnormal operations of workers (such as not wearing safety equipment, etc.), issue warnings in time and take intervention measures. Through the analysis of the equipment operating status (such as temperature, load, operating efficiency, etc.), the working status of the equipment can be understood in real time, equipment failures can be predicted, maintenance measures can be taken in advance, and downtime losses caused by equipment failures can be avoided. Regular equipment status analysis and predictive maintenance can effectively extend the service life of the equipment, reduce the frequency of failures, and improve the availability and production of equipment. Efficiency. Through the dynamic twin model, various data of the construction site (worker behavior, equipment status, environmental perception, etc.) are presented in a panoramic manner. Managers can view the overall situation of the construction site in real time and effectively grasp the dynamics of the construction site. The twin model combines virtuality and reality. Managers simulate different scenarios in a digital environment and predict the results of different decisions, so as to make more scientific management decisions. By analyzing the workers' work behavior at multiple time points, safety risks can be discovered in time, such as workers entering high-risk areas and irregular operations. The safety warning module can issue warnings in real time to remind relevant personnel to intervene. Through precise safety monitoring and warnings, the occurrence of safety accidents on the construction site can be effectively reduced to ensure the safety of workers. Based on the dynamic twin model and real-time data analysis, process resources are optimized and scheduled in real time. When a certain link lags in progress or resource conflicts, the system can automatically adjust resource allocation to ensure the smooth progress of the overall progress. The intelligent collaborative management system integrates various links, personnel, and equipment on the construction site, improves the collaborative efficiency between departments on the construction site, and avoids problems caused by information lag or uneven resource allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic diagram of the steps of a smart construction site management method based on the Internet of Things of the present invention; Figure 2 Detailed implementation flow chart of step S1; Figure 3 Detailed implementation flow chart of step S2; Figure 4 Detailed implementation flow chart of step S3. DETAILED DESCRIPTION

[0011] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0012] The present application example provides a smart construction site management method and system based on the Internet of Things. The execution subjects of the smart construction site management method and system based on the Internet of Things include but are not limited to: mechanical equipment, data processing platform, cloud server node, network upload equipment, etc. equipped with the system can be regarded as general computing nodes of the present application, and the data processing platform includes but is not limited to: at least one of an audio and image management system, an information management system, and a cloud data management system.

[0013] See also Figures 1 to 4 The present invention provides a smart construction site management method based on the Internet of Things, and the smart construction site management method based on the Internet of Things includes the following steps: Step S1: Acquire construction site environmental monitoring data in real time based on multiple IoT sensor nodes; perform real-time environmental state change mining on the construction site environmental monitoring data, perform global environmental perception modeling, and construct a digital environmental perception map of the construction site; Step S2: Obtain the all-round monitoring video of the construction site and the monitoring data of the status of the construction equipment on the construction site; perform dynamic optical flow tracking on each target in the all-round monitoring video of the construction site, and perform dynamic operation state evolution to build a real-time behavior portrait model for each worker; Step S3: Analyze the temperature change trend of the equipment in operation and the dynamic load change evolution according to the status monitoring data of the construction equipment on the construction site, and generate the comprehensive status characteristics of each equipment; Step S4: Perform panoramic visualization rendering and dynamic simulation interactive modeling on the digital environment perception map of the construction site according to the real-time behavior portrait model of each worker and the comprehensive status characteristics of each device, and build a dynamic twin model of the construction site; Step S5: Analyze the worker's work behavior at multiple time points on the construction site dynamic twin model, monitor the work safety risks, and build an abnormal work safety warning strategy; Step S6: Dynamically regulate the process resources of the dynamic twin model of the construction site, and optimize intelligent collaborative management based on the abnormal operation safety warning strategy to build a smart construction site collaborative management engine.

[0014] The present invention acquires the construction site environmental data in real time through multiple IoT sensor nodes, ensuring that environmental changes such as temperature, humidity, air quality, etc. can be sensed in the shortest time, and the safety hazards and construction conditions of the construction site can be reflected in time. Through data mining and analysis, abnormal changes in the environment (such as weather changes, pollution, etc.) can be discovered in time, and a comprehensive environmental perception model can be established. This real-time environmental perception effectively supports the safety decision-making of construction personnel and ensures that the construction site operations comply with safety regulations. By establishing a digital environmental perception map of the construction site, the overall environment of the construction site can be displayed more intuitively and accurately, providing data support for subsequent process scheduling and safety monitoring. By tracking the dynamic optical flow of the all-round video of the construction site, each construction target (such as personnel, equipment, etc.) can be realized. In this way, the activities of each worker can be monitored in real time, potential safety hazards can be discovered in time, and the real-time behavior of each worker can be portraited and modeled to ensure that the working status of the workers is fully understood. This not only helps to improve the refinement of construction site management, but also provides basic data for subsequent safety risk analysis. By obtaining the status monitoring data of the equipment, the operation status of various types of equipment can be mastered in real time, including key parameters such as temperature changes and load changes. This helps to perform predictive maintenance before the equipment fails or becomes abnormal, and reduce construction delays and safety hazards caused by equipment failures. Through dynamic load change evolution analysis, any abnormal fluctuations in the operation of the equipment can be discovered in real time to ensure that the construction equipment is in a healthy state. Healthy working status, thereby improving construction efficiency and equipment life. By combining the environmental perception map of the construction site with the worker behavior portrait and equipment status characteristics, panoramic visualization rendering is achieved. This not only helps managers monitor the construction site situation in all directions, but also simulates various scenarios that occur during the construction process through dynamic simulation modeling, and makes plans in advance. This modeling method can accurately reflect the real-time status of the construction site and provide an accurate basis for real-time decision-making. During the construction process, the dynamic changes of the construction site will receive timely feedback to ensure that the project proceeds smoothly according to the scheduled plan. By conducting multi-point analysis of the work behavior of each worker, irregular behaviors or potential risks in their work, such as workers not wearing protective equipment and improper working postures, are discovered and identified in advance. Correction, based on real-time monitoring of workers' operating behaviors, provides real-time safety risk warnings for managers to prevent accidents. When a worker enters a dangerous area, the system will automatically issue a warning and remind other staff to take preventive measures. By dynamically adjusting process resources, the construction progress and personnel allocation can be effectively adjusted to ensure that resources in each link are optimally configured. Based on real-time data, site managers can make quick decisions to avoid waste of resources. The intelligent collaborative management engine can achieve efficient coordination and information sharing between multiple links through comprehensive analysis of various types of data and safety warnings, thereby improving the overall efficiency of site construction. In addition, with the help of big data analysis, it can provide more optimization suggestions for construction management and improve management level and work efficiency.

