Construction site risk perception and dynamic management and control system integrating BIM and intelligent safety belt

Abstract

The invention discloses a construction site risk perception and dynamic management and control system integrating BIM and an intelligent safety belt, and relates to the technical field of building construction safety monitoring. Comprising a data acquisition and behavior identification module, an AI risk behavior evaluation and scoring module, a BIM model fusion and space mapping module, a three-dimensional risk thermodynamic diagram generation module, an AI behavior learning and trend prediction module and a management linkage module. According to the method, through the intelligent safety belt and the AI behavior recognition module, dangerous behaviors such as high-place operation are monitored and judged in real time, and the timeliness and initiative of safety management are greatly improved. According to the method, precise spatial mapping is realized in combination with the BIM model. According to the method, the dynamic three-dimensional risk thermodynamic diagram is generated, the three-dimensional thermodynamic diagram can be dynamically updated according to the risk behavior frequency and grade of different areas and different components, the potential safety hazard distribution and change trend are visually displayed, and a manager can visually perceive and make a judgment conveniently.

Description

Technical Field

[0001] This invention relates to the field of construction safety monitoring technology, specifically a construction site risk perception and dynamic control system that integrates BIM and intelligent safety belts. Background Technology

[0002] With the continuous improvement of informatization and intelligentization in the construction industry, utilizing BIM (Building Information Modeling) for project visualization management, safety coordination, and schedule control has become an industry trend. Simultaneously, the integrated application of emerging technologies such as smart wearable devices, artificial intelligence, and the Internet of Things on construction sites is becoming increasingly widespread, driving the transformation from traditional experience-based management to data-driven intelligent management. In high-risk construction scenarios such as high-altitude operations and edge work, how to predict and proactively intervene in safety risks based on personnel behavior data has become an important research direction in the field of construction safety. Some studies have attempted to assist in identifying safety hazards through video surveillance and sensor networks, but comprehensive risk perception and visualization at the component level, behavior level, and spatiotemporal coupling have not yet been achieved.

[0003] However, existing construction site safety management technologies still have several shortcomings. First, at the data acquisition level, traditional monitoring methods lack the ability to perceive the individual behavior of workers with fine detail, failing to achieve high-precision capture of key behaviors such as safety belt fastening status, swaying amplitude, and inertial stillness. Second, in terms of risk identification, existing systems mostly rely on rule-based judgment or image detection algorithms, which are difficult to adapt to the diverse behavioral types and prominent temporal characteristics in complex construction environments. Third, in terms of spatial mapping, behavioral risk information is difficult to bind with BIM components with high precision, resulting in spatial ambiguity and unclear management positioning. Fourth, in terms of risk response mechanisms, existing technologies have failed to form a closed-loop process of risk identification, task distribution, and educational intervention; inspection and educational tasks mostly rely on manual scheduling, resulting in delayed responses and unquantifiable effects. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by this invention is that existing construction site safety management and risk early warning methods have problems such as insufficient granularity of behavioral data collection, low integration of AI recognition and spatial structure, and lack of closed-loop feedback in risk handling mechanisms. It also addresses how to achieve component-level safety behavior perception, dynamic risk heat map generation, and joint control of inspection and education based on smart wearable devices and BIM models.