[0015] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of a method for intelligent construction site management based on the Internet of Things of the present invention. In this example, the method comprises: Step S1: Acquire construction site environmental monitoring data in real time based on multiple IoT sensor nodes; perform real-time environmental state change mining on the construction site environmental monitoring data, perform global environmental perception modeling, and construct a digital environmental perception map of the construction site; In this embodiment, suitable IoT sensors (such as temperature, humidity, gas, noise and vibration sensors) are selected to comprehensively monitor the construction site environment. According to the specific needs of the construction site, various types of sensors are deployed to achieve comprehensive data collection. Sensor nodes are arranged in key areas of the construction site (such as construction sites, material storage areas and offices) to ensure wide data coverage. Each sensor node should be able to connect to the central data processing system through a wireless network (such as Wi-Fi, Zigbee or LoRa). The sensor parameters, such as data collection frequency (such as once per minute) and alarm threshold (such as temperature exceeding a certain critical value), are configured to ensure that the sensor can capture environmental changes in real time. Through the sensor management platform (such as ThingSpeak or AWS), IoT) for remote monitoring and management to ensure the normal operation of sensors. Sensor nodes obtain site environmental data in real time and store the data in local cache to prevent data loss. Data types include temperature (°C), humidity (%), PM2.5 concentration (μg / m³), noise (dB), vibration intensity (g), etc. Ensure data diversity. Send real-time collected data to the central server or cloud database through MQTT or HTTP protocol to ensure the efficiency and security of data transmission. Set up data transmission reliability mechanisms, such as retry mechanisms and data confirmation, to ensure data integrity and accuracy. Use data processing tools (such as Python's Pandas library) to clean the data transmitted to the server, remove duplicate values ​​and outliers, and ensure data quality. Outliers are identified by Z-score or IQR method. Format the data and convert different types of data into a unified format for subsequent analysis. Use data mining technology (such as time series analysis or change point detection algorithm) for real-time analysis Changes in environmental conditions: Use the ARIMA model or moving average method to monitor the trend of data changes, set thresholds, identify the time points when environmental conditions change (such as a sharp rise in temperature or excessive PM2.5 concentration), record the specific time and magnitude of the changes, build an environmental perception model for the construction site based on real-time collected data and state change analysis, use machine learning algorithms (such as random forests or support vector machines) to classify different environmental conditions, and the model should be able to update in real time and adjust itself according to new environmental data to reflect the latest environmental conditions. Use GIS (geographic information system) tools (such as ArcGIS or QGIS) to visualize environmental data and generate a digital environmental perception map of the construction site. The map should include the distribution of various sensors, environmental status indicators (such as temperature and humidity distribution), and risk areas (such as areas where noise exceeds the standard) to facilitate monitoring and decision-making by managers. Develop a real-time monitoring platform to display site environmental monitoring data and digital environmental perception maps, and use Web technologies (such as HTML, JavaScript, and D3.js) for visual display, the platform should support real-time data update and historical data query functions, so that managers can obtain site environment information in a timely manner.

[0016] Step S2: Obtain the all-round monitoring video of the construction site and the monitoring data of the status of the construction equipment on the construction site; perform dynamic optical flow tracking on each target in the all-round monitoring video of the construction site, and perform dynamic operation state evolution to build a real-time behavior portrait model for each worker; In this embodiment, multiple high-definition cameras (such as 1080p or higher resolution) are deployed at key locations on the construction site to ensure that the entire construction area can be covered. The cameras should have night vision capabilities to obtain clear video data under different lighting conditions. Video recording parameters, such as recording frequency (such as 30 frames per second) and storage duration, are set to ensure the integrity and continuity of video data. Video data is uploaded to a central server in real time via network transmission, and RTSP or RTMP protocols are used for streaming media transmission to ensure low latency and high quality of video data. Sensors and monitoring equipment (such as GPS, temperature sensors, load measuring instruments, etc.) are installed on construction equipment to obtain the operating status of the equipment in real time. (such as position, load, running time, etc.), set the data collection frequency (such as once a minute), and upload the status data to the central database through the Internet of Things technology to ensure the real-time nature of the data, select a suitable dynamic optical flow tracking algorithm, such as the Lucas-Kanade method or the Horn-Schunck method, which can calculate the motion trajectory of the target by analyzing the changes in light intensity between video frames. When implementing the algorithm, set the parameters of the optical flow calculation, such as the window size and the number of iterations, to ensure the accuracy and stability of the tracking, and use the target detection algorithm (such as YOLO or SSD) to identify the position of workers and construction equipment in each frame of the video to ensure that each worker can be accurately located and Assign a unique ID to each worker, perform optical flow tracking on each worker, record the motion trajectory and dynamic behavior of each worker on the construction site, combine the motion trajectory and status data to form a dynamic behavior record of the worker, analyze the worker's work status (such as carrying, resting, operating equipment, etc.) based on the tracked worker's motion trajectory, use machine learning models (such as KNN or SVM) to classify the behavior, set the characteristic parameters of behavior classification, such as speed, acceleration and movement mode, in order to accurately identify the worker's work status, use time series analysis methods (such as dynamic time warping DTW) to perform evolution analysis on the worker's work status, identify the behavior pattern and its changing trend, and record each worker's work status. The work status changes, such as the time point when the work status changes from the "resting" state to the "carrying" state, are analyzed, and the influencing factors (such as environmental conditions, equipment status, etc.) are analyzed. The behavioral characteristics of each worker, including the behavior type, duration, frequency, etc., are extracted to form a behavior portrait model of the workers. The behavior portrait model is combined with the equipment status monitoring data to provide more comprehensive background information for each worker's work behavior. Data visualization tools (such as Matplotlib or D3.js) are used to display the real-time behavior portrait of the workers, including motion trajectories and work status change diagrams. The behavior portrait model is applied to the construction site management system to realize real-time monitoring and analysis of the workers' work status and help managers make timely decisions.

[0017] Step S3: Analyze the temperature change trend of the equipment in operation and the dynamic load change evolution according to the status monitoring data of the construction equipment on the construction site, and generate the comprehensive status characteristics of each equipment; In this embodiment, sensors (such as temperature sensors, load sensors, vibration sensors, etc.) installed on the construction equipment monitor the operating status of the equipment in real time. These sensors should have high precision and stability to ensure the accuracy of the data. The data collection frequency is set, for example, temperature and load data, as well as the working hours and operating status of the equipment are recorded once a minute. The real-time monitoring data is transmitted to the central database through the Internet of Things technology (such as MQTT or HTTP) to ensure the real-time and integrity of the data. The data should include the equipment ID, timestamp, temperature value, load value and other relevant status information. A data management system (such as InfluxDB or MySQL) is used to store and manage the collected data for subsequent analysis. Use data processing tools (such as Python's Pandas) to clean the collected temperature data, remove outliers and missing values, use the Z-score method to identify outliers, and correct or eliminate them, format the temperature data into time series, ensure that the data is arranged in chronological order for trend analysis, use statistical analysis methods such as linear regression or moving average method to perform trend analysis on the temperature data, set the window size (such as 7-day moving average) to smooth short-term fluctuations and extract long-term trends, and use visualization tools (such as Matplotlib or Seaborn) to draw temperature change trend graphs to show the temperature changes of the equipment in different time periods, and record the temperature change trends of each device, including temperature rise, The magnitude of the drop and its duration are recorded, and the time point of abnormal temperature change is marked to provide a basis for subsequent load change analysis. The load monitoring data is cleaned and preprocessed to ensure data integrity, remove outliers and noise, and maintain data accuracy. The load data is matched with timestamps and organized into a time series format for subsequent analysis. Dynamic time warping (DTW) or sliding window analysis methods are used to study the evolution of the load. These methods can identify patterns and trends in load changes, especially in different working conditions. Analysis parameters, such as window size and step size, are set to ensure detailed analysis of load changes. The dynamic load changes of each device are recorded, including the highest, lowest and average points of the load, as well as the load The frequency and amplitude of load changes are analyzed, and a load change trend graph is drawn using a visualization tool. The graph is compared with the temperature trend graph to analyze the relationship between load and temperature. Based on the results of the temperature change trend and load change analysis, the comprehensive status characteristics of each device are extracted, including the temperature fluctuation range, load fluctuation range, frequency of abnormal state occurrence, etc. A status evaluation index is generated for each device, such as the "Equipment Health Index" (EHI), which is used to comprehensively evaluate the operating status of the equipment. The generated comprehensive status characteristics are recorded in the database to form a device status characteristic data set for subsequent analysis and comparison. Data mining techniques (such as cluster analysis) are used to classify the comprehensive status characteristics of the equipment to identify equipment with good performance, equipment that requires maintenance, or equipment with potential failures.

[0018] Step S4: Perform panoramic visualization rendering and dynamic simulation interactive modeling on the digital environment perception map of the construction site according to the real-time behavior portrait model of each worker and the comprehensive status characteristics of each device, and build a dynamic twin model of the construction site; In this embodiment, the real-time behavior portrait model data of the workers are collected, including the workers' location, work status and behavior characteristics. At the same time, the comprehensive status characteristics of the equipment are obtained, including information such as operating status, temperature, load, etc. All data are integrated into a unified database to ensure that the data format is consistent and convenient for subsequent processing and analysis. A digital environment perception map of the construction site is obtained, usually in GIS format (such as Shapefile or GeoJSON), to ensure that the map contains the layout information of the workers and equipment, to pre-process the map, to mark key areas (such as worker activity areas, equipment storage areas, etc.) and safety hazard areas, so as to emphasize them in the visualization process, to select suitable visualization tools (such as Unity, Three.js or UnrealEngine) for panoramic rendering, which support 3D graphics rendering and interactive display, and can provide dynamic and real-time visual effects, to set visualization parameters such as perspective, lighting and material to improve the realism of the rendering effect, to use high-quality texture and light source models to enhance the details of the scene, to input the integrated data into the visualization system, to dynamically display the location, work status and behavior characteristics of each worker, and to display the comprehensive status characteristics of the equipment (such as color coding to display the equipment status), to achieve real-time updates in a three-dimensional environment, to ensure When the status of workers or equipment changes, the representation on the map can also change accordingly. Develop a user interaction interface that allows managers to interact with the visualization model through a mouse or touch screen, such as selecting a specific worker or equipment to view detailed information. Design functions such as zooming, rotating, and switching perspectives to allow users to observe the construction site from different angles. Based on the worker behavior portrait and equipment status characteristics, establish a dynamic simulation model, use discrete event simulation (DES) or agent-based model (ABM) to simulate the interactive behavior of workers and equipment, set model parameters, such as worker work efficiency, equipment operating efficiency, and environmental factors, to ensure that the model truly reflects the construction site. Dynamic characteristics, run dynamic simulation models, observe the performance of workers and equipment under different working conditions, record key performance indicators (KPIs), such as work efficiency, equipment utilization and safety incident rate, adjust simulation parameters, simulate different work scenarios (such as high-intensity work, equipment failure, etc.), evaluate the impact of these conditions on the overall operation of the construction site, integrate visualization rendering and dynamic simulation results to form a dynamic twin model of the construction site. The model should include the real-time behavior of workers, the comprehensive status of equipment and their mutual influence, update model data regularly, ensure that the dynamic twin model can reflect changes in the construction site in real time, establish a feedback mechanism, and adjust model parameters according to actual conditions.

[0019] Step S5: Analyze the worker's work behavior at multiple time points on the construction site dynamic twin model, monitor the work safety risks, and build an abnormal work safety warning strategy; In this embodiment, data at multiple time points are extracted from the dynamic twin model of the construction site, including workers' location information, job type, job time, environmental variables (such as temperature, humidity) and equipment status. These data should include workers' job conditions in different time periods (such as morning, noon and evening on working days). The time interval for data collection is set, for example, job data is obtained once an hour, and timestamps are marked to ensure the continuity and integrity of the data. Data processing tools (such as Python's Pandas library) are used to clean the extracted data, remove duplicate values ​​and missing values, ensure data accuracy, and integrate data at different time points into a unified data frame to facilitate subsequent actions. Analyze and compare to ensure that the data format is unified for subsequent processing. Apply statistical analysis methods (such as descriptive statistics and cluster analysis) to extract features of workers' work behaviors. The extracted features include work efficiency, frequency of work types, work duration, and work status transitions. Set key performance indicators (KPIs), such as the number of working hours per day for each worker, the amount of work completed, and the proportion of rest time, to facilitate the evaluation of work behavior performance. Use machine learning methods (such as K-means clustering or decision trees) to identify workers' work behavior patterns. By analyzing the behavioral differences between different workers, identify efficient and inefficient workers. Record the changing trends of behavioral patterns for subsequent Provide basic data for safety risk monitoring. According to the characteristics of workers' work behavior, set indicators for work safety risk monitoring, including high-risk work types (such as high-altitude work), work time (such as night shift work), environmental risks (such as extreme weather), etc. Set risk thresholds. For example, under certain conditions (such as work time exceeding 8 hours, temperature exceeding 35°C), mark it as high-risk work. Combined with real-time data streams, monitor whether the workers' work behavior meets safety standards. When it is detected that the workers' work behavior exceeds the set safety threshold, the system automatically triggers a risk alarm and uses visualization tools (such as Grafana or Tableau) to display the real-time work risk status for easy management. Personnel should promptly understand the safety status, build abnormal operation safety warning strategies based on the results of workers' work behavior analysis and risk monitoring data, use rule-based models or machine learning models (such as random forests, support vector machines) to make warning decisions, set warning parameters, including workers' work behavior characteristics, external environmental conditions and equipment status, ensure that the warning model can accurately identify abnormal work conditions, implement the early warning system, and analyze workers' work behavior and environmental data in real time. When an abnormal situation is detected, the system automatically triggers an early warning, notifies safety management personnel, and records warning events, including warning time, worker ID, work behavior characteristics and risk level, for subsequent analysis and improvement of early warning strategies.

[0020] Step S6: Dynamically regulate the process resources of the dynamic twin model of the construction site, and optimize intelligent collaborative management based on the abnormal operation safety warning strategy to build a smart construction site collaborative management engine.

[0021] In this embodiment, according to the actual needs of the construction site, the resources required for each process (including manpower, equipment and materials) are analyzed, and the priorities of different processes are determined. The historical data are reviewed using data analysis tools (such as Excel or Python), the peak and trough periods of resource demand are identified, and resource control goals are set. For example, the resource supply of key processes is ensured during peak periods to avoid delays in construction due to insufficient resources. A dynamic control model is established, and an optimization algorithm (such as linear programming, integer programming or genetic algorithm) is used to formulate a resource allocation plan. The objective function (such as minimizing the total cost or minimizing the construction period) and constraints (such as total resource limit, process dependency, etc.) are set. According to the characteristics of the process and resource availability, control parameters (such as control frequency, resource allocation ratio) are set to achieve flexible resource management. Under the guidance of the dynamic control model, the allocation of resources is adjusted in real time. When a certain process is in progress, more manpower and equipment are automatically allocated to ensure the smooth progress of the process. Through the scheduling system (such as SAP or Oracle Primavera) to implement resource allocation, ensure the real-time update and execution of the scheduling plan, set up a real-time monitoring system, monitor resource usage and process progress, ensure the effective use of resources, use sensors and monitoring equipment to obtain real-time data, and display key indicators (such as resource utilization, process progress, etc.) through a visual panel. According to the real-time monitoring results, timely adjust resource allocation to respond to emergencies or process changes to ensure the smoothness of the construction process, integrate the abnormal operation safety warning strategy with the resource dynamic control model to ensure that safety risk factors are considered in the resource control process. When the operation behavior of a process is marked as high risk, the system automatically adjusts the resource allocation, reduces the operation intensity of the process or suspends the operation, and sets warning parameters and thresholds, such as when workers exceed the set time or environmental conditions deteriorate When a task is completed, the system automatically triggers an early warning and adjusts resource allocation. Based on process requirements and real-time data, the collaborative management process is optimized to ensure effective coordination of resources between processes. If a process requires a large number of equipment while another process is temporarily idle, the system should automatically adjust equipment scheduling to improve resource utilization efficiency. The collaborative management strategy is continuously optimized through machine learning algorithms (such as reinforcement learning). According to historical data and real-time feedback, the resource control strategy is adjusted to adapt to the changing construction environment. The dynamic resource control and safety early warning strategy are integrated to build a collaborative management engine for smart construction sites. The engine should have functions such as real-time data processing, resource scheduling and safety monitoring to form a closed-loop management. Cloud computing and big data technologies are used to ensure that the engine has strong data processing capabilities and real-time response capabilities, and can handle large amounts of sensor data and monitoring information.