[0006] To address the aforementioned technical problems, this invention provides the following technical solution: a construction site risk perception and dynamic control system integrating BIM and intelligent safety belts, comprising a data acquisition and behavior recognition module for real-time monitoring of workers' harness status, posture changes, sway amplitude, inertial stillness, and tension changes, and uploading the data to an edge computing node via a wireless network as input to an AI risk recognition module, while simultaneously sharing location information with a BIM model fusion and spatial mapping module; and an AI risk behavior assessment and scoring module for identifying behaviors such as not wearing a safety belt, suspended stillness, and low-hanging-high-use based on the collected raw behavior data using an AI model, calculating the duration of these behaviors, and combining personnel type and construction stage to calculate component-level risk scores, outputting the results to a BIM model mapping and heatmap rendering module; The IM model fusion and spatial mapping module integrates personnel location data with BIM component coordinates, binds identified behaviors to components, extracts component ID, type, floor, and responsible job information, and generates structured risk records for use by the heatmap rendering and AI learning modules. The 3D risk heatmap generation module uses red, orange, yellow, and green to mark risk levels in the 3D model based on BIM model spatial mapping data. The AI ​​behavior learning and trend prediction module automatically generates component-level inspection tasks and safety education content based on high-risk areas in the heatmap. The management linkage module integrates historical risk labels, component characteristics, and environmental parameters to train the AI ​​model to predict future risk trends in specific areas for early warning and task scheduling, and provides dynamic learning data for the AI ​​risk identification module.

[0007] As a preferred embodiment of the construction site risk perception and dynamic control system integrating BIM and intelligent safety belts described in this invention, the data acquisition and behavior recognition module includes a posture sensor, a tension sensor, and a positioning module; the posture sensor is used to detect the angle of personnel movement and the amplitude of swaying; the tension sensor is used to monitor the fastening status and impact force changes of the safety belt; and the positioning module is used to obtain the real-time position coordinates of the workers in three-dimensional space.

[0008] As a preferred embodiment of the construction site risk perception and dynamic control system integrating BIM and intelligent safety belts described in this invention, the AI ​​risk behavior assessment and scoring module includes a hybrid neural network model based on a deep time series structure, which performs feature extraction and classification on the input multimodal behavior data, and outputs behavior type, behavior occurrence duration, personnel level, and component risk score.

[0009] As a preferred embodiment of the construction site risk perception and dynamic control system integrating BIM and intelligent safety belts described in this invention, the BIM model fusion and spatial mapping module includes: binding behavior recognition results to the center or boundary coordinates of BIM model components based on UWB positioning or inertial navigation data, and outputting a structured dataset of component number, risk type, risk score and construction stage information.

[0010] As a preferred embodiment of the construction site risk perception and dynamic management system integrating BIM and intelligent safety belts described in this invention, the three-dimensional risk heat map generation module includes: supporting the visualization and coloring of component risk levels, dynamically updating the heat map content according to the time dimension, and viewing historical risk trajectories and predicted trends through an interactive interface.

[0011] As a preferred embodiment of the construction site risk perception and dynamic control system integrating BIM and intelligent safety belts described in this invention, the AI ​​behavior learning and trend prediction module includes generating a risk trend curve for a specific component area in the future time period based on heat map rendering results and historical behavior records, and recommending safety inspection tasks and educational content.

[0012] As a preferred embodiment of the construction site risk perception and dynamic control system integrating BIM and intelligent safety belts described in this invention, the management linkage module includes: dynamically adjusting AI model parameters by learning from historical risk events, construction environment parameters, and behavioral consequences; and recording and providing feedback on inspection results and education completion status to achieve closed-loop management of risk identification, task generation, and behavioral intervention.

[0013] Another objective of this invention is to provide a construction site risk perception and dynamic control system that integrates BIM and intelligent safety belts. This system can solve the problem of low integration between AI recognition and spatial structure in current construction site safety management and risk early warning methods through an AI risk behavior assessment and scoring module.

[0014] As a preferred embodiment of the construction site risk perception and dynamic control method integrating BIM and intelligent safety belts described in this invention, the method includes: collecting multimodal behavioral data of workers through intelligent safety belts and uploading it to edge computing nodes for processing; using AI models to classify, identify, and score the behavioral data, and binding and mapping it with the spatial location of BIM model components; and generating a three-dimensional risk heat map based on the identification results to automatically push and provide feedback on risk-driven inspection and education tasks.

[0015] Another object of the present invention is to provide a construction site risk perception and dynamic management device that integrates BIM and intelligent safety belts, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the construction site risk perception and dynamic management method that integrates BIM and intelligent safety belts.