[0022] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Step S11: Acquire construction site environment monitoring data in real time based on multiple IoT sensor nodes; Step S12: performing construction site location calculation on multiple IoT sensor nodes to obtain the construction site location coordinates of each sensor; Step S13: performing construction site area type analysis based on the construction site positioning coordinates of each sensor to obtain multiple construction site area type features; Step S14: mining the real-time environmental state change of the construction site environmental monitoring data according to the characteristics of multiple construction site area types to obtain the environmental state change characteristics of each area; Step S15: Perform global environmental perception modeling based on the environmental state change characteristics of each area to build a digital environmental perception map of the construction site.

[0023] In this embodiment, the environmental parameters that need to be monitored are determined, such as temperature (using a digital temperature sensor), humidity (using a humidity sensor), noise (using a sound level meter), and air quality (using a gas sensor), and a suitable sensor model is selected to ensure that its measurement range and accuracy meet the needs of site environmental monitoring. The sampling frequency of the sensor (such as sampling once per minute) and the data transmission method (such as Wi-Fi, LoRa, Zigbee) are configured, and sensor nodes are installed at key locations on the site to ensure that the environmental monitoring needs of the entire site are covered and blind spots are avoided. The sensor is connected to the network, and data is uploaded and managed through an IoT platform (such as AWSIoT, ThingSpeak), ensuring that the network connection is stable so that real-time data transmission can be achieved. Transmission, configure sensor nodes to collect environmental data in real time, and transmit the data to the cloud or local server, ensure that the data format is unified (such as JSON, CSV) for subsequent processing, set data storage strategies, including data retention time, backup frequency, etc., to ensure data integrity and availability, use data visualization tools (such as Grafana, Tableau) to display real-time monitoring data, monitor environmental parameter change trends, record data transmission delays and packet loss rates to evaluate system stability and reliability, select appropriate positioning technology, such as GPS, Wi-Fi positioning or Bluetooth positioning. For small-scale construction sites, Wi-Fi or Bluetooth positioning is more accurate. Determine the positioning accuracy requirements. For example, the accuracy of GPS is generally 5-10 meters, Wi-Fi positioning is accurate to 1-3 meters. Install positioning modules on sensor nodes to obtain the geographic coordinates of sensors in real time. For solutions using GPS, ensure that sensors are in open spaces to obtain the best signal. Use existing base stations or Wi-Fi hotspots for positioning. Record the location information of each sensor node, including longitude, latitude, and altitude. Integrate the positioning data of each sensor with its environmental monitoring data to generate an environmental data set containing location information. Verify positioning accuracy. Compare the error between GPS positioning results and known locations to ensure the accuracy of positioning data. Define area types, such as construction areas, storage areas, rest areas, and safety areas, according to the different functions and usage of the construction site. Set each area The characteristics of the type, for example, the construction area has high noise and dust, while the rest area is relatively quiet and has a good environment. Use geographic information system (GIS) software (such as ArcGIS, QGIS) to visualize the sensor location data, assign sensors to different area types, identify the environmental characteristics of each area through spatial analysis methods (such as buffer analysis, point density analysis), and match them with sensor data to generate a construction site area type analysis report, record the environmental characteristics and sensor data of each area type, and ensure that the analysis results can provide a basis for subsequent environmental monitoring and management. According to different area types, define the characteristics of environmental state changes, such as temperature fluctuations, humidity changes, noise peaks, and air quality index, and set change thresholds.For example, temperature changes exceeding 2°C, humidity changes exceeding 10%, etc., are used as the basis for determining state changes. Statistical analysis methods (such as time series analysis, moving average) are used to mine monitoring data in real time to identify the changing trend of environmental states. Anomaly detection algorithms (such as Z-score, anomaly detection based on machine learning) are used to identify abnormal environmental states in different areas. Changes in environmental states in each area, including the time, type, and degree of change, are recorded, and environmental state change reports are generated. Real-time change results are fed back to the monitoring system for timely environmental regulation. Appropriate environmental perception modeling methods are selected, such as physical model-based modeling, data-based machine learning models, or hybrid models. Model inputs are determined, including sensor data, regional type characteristics, and state change characteristics. Machine learning algorithms (such as random forests and support vector machines) are used to build environmental perception models. The models are trained to identify the relationship between environmental states and influencing factors, ensuring that the models can process multidimensional data, and feature selection and dimensionality reduction are performed to improve the accuracy and efficiency of the models. Digital environmental perception maps are generated based on the constructed environmental perception models to display the environmental state of the construction site and its changing trends. GIS software is used to visualize the perception maps and sensor data to provide intuitive environmental monitoring results.

[0024] In this embodiment, the specific steps of step S14 are: Based on the construction site positioning coordinates of each sensor, the construction site environmental monitoring data is divided into different construction site area parameters to obtain environmental monitoring parameters of each construction site area, wherein the environmental monitoring parameters include air humidity parameters, air dust concentration parameters, noise intensity and ground vibration amplitude; Conduct time series humidity change analysis on air humidity parameters to generate time series humidity change characteristics; Conduct dust concentration pollution assessment on air dust concentration parameters to generate air quality pollution assessment values; Perform multi-time point discrete fitting on the noise intensity and construct the noise fluctuation intensity curve; Conduct real-time amplitude fluctuation analysis on the ground vibration amplitude to obtain the real-time amplitude fluctuation characteristics of the ground; The environmental state change characteristics of each region are mined by the time series humidity change characteristics, air quality pollution assessment values, noise fluctuation intensity curves and real-time ground amplitude fluctuation characteristics to obtain the environmental state change characteristics of each region.

[0025] In this embodiment, the construction site is divided into different areas according to the layout and function of the construction site, such as construction area, storage area, rest area, etc. The division standard of each area should take into account the distribution and environmental characteristics of the sensors, set the boundary conditions of the area, and according to the geographical coordinates and environmental characteristics of the sensors, regard the areas within a certain range (such as 50 meters) from the sensors as the same area. The environmental monitoring data of the sensors (including air humidity, dust concentration, noise intensity and vibration amplitude) are associated with their location information. A database (such as SQL) or a data processing tool (such as Pandas) is used to integrate the data to ensure that the data of each sensor is correctly classified into the corresponding area, so as to form a data set containing the environmental monitoring parameters of each area. Extract the air humidity parameters of each area from the dataset, record the humidity data of each sensor and its corresponding timestamp for time series analysis, ensure the integrity of the data, remove missing values ​​and outliers, use interpolation methods (such as linear interpolation) to fill in the missing parts of the data, use time series analysis methods (such as ARIMA, SARIMA) to model the humidity data, analyze the humidity change trend, generate a humidity change characteristic graph, record the peak value, valley value and fluctuation range of humidity, and calculate the rate of change (such as daily change amplitude), extract the characteristic parameters of time series humidity change, such as the mean, standard deviation, maximum and minimum values ​​of humidity, generate a humidity change characteristic report, record the humidity change characteristics of each area for subsequent analysis and comparison, from the environment Extract the air dust concentration data of each region from the monitoring data set to ensure the accuracy and completeness of the data, clean the data, remove outliers, and use statistical methods (such as Z-score) to identify and process extreme values. Set the pollution assessment value of dust concentration according to the national or regional air quality standards. When the PM10 concentration exceeds 100μg / m³, it can be regarded as poor air quality. Calculate the air quality index (AQI) of each region, generate an air quality pollution assessment report, record the air quality pollution assessment value of each region, including the pollution level (such as good, light pollution, moderate pollution, etc.), organize the assessment report, and form an archive for subsequent tracking and management. Extract noise intensity data from the environmental monitoring data set, and perform time series cleaning and Processing, ensure the timeliness and accuracy of data, display the noise intensity change trend through visualization tools (such as Matplotlib), use discretization method to fit the noise data, construct the noise fluctuation intensity curve, use polynomial regression or spline interpolation method to fit, record the parameters of the fitting curve, including goodness of fit (R²), fluctuation range and frequency characteristics, extract ground vibration amplitude data from the environmental monitoring data set, ensure data integrity and accuracy, monitor and record vibration data in real time, set the sampling frequency (such as sampling once per second), use Fourier transform (FFT) to analyze ground vibration data, identify vibration frequency and amplitude characteristics, generate real-time amplitude fluctuation characteristic diagram, record the peak value, frequency and fluctuation range of vibration,Integrate the time series humidity change characteristics, air quality pollution assessment values, noise fluctuation intensity curves and ground real-time amplitude fluctuation characteristics into a comprehensive data set, use multivariate analysis methods (such as principal component analysis PCA, cluster analysis) to identify the environmental state change characteristics of each area, record the environmental state change characteristics of each area, form an environmental state change analysis report, and provide it to relevant management departments to ensure that the analysis results can provide a basis for subsequent environmental regulation and management decisions.

[0026] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Step S21: Obtaining a full-scale monitoring video of the construction site and monitoring data of the status of construction equipment on the construction site; performing time-series frame detail enhancement processing on the full-scale monitoring video of the construction site to obtain a detail-enhanced monitoring video; Step S22: performing target visual recognition on the detail-enhanced surveillance video and marking each construction site worker; Step S23: performing image frame segmentation according to each worker on the construction site to obtain multiple worker image frames; Step S24: performing dynamic optical flow tracking on multiple worker image frames one by one to generate a dynamic movement trajectory of each worker; Step S25: performing real-time production action analysis on multiple worker image frames to obtain real-time production action features of each worker; Step S26: Based on the dynamic movement trajectory of each worker, the real-time production action characteristics of each worker are dynamically evolved to form a real-time behavior portrait model for each worker.

[0027] In this embodiment, multiple high-resolution surveillance cameras (such as 1080p or 4K) are deployed to cover the entire construction site to ensure all-round monitoring at different angles and locations. The cameras should have night vision capabilities to work in low-light conditions. The frame rate of video acquisition (such as 30 frames per second) and the length of recording time are set to ensure that the dynamic changes of construction site activities can be recorded. The status of construction equipment is monitored through IoT sensors (such as temperature sensors, pressure sensors, and vibration sensors), and the operating parameters of the equipment (such as working hours, load, and fault alarms) are obtained in real time. The video and monitoring data are stored in the cloud or local server to ensure data security and accessibility. Video processing tools (such as OpenCV, FFmpeg) are used to enhance the time-series frame details of the captured video to improve video quality and enhance detail clarity. Image enhancement techniques such as histogram equalization, denoising, and super-resolution reconstruction are used to improve the visual effect of the video, making the movements of workers and equipment clearer. A suitable target detection algorithm (such as YOLO, SSD, or Faster) is selected. Use R-CNN to perform target visual recognition, identify workers in the video, train the model, use a training set containing worker annotations (such as the COCO dataset) to ensure that the model can accurately identify workers, mark the identified workers, generate surveillance videos with worker identification, record the location information and timestamp of each worker for subsequent analysis, use the target detection results to segment the image frame of each worker, extract the image area of ​​the worker, form multiple worker image frames, ensure that the size of each image frame is suitable for subsequent processing, use a fixed-size image frame (such as 224x224 pixels), store the extracted worker image frames in a specified database or file system for subsequent dynamic optical flow tracking and motion analysis, and select an appropriate optical flow tracking algorithm (such as the Lucas-Kanade optical flow method or the Dense Optical Flow Method). The optical flow tracking system dynamically tracks multiple worker image frames, sets the parameters of optical flow tracking, including window size and pyramid layers, to improve the tracking accuracy, performs optical flow calculation on each worker's image frame, extracts the dynamic movement trajectory of each worker, records the movement path of each worker during the entire monitoring period, including the time series data of position changes, uses action recognition algorithms (such as LSTM, CNN-RNN hybrid model) to perform real-time production action analysis on multiple worker image frames, identifies different production actions of workers (such as carrying, loading and unloading, and resting), ensures that the model is fully trained and can accurately identify different action types, and calculates the duration and frequency of the action, records the real-time production action characteristics of each worker, including action type, frequency, and duration, forms an action feature database, visualizes real-time production action data, displays the working status and production efficiency of each worker through charts, and constructs a real-time behavior portrait model of the worker based on the dynamic movement trajectory and real-time production action characteristics of each worker. The model should include comprehensive information such as the worker's position, action status, and work efficiency.Machine learning methods (e.g., decision trees, random forests) are used to classify workers’ behaviors and identify normal and abnormal behavior patterns.

[0028] In this embodiment, refer to Figure 4 , is a flowchart of detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include: Step S31: classifying the types of construction equipment according to the status monitoring data of construction equipment at the construction site to obtain status monitoring data of different types of equipment; Step S32: performing temperature change trend analysis of the equipment in operation state on the status monitoring data of different equipment types, and generating temperature change trend characteristics of each construction equipment; Step S33: Calculate the real-time operating load of the status monitoring data of different equipment types to obtain the operating load parameters of each equipment; Step S34: performing dynamic load change evolution on the operating load parameters of each device to generate dynamic device load change data; Step S35: Perform equipment comprehensive status mining based on the temperature change trend characteristics of each construction equipment and the dynamic equipment load change data to generate comprehensive status characteristics of each equipment.

[0029] In this embodiment, the temperature changes of construction equipment are monitored in real time through IoT sensors (such as temperature sensors), ensuring that each device has an independent sensor, and recording its temperature data, timestamp and device ID. The data collection frequency is set, for example, temperature data is collected once a minute to ensure the real-time and accuracy of temperature changes. Data cleaning tools (such as Python's Pandas library) are used to clean the collected data, remove missing values ​​and outliers, and identify outliers using the Z-score method. The temperature data is sorted by device type and timestamp to form a structured data set for subsequent analysis. The temperature data of each device is modeled using time series analysis methods (such as ARIMA or exponential smoothing). Select appropriate model parameters to obtain the best fit effect, generate a temperature change trend graph, display the historical change trajectory of the equipment temperature, identify the peak, valley and change range of the temperature, extract the temperature change characteristics of each device, including the average, standard deviation, maximum and minimum values ​​of the temperature, and record the rate of temperature change (such as the change per hour), generate a temperature change trend characteristic report, provide the temperature change of the equipment and potential fault warning information, monitor the operating load of the equipment (such as current, power, pressure, etc.) through sensors, ensure that each device has a corresponding load monitoring device, set the data collection frequency, such as recording load data once a minute for real-time monitoring, clean the load data, remove missing values ​​and outliers, and use IQR (quartile range) to calculate the load data. The load data is sorted by device category and timestamp to form an analyzable data set. The real-time operating load parameters of the device are calculated, such as the power (P=VI, V is voltage, I is current) through the formula, and the results are recorded in the data set. The load benchmark value is set for subsequent performance evaluation and comparison. The operating load data of each device is recorded to form a load feature database. The load change trend is displayed through visualization tools (such as Grafana). An operating load feature report is generated to provide the load status and operating efficiency information of the device. The dynamic system model (such as the state space model or LSTM network) is used to analyze the dynamic changes of the operating load parameters of the device. The model should be able to capture the load changes over time. The change trend of the temperature change of the equipment is determined, and the input (such as historical load data, equipment working status) and output (such as future load prediction) of the model are determined. The load change characteristics are extracted from the dynamic load model, including the load fluctuation range, change rate and periodicity, etc., and dynamic load change data is generated. The load change of each device in a specific time period is recorded, and the appropriate comprehensive state analysis method is selected, such as fuzzy logic, decision tree or machine learning model, to conduct a comprehensive analysis of the temperature change trend and dynamic load change of the equipment, and the input characteristics of the model are determined, including temperature change trend characteristics, dynamic load change data, historical fault records, etc. The constructed model is used to conduct comprehensive state mining for each device to evaluate the operating status of the equipment. The model should be able to identify normal and abnormal states.Record comprehensive status characteristics, including comprehensive status score, potential failure risk and maintenance recommendations.