[0016] Another object of the present invention is to provide a construction site risk perception and dynamic management storage medium that integrates BIM and intelligent safety belts, wherein a computer program is stored thereon, and when the computer program is executed by a processor, the steps of the construction site risk perception and dynamic management method integrating BIM and intelligent safety belts are implemented.

[0017] The beneficial effects of this invention are as follows: The construction site risk perception and dynamic control system integrating BIM and intelligent safety belts provided by this invention achieves real-time monitoring and judgment of dangerous behaviors such as working at heights through intelligent safety belts and AI behavior recognition modules, significantly improving the timeliness and proactivity of safety management. This invention combines BIM models to achieve precise spatial mapping; risky behaviors are not only identified but also accurately mapped to construction components or spatial locations, overcoming the problem that traditional two-dimensional maps cannot accurately reflect the actual construction structure, and providing support for precise management based on component dimensions.

[0018] This invention generates a dynamic three-dimensional risk heat map. The system can dynamically update the three-dimensional heat map according to the frequency and level of risk behaviors in different areas and components, and visualize the distribution and changing trends of safety hazards, making it easier for managers to intuitively perceive and make judgments.

[0019] This invention introduces an AI learning mechanism to form risk prediction. By learning the correlation between historical behavioral data and environmental factors (such as weather, construction stage, and job distribution), the system can predict the future risk trend of a specific area and provide data support for early warning.

[0020] This invention establishes a closed-loop linkage between safety inspections and education. After identifying high-risk areas, the system automatically generates inspection tasks and educational reminders, and pushes these to the relevant responsible persons based on BIM location data, achieving a complete closed loop of risk discovery, rectification, learning, and recording. This invention improves regulatory efficiency and reduces reliance on manual labor. By replacing a large amount of manual inspection and judgment with a platform-based approach, it can improve management efficiency, reduce omissions and misjudgments, and maintain continuous and stable operation in complex construction scenarios.

[0021] This invention is highly adaptable and can be extended to various construction scenarios. The system is applicable to various construction forms such as high-rise buildings, bridges, industrial plants, and tunnels. It has good scalability and secondary development capabilities, which is conducive to its widespread application. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a module diagram of a construction site risk perception and dynamic control system that integrates BIM and intelligent safety belts, as provided in Embodiment 1 of the present invention. Detailed Implementation

[0024] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0025] Example 1, referring to Figure 1 As an embodiment of the present invention, a construction site risk perception and dynamic control system integrating BIM and intelligent safety belt is provided, including: a data acquisition and behavior recognition module 100, an AI risk behavior assessment and scoring module 200, a BIM model fusion and spatial mapping module 300, a three-dimensional risk heat map generation module 400, an AI behavior learning and trend prediction module 500, and a management linkage module 600.

[0026] Furthermore, the data acquisition and behavior recognition module 100 monitors the workers' harness status, posture changes, sway amplitude, inertial stillness, and tension changes in real time, and uploads this data to the edge computing node via a wireless network as input to the AI ​​risk recognition module. It also shares location information with the BIM model spatial mapping module 300. The AI ​​risk behavior assessment and scoring module 200, based on the collected raw behavior data, uses an AI model to identify behaviors such as not wearing a safety belt, being suspended in mid-air, and low-hanging-high-use, calculates the duration of these behaviors, and combines this with personnel type and construction stage to calculate component-level risk scores. The output results are then transmitted to the BIM model mapping and heatmap rendering module. The BIM model fusion and spatial mapping module 300 integrates personnel location data. Data and BIM component coordinates bind identification behavior to components, extracting component ID, type, floor, and responsible job information to generate structured risk records for use by the heatmap rendering and AI learning modules; the 3D risk heatmap generation module 400 marks risk levels in the 3D model using four colors: red, orange, yellow, and green, based on the spatial mapping data of the BIM model; the AI ​​behavior learning and trend prediction module 500 automatically generates component-level inspection tasks and safety education content based on high-risk areas in the heatmap; the management linkage module 600 integrates historical risk labels, component characteristics, and environmental parameters to train an AI model to predict future risk trends in specific areas for early warning and task scheduling, and provides dynamic learning data for the AI ​​risk identification module.

[0027] It should be noted that this module is used to realize the real-time perception and data reporting of the behavior of workers performing high-altitude operations at construction sites, and is the perception foundation layer for the system to realize risk identification and spatial mapping. By integrating this module into the smart safety belt worn by workers, the system can perform multi-dimensional and continuous monitoring of the key behavioral states of personnel, ensuring the integrity and reliability of the collected data.

[0028] Specifically, the data acquisition and behavior recognition module 100 integrates multiple sensing sub-modules and communication units, including but not limited to: a seatbelt fastening status sensor: used to identify whether the seatbelt is correctly fastened to the anchor point; a three-axis gyroscope and accelerometer: to identify personnel posture, tilt angle, and vibration characteristics; a tension sensing module: to monitor instantaneous tension changes and identify risky actions such as low fastening and high use, and violent shaking; an inertial measurement unit (IMU): to determine whether there is an abnormal state of prolonged suspension and stillness; a UWB or BLE positioning module: to achieve three-dimensional spatial positioning with centimeter-level accuracy; and an edge computing unit (optional): to support data compression, anomaly filtering, and local preprocessing.

[0029] The module supports multiple wireless communication methods (such as Wi-Fi 6, LoRa, NB-IoT, and BLEMesh) to upload behavioral data to edge computing nodes or cloud systems in real time. The sampling frequency is set to 50Hz by default, and can be automatically increased to 100Hz to improve recognition accuracy when abnormal behavior occurs. It has a network outage caching and retransmission mechanism to ensure data continuity.

[0030] The system adopts a standardized data encapsulation format. Each data record includes: (personnel ID, equipment number, timestamp, component coordinates (X,Y,Z), action label, action amplitude, angular velocity, attachment status, tension value, and stationary time).

[0031] The module supports integration with BIM models and spatial mapping modules 300, sharing personnel location information to provide a spatial foundation for subsequent component binding and heat map generation. To ensure data traceability, the system supports personnel and equipment binding management: operators can bind themselves to specific equipment by scanning a code through the App. The binding information includes name, job type, work group, construction stage, etc., and the binding relationship runs through the entire data lifecycle.

[0032] In addition, the module has a good anti-interference and safety design, adopts EMC standard shielded circuit, supports AES128 or SM4 data encryption, and has a low-power sleep mechanism and drop alarm linkage function to ensure that the system operates stably, safely and reliably in complex construction environments.

[0033] It should also be noted that the AI ​​risk behavior assessment and scoring module 200 is used to intelligently identify and quantitatively assess the raw multimodal behavior data uploaded by the data collection and behavior recognition module 100. This is a core processing step in realizing risk assessment of on-site work behavior. By integrating a pre-trained AI behavior recognition model, the module can accurately classify the behavior types of workers and output component-level risk scores based on relevant factors, providing effective analysis results for subsequent spatial binding and heatmap rendering.

[0034] Specifically, the AI ​​risk behavior assessment and scoring module 200 uses a hybrid neural network model based on a deep time series structure (such as a fusion of CNN and LSTM structures) to extract features and classify the input time series data. The types of behaviors identified by the system include, but are not limited to: not wearing a seatbelt, remaining suspended in the air for an extended period, using the vehicle at a low speed while it is in use, and high-frequency vertical vibration.

[0035] Furthermore, each behavior type is assigned a unique behavior code. The model can simultaneously output the duration of this behavior. Types of workers involved (such as ordinary workers, team leaders, special operations personnel, etc.) and their respective construction phases. .