[0030] In this embodiment, step S4 includes the following steps: Step S41: accurately positioning the construction site environment on the digital environment perception map of the construction site according to the multiple worker image frames to obtain the real-time positioning coordinates of each worker; Step S42: performing panoramic visualization rendering of the real-time behavior portrait model of each worker on the digital environment perception map of the construction site according to the real-time positioning coordinates of each worker, and generating a real-time operation visualization model; Step S43: Analyze the construction equipment distribution on the construction site based on the all-round monitoring video of the construction site to obtain the spatial distribution data of the construction equipment; Step S44: Based on the spatial distribution data of construction equipment, the comprehensive status characteristics of each equipment are used to perform dynamic simulation and interactive modeling on the real-time operation visualization model to construct a dynamic twin model of the construction site.

[0031] In this embodiment, the image frame data of the workers are extracted from the target visual recognition stage, including the position (image frame coordinates) and timestamp of each worker. These data are aligned with the digital environmental perception map of the construction site, and the image frame coordinates are converted into map coordinates through image processing technology (such as perspective transformation). A positioning algorithm based on computer vision (such as image matching, feature point matching algorithm) can be used to combine existing map information and worker image frame data for positioning. Algorithm parameters, such as feature point extraction algorithm (such as SIFT, ORB), are set to improve the accuracy of positioning. The real-time positioning coordinates of each worker are calculated through the algorithm and marked on the digital environmental perception map. The positioning data of each worker, including timestamp, spatial coordinates and corresponding worker ID, are recorded for subsequent analysis. The real-time behavior portrait model of each worker is combined with their positioning coordinates to form a positioning data set containing behavior information, ensuring that the behavior characteristics of each worker (such as action type, working status) correspond to their positioning information. 3D visualization tools (such as Unity and Three.js) are used to render the digital environment perception map of the construction site in full view, and visualization parameters such as viewing angle, lighting, and material are set to make the rendering effect more realistic. A visualization model containing the real-time positioning and behavior information of each worker is generated, and the working status of the workers is dynamically displayed on the map. The working status of each worker and the interaction with the environment are recorded for subsequent analysis and optimization. Construction equipment is extracted and identified from the omnidirectional monitoring video, and target detection algorithms (such as YOLO, Faster The equipment is automatically identified by R-CNN, and its location information and status (such as working / idle) are recorded with timestamps. The identified equipment location is combined with the digital environment perception map of the construction site, and the spatial distribution data of each equipment is calculated to generate a spatial distribution map of the construction equipment. The distribution density and usage of different equipment are marked. According to the spatial distribution data and comprehensive status characteristics of the construction equipment, a dynamic twin model is constructed using simulation tools (such as MATLAB and Simulink). The real-time status and operation information of the equipment are integrated into the model to simulate the dynamic behavior of the equipment in the construction site environment.

[0032] In this embodiment, the specific steps of step S5 are: Step S51: Analyze the worker operation behaviors at multiple time points on the construction site dynamic twin model to obtain the worker operation behavior characteristics at multiple time points; Step S52: Perform deep action fine-grained division on the worker's work behavior characteristics at multiple time points to generate type characteristics for each work behavior; Step S53: identifying the work area based on the work behavior characteristics of the workers at multiple time points to obtain the work area of ​​the construction site at each time point; Step S54: Performing operation safety risk monitoring on the type characteristics of each operation behavior according to the construction site operation area at each time point, and generating abnormal operation safety risk data; Step S55: Make operation behavior safety warning decisions based on abnormal operation safety risk data and build abnormal operation safety warning strategies.

[0033] In this embodiment, worker operation data at multiple time points are extracted from the construction site dynamic twin model, including the workers' positioning coordinates, behavior portraits, and operation status. These data should include the workers' operation status in different time periods (such as different time periods on working days), and the timestamp of the data is accurate to facilitate subsequent analysis. The extracted data is cleaned to remove duplicate values ​​and missing values ​​to maintain the integrity and accuracy of the data. The Python Pandas library is used for data processing to integrate the data at different time points into a unified data frame to facilitate subsequent analysis and comparison. A deep learning model (such as a convolutional neural network CNN or a long short-term memory network LSTM) is used to perform fine-grained division of the workers' operation behaviors. A data set containing worker action labels is used when training the model to ensure that the model can accurately identify different types of operation behaviors. The model predicts and generates type features for each operation behavior, such as "carrying", "loading and unloading", "resting", etc. The type of worker operation behavior at each time point is recorded and integrated into the behavior feature data set. A spatial analysis algorithm (such as K-means clustering or DBSCAN) is used to identify the operation area of ​​the worker operation behavior. The work area is determined based on the work behavior data, and the parameters for area identification, such as clustering radius, are set to ensure the accuracy of area division. The work area of ​​each worker is defined according to the algorithm results, and the corresponding time point information is recorded to form a work area feature data set for subsequent analysis. According to the worker's work behavior type and the work area characteristics, the work safety risk monitoring parameters are set, such as high-risk areas, dangerous action types, etc., and the safety risk threshold is set in combination with historical data for subsequent monitoring. The risk level of the worker's work behavior and the work area is analyzed through monitoring algorithms (such as anomaly detection algorithms, machine learning classifiers) to generate abnormal work safety risk data, and the abnormal risk situation at each time point is recorded for subsequent analysis and decision-making. Based on the abnormal work safety risk data, an early warning decision model is constructed, and machine learning models such as decision trees and random forests are used for training based on historical data. The input features of the model are set, including work behavior type, work area characteristics and safety risk data, and the work behavior and risk data are analyzed in real time to generate a work safety early warning. When the risk data exceeds the set threshold, the early warning is triggered immediately, and the early warning information is recorded, including the warning time, worker ID, work behavior type and risk level.

[0034] In this embodiment, step S6 includes the following steps: Step S61: Perform a comprehensive analysis of the production progress of the construction site dynamic twin model to obtain a real-time production progress comprehensive report; Step S62: performing production progress deviation analysis on the real-time production progress comprehensive report based on the preset production progress log to obtain current production progress deviation data; Step S63: Perform construction process dependency analysis on the construction site dynamic twin model to generate construction process logic; Step S64: dynamically control process resources based on current production progress deviation data and construction process logic, and construct a dynamic process resource control strategy; Step S65: Perform intelligent collaborative management optimization on the abnormal operation safety warning strategy and dynamic process resource control strategy to build a smart construction site collaborative management engine.