[0036] After identification, the system uses the aforementioned factors to call a risk scoring function to quantitatively calculate the component-level behavioral risk. The structure of this scoring function is as follows: , in, To identify and score the risks of components, For behavior type coefficients, For behavior coding values ​​(e.g., 1 = not wearing a seatbelt, (e.g., suspended in mid-air and stationary) This is the duration coefficient of the behavior. The cumulative duration of this type of behavior within a unit of time (unit: seconds). The risk sensitivity level of personnel is categorized as follows: 1 for ordinary workers, 2 for special operations, and 3 for responsible persons and managers. The sensitivity adjustment coefficient for the role. Code the construction phase (e.g.) Erecting, Formwork (steel structure work) This represents the impact factor during the construction phase. These are the phase period amplitude modulation parameters.

[0037] The output value is defined in In conventional applications, it can be standardized as ;when Low-risk areas, only recorded; when The system will issue a warning if the area is classified as medium-risk; when... In high-risk areas, a joint inspection and education mechanism will be immediately triggered.

[0038] It should be noted that the AI ​​risk behavior assessment and scoring module 200 has continuous learning capabilities, which can continuously optimize the model recognition accuracy during task execution and behavior feedback. The module synchronously transmits the recognition results to the BIM model fusion and spatial mapping module 300 and the 3D risk heat map generation module 400, and supports dynamic data interaction with the AI ​​behavior learning and trend prediction module 500 to achieve real-time model updates and accuracy enhancement.

[0039] Through the AI ​​risk behavior assessment and scoring module 200, the system has achieved automatic identification of personnel behavior at the construction site, component-level quantitative scoring, and efficient data structuring processing, thereby improving the data support capability of safety management and the level of intelligent behavior perception.

[0040] It should also be noted that the BIM model fusion and spatial mapping module 300 is used to establish a precise spatial binding relationship between the identified personnel behaviors and the specific components in the BIM model, realizing a structured mapping from personnel behavior identification to construction component risk modeling. It is the core bridging unit for the association between risk information and building space.

[0041] The BIM model fusion and spatial mapping module 300 receives positioning information from the data acquisition and behavior recognition module 100 and risk score results output by the AI ​​risk behavior assessment and scoring module 200. Through coordinate projection and spatial matching algorithms, it binds behavioral events to specific components in the BIM model. The system supports centimeter-level positioning technologies such as UWB and BLE. Combining the three-dimensional spatial boundary coordinates of BIM components, the system performs the following binding process for personnel behavior points: Coordinate transformation and component matching: The system will transfer the collected behavioral coordinates... Mapping to the coordinate system used in the BIM model, and setting a spatial buffer threshold based on the geometric boundaries of the components. The system determines whether a specific component has been matched; component attribute extraction: once the binding is successful, the system extracts information such as the component's ID, type, floor, responsible job, and current construction stage; structured record output: standardized component risk information items are generated: component ID, risk score R, behavior type B, duration D, personnel type P, construction stage C, spatial coordinates (x, y, z), and timestamp T.

[0042] Furthermore, the generated data records will be simultaneously provided to the 3D risk heat map generation module 400 for visualization and coloring processing, and will also serve as key spatial dimension data input for the subsequent training and trend modeling of the AI ​​behavior learning and trend prediction module 500.

[0043] Furthermore, the module supports aggregated analysis of multiple risk behaviors of the same component within a continuous time period. It calculates the time-weighted average risk score or maximum risk score of the component using a sliding window to dynamically display risk evolution trends. The module also supports index retrieval and spatial hierarchical management by component ID, facilitating rapid location and coordinated push notifications in subsequent heatmap rendering and task distribution.

[0044] Through the BIM model fusion and spatial mapping module 300, the system achieves high-precision fusion of behavioral information and BIM structural model, providing a spatial semantic basis for component-level risk visualization and safety decision-making.

[0045] It should be noted that this module is used to dynamically map component-level risk scores into the BIM 3D model in a visual form, forming a 3D risk heat map with clear spatial distribution, visible temporal evolution, and user-friendly management interaction. It is the core interface layer for realizing the visualization and proactive management of risks on the construction site.

[0046] The 3D risk heat map generation module 400 receives structured risk record data from the BIM model fusion and spatial mapping module 300, and assigns a risk score to each component based on the component ID and spatial coordinates. The model is rendered using color grading. The module employs a four-color grading strategy: red-orange-yellow-green. The risk grading rules are as follows: Green (safe); Yellow (Medium risk); Orange (High Risk); Red (extremely high risk).

[0047] In addition, the 3D risk heat map generation module 400 supports the following key functions: Dynamic playback of time axis: The system can generate time series heat map animations based on the timestamp field in the component risk data, realizing the retrospective and predictive playback of the risk evolution process at the construction site; Component focus and information query: Users can click on any risk component in the model interface to view its risk behavior type, frequency of occurrence, personnel information and responsible unit; Spatial layer filtering: Supports filtering and display by floor, construction stage, risk level, team responsibility and other conditions to achieve management focus; Linkage triggering mechanism: When a component continuously exceeds the high risk threshold, the module will trigger an event signal, driving the AI ​​behavior learning and trend prediction module 500 to automatically generate inspection tasks and educational pushes.

[0048] The 3D risk heat map generation module 400 supports integration with the native model of the BIM platform, has strong real-time rendering capabilities and graphical interactivity, and can be deployed on PC clients, BIM visualization platforms or engineering large screen display systems.

[0049] With the 3D risk heat map generation module 400, managers can intuitively identify high-risk areas, grasp risk trends, and achieve a visual closed-loop support from risk triggering logic to intervention execution path. This is an important manifestation of the system's "digital risk perception" capability.

[0050] It should also be noted that the AI ​​behavior learning and trend prediction module 500 is used to build an adaptive risk evolution model based on heat map annotation results and historical risk behavior data, and to automatically generate the prediction and intervention tasks for future risk trends of specific component areas at the construction site. It is a key logical unit for the system to realize the forward-moving risk management and behavior improvement feedback loop.

[0051] The AI ​​behavior learning and trend prediction module 500 receives high-risk component information from the 3D risk heat map generation module 400 and component spatial attribute data from the BIM model fusion and spatial mapping module 300. It combines historical behavior tag data, personnel behavior trajectory data and external environmental parameters (such as weather, construction stage, work density, etc.) to establish a risk trend prediction model through AI model.

[0052] The module adopts a multi-factor learning mechanism. The training model mainly includes the following input elements: behavior type sequence (such as a component being "unattached" for 3 consecutive days); component structural parameters and operation attributes (such as formwork platform, cantilever steel structure); type and distribution of workers; environmental disturbance factors such as weather and wind speed; and feedback information on the completion of inspection and training (as the basis for model iteration and optimization).

[0053] By aggregating the above multimodal information, the system outputs the following two types of results: Future risk trend prediction results: For a specified component, predict the risk score change trend within a future time period (such as 24 hours or 72 hours), output the value range and rate of change, and mark the potential risk increase time window.

[0054] Automatic Intervention Task Generation: If the predicted score of a component is about to exceed the high-risk threshold, the system will automatically generate an inspection task, including the component number, inspection content, and suggested time window. Simultaneously, it will match relevant educational resources (micro-lecture videos, operation specification PDFs, typical case animations, etc.) and push them to the responsible operator's terminal. The task information will be further pushed and executed through the management linkage module 600, and upon completion, the inspection and educational results will be returned as feedback input for the module's continuous learning and model optimization.

[0055] In addition, the AI ​​behavior learning and trend prediction module 500 has self-optimization capabilities, which can dynamically adjust model parameters based on the intervention effect (such as the degree of risk reduction and changes in the rate of repeated violations) to continuously improve the prediction accuracy and intervention timeliness.