[0035] In this embodiment, real-time production progress data, including worker operation time, equipment operation status, material usage, etc., are extracted from the dynamic twin model to ensure that the data timestamp is accurate for time series analysis. Data from various monitoring systems (such as video monitoring and sensor monitoring) are integrated to form a comprehensive production progress data set. Data cleaning tools (such as Python's Pandas library) are used to remove missing values ​​and outliers to ensure data integrity and accuracy. The cleaned data is arranged in chronological order and formatted for subsequent analysis. Statistical analysis methods (such as regression analysis and time series analysis) are used to conduct a comprehensive analysis of the production progress data, and the ARIMA model is used to predict future production. Production progress, calculate key performance indicators (KPIs) such as production efficiency and process completion rate, and record the results in a comprehensive report to generate a real-time production progress comprehensive report, which includes an overview of the production progress, key indicator analysis, trend forecast and other information. Use visualization tools (such as Tableau or Matplotlib) to display production progress trends and key indicators to facilitate management personnel to understand and make decisions. Collect and organize preset production progress logs, including planned production nodes, time arrangements and resource allocation. Ensure that these logs are detailed to facilitate subsequent deviation analysis. Compare the preset production progress with the real-time production progress data, identify the source of the deviation, and calculate the deviation value. Use the formula: Deviation = Actual progress - Preset the progress, record the deviation of each node, and analyze the positive and negative values ​​of the deviation, identify the processes and nodes with large deviations, analyze the reasons (such as equipment failure, low worker efficiency, etc.), and organize the deviation analysis results into a data report, including deviation values, deviation trends and cause analysis, and generate deviation analysis charts to facilitate management to quickly understand the deviation of production progress. Collect relevant data on construction processes, including the start and end time of each process, required resources, dependencies between processes, etc., and organize them into structured data sets for subsequent analysis. Use graph theory analysis methods (such as directed graphs or network analysis) to model process dependencies. Each process is regarded as a node in the graph, and the dependencies are regarded as edges in the graph. The order of processes is determined through algorithms (such as topological sorting). The system can analyze the sequence and dependency of each process, record the analysis results, generate a logical diagram of the construction process, clearly display the dependency and sequence between each process, and facilitate subsequent scheduling and resource allocation. According to the production progress deviation data and the construction process logic, a dynamic resource control strategy is designed, and linear programming or optimization algorithms (such as genetic algorithms) are used to optimize resource allocation. Control parameters are set, including resource availability, process priority, deviation threshold, etc. Control measures are implemented: according to the dynamic control strategy, the allocation of resources is adjusted in real time, and manpower and equipment are allocated to processes with larger deviations first to ensure the recovery of production progress. The implementation of control measures is recorded, and the control effect is monitored in real time. Combined with the abnormal operation safety warning strategy and the dynamic process resource control strategy, an intelligent collaborative management model is constructed.Use rule-based reasoning systems or multi-agent systems to make management decisions, set collaborative management parameters, including priority, response time, and decision thresholds, implement intelligent collaborative management strategies, monitor work conditions and resource status in real time, ensure a balance between safety and progress, collect feedback data after management implementation, and conduct effect evaluation and optimization adjustments.

[0036] In this embodiment, a smart construction site management system based on the Internet of Things is provided, which is used to execute the smart construction site management method based on the Internet of Things as described above, including: The environmental perception module is used to obtain the construction site environmental monitoring data in real time based on multiple IoT sensor nodes; to mine the real-time environmental state changes of the construction site environmental monitoring data, to perform global environmental perception modeling, and to construct a digital environmental perception map of the construction site; The behavior profiling module is used to obtain all-round monitoring videos of the construction site and monitoring data on the status of construction equipment on the construction site; it tracks the dynamic optical flow of each target in the all-round monitoring videos of the construction site, and performs dynamic operation status evolution to build a real-time behavior profiling model for each worker; The equipment status module is used to analyze the temperature change trend of the operating equipment and the dynamic load change evolution based on the status monitoring data of the construction equipment on the construction site, and generate the comprehensive status characteristics of each equipment; The dynamic twin model module is used to perform panoramic visualization rendering and dynamic simulation interactive modeling on the digital environment perception map of the construction site based on the real-time behavior portrait model of each worker and the comprehensive status characteristics of each device, and to build a dynamic twin model of the construction site; The safety warning module is used to analyze the worker's operating behavior at multiple time points on the construction site dynamic twin model, monitor the operating safety risks, and build a safety warning strategy for abnormal operations; The intelligent collaborative management module is used to dynamically control the process resources of the dynamic twin model of the construction site, and to optimize the intelligent collaborative management based on the abnormal operation safety warning strategy to build a smart construction site collaborative management engine.

[0037] The present invention monitors environmental parameters, and managers can adjust the construction methods of the construction site in time or take necessary protective measures to ensure that the workers' working environment meets safety standards, thereby improving work efficiency. Through real-time mining of environmental changes, potential environmental problems (such as air pollution, excessive noise, etc.) can be quickly identified, and data support can be provided for subsequent safety warnings and adjustments. Through dynamic optical flow tracking, the movements and trajectories of each worker are analyzed in real time, and managers have a clear understanding of the work situation of each worker on the construction site, thereby optimizing human resource allocation. The real-time behavior portrait model can effectively identify unsafe work behaviors or abnormal operations of workers (such as not wearing safety equipment, etc.), issue warnings in time and take intervention measures. Through the analysis of the equipment operating status (such as temperature, load, operating efficiency, etc.), the working status of the equipment can be understood in real time, equipment failures can be predicted, maintenance measures can be taken in advance, and downtime losses caused by equipment failures can be avoided. Regular equipment status analysis and predictive maintenance can effectively extend the service life of the equipment, reduce the frequency of failures, and improve the availability and production of equipment. Efficiency. Through the dynamic twin model, various data of the construction site (worker behavior, equipment status, environmental perception, etc.) are presented in a panoramic manner. Managers can view the overall situation of the construction site in real time and effectively grasp the dynamics of the construction site. The twin model combines virtuality and reality. Managers simulate different scenarios in a digital environment and predict the results of different decisions, so as to make more scientific management decisions. By analyzing the workers' work behavior at multiple time points, safety risks can be discovered in time, such as workers entering high-risk areas and irregular operations. The safety warning module can issue warnings in real time to remind relevant personnel to intervene. Through precise safety monitoring and warnings, the occurrence of safety accidents on the construction site can be effectively reduced to ensure the safety of workers. Based on the dynamic twin model and real-time data analysis, process resources are optimized and scheduled in real time. When a certain link lags in progress or resource conflicts, the system can automatically adjust resource allocation to ensure the smooth progress of the overall progress. The intelligent collaborative management system integrates various links, personnel, and equipment on the construction site, improves the collaborative efficiency between departments on the construction site, and avoids problems caused by information lag or uneven resource allocation.

[0038] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0039] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A smart construction site management method based on the Internet of Things, characterized in that: The following steps are involved: Step S1: Acquire construction site environmental monitoring data in real time based on multiple IoT sensor nodes; perform real-time environmental state change mining on the construction site environmental monitoring data, perform global environmental perception modeling, and construct a digital environmental perception map of the construction site; Step S2: Obtain the all-round monitoring video of the construction site and the monitoring data of the status of the construction equipment on the construction site; perform dynamic optical flow tracking on each target in the all-round monitoring video of the construction site, and perform dynamic operation state evolution to build a real-time behavior portrait model for each worker; Step S3: Analyze the temperature change trend of the equipment in operation and the dynamic load change evolution according to the status monitoring data of the construction equipment on the construction site, and generate the comprehensive status characteristics of each equipment; Step S4: Perform panoramic visualization rendering and dynamic simulation interactive modeling on the digital environment perception map of the construction site according to the real-time behavior portrait model of each worker and the comprehensive status characteristics of each device, and build a dynamic twin model of the construction site; Step S5: Analyze the worker's work behavior at multiple time points on the construction site dynamic twin model, monitor the work safety risks, and build an abnormal work safety warning strategy; Step S6: Dynamically regulate the process resources of the dynamic twin model of the construction site, and optimize intelligent collaborative management based on the abnormal operation safety warning strategy to build a smart construction site collaborative management engine.