[0056] The AI ​​behavior learning and trend prediction module 500 enables a shift from passive identification to proactive prediction in risk management, and establishes an AI-driven task assignment and personalized education recommendation mechanism, which is the core control layer for promoting a closed loop of intelligent safety governance at construction sites.

[0057] The management linkage module 600 is used to issue inspection tasks for high-risk component areas on the construction site, push safety education content, record behavior improvement and update AI model feedback. It is the execution layer and backtracking layer of the system to build a closed-loop logic of "identification-intervention-feedback-optimization".

[0058] The management linkage module 600 receives the task generation results output by the AI ​​behavior learning and trend prediction module 500, arranges tasks according to component ID and risk type, and accurately pushes inspection plans and educational resources to the corresponding responsible persons, operators or management terminals to ensure that risk identification can be quickly transformed into action paths for on-site intervention and personnel awareness enhancement.

[0059] The 600-function management linkage module includes: Linked execution of safety inspection tasks: Inspection tasks include target component IDs, suggested inspection periods, and key inspection items (such as attachment point settings and scaffold stability); inspection tasks are issued via mobile app or wearable terminal, supporting real-time location tracking, on-site photography, video uploads, and voice recording; after inspection, the system records the task status and handling details as a basis for subsequent evaluation and feedback. Intelligent recommendation and push of safety education content: The system extracts keywords from risk behavior tags and matches them with micro-lesson videos, operating procedures, standard manuals, or simulation animations; educational resources are precisely pushed according to the operator's role and can be displayed on various devices such as terminal apps, on-site electronic screens, and voice terminals; the system records each operator's learning time, completion status, and simple assessment results, and includes them in their personal safety file. Behavior improvement assessment and data feedback mechanism: The system compares the changes in component risk scores and the frequency of violations before and after the implementation of inspection and education tasks to evaluate the rectification effect; if the risk level drops significantly, it is automatically recorded as "effective intervention"; if multiple education sessions are ineffective or the risk score increases, the model sensitivity is increased, the scoring weight is adjusted, and the area is marked as a key supervision area; all results will be fed back as feedback data to the AI ​​risk behavior assessment and scoring module 200 and the AI ​​behavior learning and trend prediction module 500, forming a complete data cycle and model relearning path.

[0060] The management linkage module 600 realizes a closed loop throughout the entire process, from high-risk identification to precise inspection and personalized education, and then to behavior effectiveness tracking and AI retraining. It is a key execution and evaluation fulcrum that supports the long-term self-optimization and self-enhancement of the system.

[0061] If the functions of this invention are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0062] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0063] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0064] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc. It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0065] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0066] Example 2, one embodiment of the present invention, provides a method for risk perception and dynamic control of construction sites that integrates BIM and smart safety belts. The method includes collecting multimodal behavioral data of workers through smart safety belts and uploading it to an edge computing node for processing; using an AI model to classify, identify and score the behavioral data, and binding and mapping it with the spatial location of BIM model components; and generating a three-dimensional risk heat map based on the identification results to automatically push and provide feedback on risk-driven inspection and education tasks.

[0067] This embodiment also provides a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements a construction site risk perception and dynamic control method that integrates BIM and intelligent safety belts as proposed in the above embodiment.

[0068] This embodiment also provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements a construction site risk perception and dynamic control method that integrates BIM and intelligent safety belts as proposed in the above embodiment.