2. The method for intelligent construction site management based on the Internet of Things according to claim 1 is characterized in that: The specific steps of step S1 are: Step S11: Acquire construction site environment monitoring data in real time based on multiple IoT sensor nodes; Step S12: performing construction site location calculation on multiple IoT sensor nodes to obtain the construction site location coordinates of each sensor; Step S13: performing construction site area type analysis based on the construction site positioning coordinates of each sensor to obtain multiple construction site area type features; Step S14: mining the real-time environmental state change of the construction site environmental monitoring data according to the characteristics of multiple construction site area types to obtain the environmental state change characteristics of each area; Step S15: Perform global environmental perception modeling based on the environmental state change characteristics of each area to build a digital environmental perception map of the construction site.

3. The method for intelligent construction site management based on the Internet of Things according to claim 2 is characterized in that: The specific steps of step S14 are: Based on the construction site positioning coordinates of each sensor, the construction site environmental monitoring data is divided into different construction site area parameters to obtain environmental monitoring parameters of each construction site area, wherein the environmental monitoring parameters include air humidity parameters, air dust concentration parameters, noise intensity and ground vibration amplitude; Conduct time series humidity change analysis on air humidity parameters and generate time series humidity change characteristics; Conduct dust concentration pollution assessment on air dust concentration parameters to generate air quality pollution assessment values; Perform multi-time point discrete fitting on the noise intensity and construct the noise fluctuation intensity curve; Conduct real-time amplitude fluctuation analysis on the ground vibration amplitude to obtain the real-time amplitude fluctuation characteristics of the ground; The environmental state change characteristics of each region are mined by the time series humidity change characteristics, air quality pollution assessment values, noise fluctuation intensity curves and real-time ground amplitude fluctuation characteristics to obtain the environmental state change characteristics of each region.

4. The method for intelligent construction site management based on the Internet of Things according to claim 1 is characterized in that: The specific steps of step S2 are: Step S21: Obtaining a full-scale monitoring video of the construction site and monitoring data of the status of construction equipment on the construction site; performing time-series frame detail enhancement processing on the full-scale monitoring video of the construction site to obtain a detail-enhanced monitoring video; Step S22: performing target visual recognition on the detail-enhanced surveillance video and marking each construction site worker; Step S23: performing image frame segmentation according to each worker on the construction site to obtain multiple worker image frames; Step S24: performing dynamic optical flow tracking on multiple worker image frames one by one to generate a dynamic movement trajectory of each worker; Step S25: performing real-time production action analysis on multiple worker image frames to obtain real-time production action features of each worker; Step S26: Based on the dynamic movement trajectory of each worker, the real-time production action characteristics of each worker are dynamically evolved to form a real-time behavior portrait model for each worker.

5. The method for intelligent construction site management based on the Internet of Things according to claim 1 is characterized in that: The specific steps of step S3 are: Step S31: classifying the types of construction equipment according to the status monitoring data of construction equipment at the construction site to obtain status monitoring data of different types of equipment; Step S32: performing temperature change trend analysis of the equipment in operation state on the status monitoring data of different equipment types, and generating temperature change trend characteristics of each construction equipment; Step S33: Calculate the real-time operating load of the status monitoring data of different equipment types to obtain the operating load parameters of each equipment; Step S34: performing dynamic load change evolution on the operating load parameters of each device to generate dynamic device load change data; Step S35: Perform equipment comprehensive status mining based on the temperature change trend characteristics of each construction equipment and the dynamic equipment load change data to generate comprehensive status characteristics of each equipment.

6. The method for intelligent construction site management based on the Internet of Things according to claim 1 is characterized in that: The specific steps of step S4 are: Step S41: accurately positioning the construction site environment on the digital environment perception map of the construction site according to the multiple worker image frames to obtain the real-time positioning coordinates of each worker; Step S42: performing panoramic visualization rendering of the real-time behavior portrait model of each worker on the digital environment perception map of the construction site according to the real-time positioning coordinates of each worker, and generating a real-time operation visualization model; Step S43: Analyze the construction equipment distribution on the construction site based on the all-round monitoring video of the construction site to obtain the spatial distribution data of the construction equipment; Step S44: Based on the spatial distribution data of construction equipment, the comprehensive status characteristics of each equipment are used to perform dynamic simulation and interactive modeling on the real-time operation visualization model to construct a dynamic twin model of the construction site.

7. The method for intelligent construction site management based on the Internet of Things according to claim 1 is characterized in that: The specific steps of step S5 are: Step S51: Analyze the worker operation behaviors at multiple time points on the construction site dynamic twin model to obtain the worker operation behavior characteristics at multiple time points; Step S52: Perform deep action fine-grained division on the worker's work behavior characteristics at multiple time points to generate type characteristics for each work behavior; Step S53: identifying the work area based on the work behavior characteristics of the workers at multiple time points to obtain the work area of ​​the construction site at each time point; Step S54: Performing operation safety risk monitoring on the type characteristics of each operation behavior according to the construction site operation area at each time point, and generating abnormal operation safety risk data; Step S55: Make operation behavior safety warning decisions based on abnormal operation safety risk data and build abnormal operation safety warning strategies.

8. The method for intelligent construction site management based on the Internet of Things according to claim 1 is characterized in that: The specific steps of step S6 are: Step S61: Perform a comprehensive analysis of the production progress of the construction site dynamic twin model to obtain a real-time production progress comprehensive report; Step S62: performing production progress deviation analysis on the real-time production progress comprehensive report based on the preset production progress log to obtain current production progress deviation data; Step S63: Perform construction process dependency analysis on the construction site dynamic twin model to generate construction process logic; Step S64: dynamically control process resources based on current production progress deviation data and construction process logic, and construct a dynamic process resource control strategy; Step S65: Perform intelligent collaborative management optimization on the abnormal operation safety warning strategy and dynamic process resource control strategy to build a smart construction site collaborative management engine.

9. A smart construction site management system based on the Internet of Things, characterized in that: Used to execute the smart construction site management method based on the Internet of Things as claimed in claim 1, comprising: The environmental perception module is used to obtain the construction site environmental monitoring data in real time based on multiple IoT sensor nodes; to mine the real-time environmental state changes of the construction site environmental monitoring data, to perform global environmental perception modeling, and to construct a digital environmental perception map of the construction site; The behavior profiling module is used to obtain all-round monitoring videos of the construction site and monitoring data on the status of construction equipment on the construction site; it tracks the dynamic optical flow of each target in the all-round monitoring videos of the construction site, and performs dynamic operation status evolution to build a real-time behavior profiling model for each worker; The equipment status module is used to analyze the temperature change trend of the operating equipment and the dynamic load change evolution based on the status monitoring data of the construction equipment on the construction site, and generate the comprehensive status characteristics of each equipment; The dynamic twin model module is used to perform panoramic visualization rendering and dynamic simulation interactive modeling on the digital environment perception map of the construction site based on the real-time behavior portrait model of each worker and the comprehensive status characteristics of each device, and to build a dynamic twin model of the construction site; The safety warning module is used to analyze the worker's operating behavior at multiple time points on the construction site dynamic twin model, monitor the operating safety risks, and build a safety warning strategy for abnormal operations; The intelligent collaborative management module is used to dynamically adjust the process resources of the dynamic twin model of the construction site, and to optimize the intelligent collaborative management based on the abnormal operation safety warning strategy to build a smart construction site collaborative management engine.

Images