[0069] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0070] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0071] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0072] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0073] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A construction site risk perception and dynamic control system integrating BIM and intelligent safety belts, characterized in that, include: The data acquisition and behavior recognition module is used to monitor the worker's attachment status, posture changes, sway amplitude, inertial stillness and tension changes in real time, and uploads the data to the edge computing node via wireless network as input to the AI ​​risk recognition module. At the same time, it shares positioning information with the BIM model fusion and spatial mapping module. The AI ​​risk behavior assessment and scoring module is used to identify behaviors such as not wearing a safety belt, being suspended in the air and being used at a low position based on the collected raw behavior data using an AI model, count the duration of the behavior, and calculate the component-level risk score by combining personnel type and construction stage. The output results are transmitted to the BIM model mapping and heat map rendering module. The BIM model fusion and spatial mapping module is used to fuse personnel positioning data with BIM component coordinates, bind identified behaviors to components, extract component ID, type, floor and responsible job information, and generate structured risk records for use by the heat map rendering and AI learning modules. The 3D risk heat map generation module is used to mark the risk level in the 3D model using four colors: red, orange, yellow, and green, based on the spatial mapping data of the BIM model. The AI ​​behavior learning and trend prediction module is used to automatically generate component-level inspection tasks and safety education content based on high-risk areas in the heat map. The management linkage module is used to integrate historical risk labels, component characteristics and environmental parameters to train AI models to predict future risk trends in specific areas, for early warning and task scheduling, and to provide dynamic learning data for the AI ​​risk identification module.

2. The construction site risk perception and dynamic control system integrating BIM and intelligent safety belts as described in claim 1, characterized in that: The data acquisition and behavior recognition module includes, Attitude sensor, force sensor, and positioning module; The posture sensor is used to detect the angle of a person's movement and the amplitude of swaying. The tension sensor is used to monitor the fastening status and impact force changes of the seat belt; The positioning module is used to obtain the real-time position coordinates of the operator in three-dimensional space.

3. The construction site risk perception and dynamic control system integrating BIM and intelligent safety belts as described in claim 1 or 2, characterized in that: The AI-based risk behavior assessment and scoring module includes, A hybrid neural network model based on deep time series structure is used to extract features and classify the input multimodal behavioral data. The output includes behavior type, duration of behavior occurrence, personnel level, and component risk score.

4. The construction site risk perception and dynamic control system integrating BIM and intelligent safety belts as described in claim 3, characterized in that: The BIM model fusion and spatial mapping module includes, Based on UWB positioning or inertial navigation data, the behavior recognition results are bound to the center or boundary coordinates of BIM model components, and a structured dataset containing component number, risk type, risk score, and construction stage information is output.

5. The construction site risk perception and dynamic control system integrating BIM and intelligent safety belts as described in claim 1, 2, or 4, characterized in that: The three-dimensional risk heatmap generation module includes, It supports visual coloring of component risk levels and dynamically updates heatmap content according to the time dimension. Historical risk trajectories and predicted trends can be viewed through an interactive interface.

6. The construction site risk perception and dynamic control system integrating BIM and intelligent safety belts as described in claim 5, characterized in that: The AI ​​behavior learning and trend prediction module includes, Based on heatmap rendering results and historical behavior records, a multi-factor modeling approach is used to generate risk trend curves for specific component areas over future time periods, and to recommend safety inspection tasks and educational content.

7. The construction site risk perception and dynamic control system integrating BIM and intelligent safety belts as described in claim 1, 2, 4 or 6, characterized in that: The management linkage module includes, By learning from historical risk events, construction environment parameters, and behavioral consequences, the AI ​​model parameters are dynamically adjusted, and the inspection results and education completion status are recorded and fed back, thus achieving closed-loop management of risk identification, task generation, and behavioral intervention.

8. A method for risk perception and dynamic control of construction sites integrating BIM and intelligent safety belts, employing the construction site risk perception and dynamic control system integrating BIM and intelligent safety belts as described in any one of claims 1 to 7, characterized in that: This includes collecting multimodal behavioral data of workers through smart safety belts and uploading it to edge computing nodes for processing; AI models are used to classify, identify, and score behavioral data, and then they are bound and mapped to the spatial locations of BIM model components. The system generates a 3D risk heat map based on the identification results, enabling the automatic push and feedback of risk-driven inspection and education tasks.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the construction site risk perception and dynamic control method integrating BIM and intelligent safety belt as described in claim 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the construction site risk perception and dynamic control method integrating BIM and intelligent safety belt as described in claim 8.

Images