Intelligent monitoring and early warning system for potential safety hazards in construction of traffic engineering project

Abstract

The invention relates to the technical field of intelligent construction safety monitoring, and discloses a traffic engineering project construction potential safety hazard intelligent monitoring and early warning system, which comprises a sensing layer, a transmission layer, an edge calculation layer, an intelligent analysis layer, a cloud decision-making layer, an application layer, a collaboration layer and an iteration layer. According to the traffic engineering project construction potential safety hazard intelligent monitoring and early warning system, through multi-dimensional monitoring (personnel / equipment / environment / process) of a sensing layer, a transmission layer security network and real-time feature processing and multi-source fusion of an edge calculation layer, a'total factor-whole process' data closed loop is constructed, construction scene fragmentation risk points (such as equipment faults and personnel violation) are covered, and the safety risk points of the construction scene are also covered. And near-end fusion analysis of data is realized through an AI edge box, accurate identification (such as linkage risks of environmental abrupt change linkage equipment abnormity) is realized in a hidden danger germination stage, the problems of missed judgment and lag in traditional monitoring are solved, and the hidden danger capturing precision and efficiency are greatly improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent construction safety monitoring technology, specifically to an intelligent monitoring and early warning system for construction safety hazards in transportation engineering projects. Background Technology

[0002] Traffic engineering refers to the engineering field of planning, designing, constructing, and maintaining urban and rural transportation systems. This includes the planning and construction of infrastructure such as roads, bridges, tunnels, traffic signal systems, public transportation facilities, and parking facilities. It aims to ensure the safe, efficient, and sustainable development of transportation systems and to solve problems such as traffic congestion, traffic safety, environmental protection, and transportation efficiency in order to improve people's travel convenience and quality of life.

[0003] Currently, the safety management of transportation engineering projects mainly relies on manual inspections, traditional sensor monitoring, and decentralized management systems, which present the following technical bottlenecks:

[0004] 1. Data silo problem: Data from multiple sources such as personnel, machinery, and environment are collected independently, lacking the ability to integrate and analyze them, making it difficult to comprehensively assess overall risks;

[0005] 2. Response lag: The existing system mainly relies on post-event alarms and lacks an active early warning mechanism based on spatiotemporal prediction;

[0006] 3. Poor adaptability to different scenarios: Traditional monitoring methods are difficult to cope with complex geological conditions (such as slope deformation) and dynamic construction environments (such as collaborative operation of machinery groups);

[0007] 4. Low collaboration efficiency: Early warning information is not integrated with emergency command and cross-departmental supervision, and there is a lack of reliable evidence storage mechanisms. Summary of the Invention

[0008] (a) Technical problems to be solved

[0009] To address the shortcomings of existing technologies, this invention provides an intelligent monitoring and early warning system for construction safety hazards in transportation engineering projects. It has advantages such as multimodal data fusion, intelligent prediction and early warning, and full-chain collaborative response. It solves the problems that current construction safety management in transportation engineering projects mainly relies on manual inspections, traditional sensor monitoring, and decentralized management systems, which have the following technical bottlenecks: data silos, response lag, poor scenario adaptability, and low collaborative efficiency.

[0010] The problem.

[0011] (II) Technical Solution

[0012] To achieve the aforementioned goals of multimodal data fusion, intelligent prediction and early warning, and full-chain collaborative response, this invention provides the following technical solution: an intelligent monitoring and early warning system for construction safety hazards in transportation engineering projects, comprising an intelligent monitoring and early warning system, wherein the intelligent monitoring and early warning system includes a perception layer, a transmission layer, an edge computing layer, an intelligent analysis layer, a cloud decision-making layer, an application layer, a collaboration layer, and an iteration layer;

[0013] The perception layer is used to collect construction scene data in all dimensions, covering the core elements of traffic engineering construction safety, monitoring personnel status (fatigue, violation of regulations), mechanical equipment operating parameters (vibration, energy consumption, fault codes), environmental and geological indicators (wind speed, noise, slope displacement, tunnel deformation), and collecting management process data (process progress, safety briefing records), providing the system with a multi-dimensional "data source" for safety hazards.

[0014] The transmission layer is used to build a secure and stable construction data transmission channel. It adapts to complex traffic engineering scenarios (such as network differences in tunnels and open-air work areas) through a hybrid network architecture. Combined with data security mechanisms (encryption and access control), it ensures that the multi-source data (video, sensor, process information) collected by the perception layer is reliably transmitted to the edge computing layer and the cloud.

[0015] The edge computing layer is used for near-end real-time processing of construction data, quickly identifying potential hazards, performing "low-latency" processing on transmission layer data, and through real-time feature processing (extracting abnormal vibration frequencies of equipment and characteristics of personnel violations), combined with multi-source data fusion (associating equipment status + environmental parameters + process progress), it can identify potential safety hazards (such as early signs of equipment failure and trends of minor slope deformation) at the construction front line (such as tunnel entrances and bridge work faces), thus gaining response time for subsequent analysis.

[0016] The intelligent analysis layer is used to deeply mine the value of data and build a dual engine of "knowledge + prediction" for construction safety. Relying on knowledge graphs and risk identification, it links with historical accident databases of traffic engineering and standards (such as bridge construction safety regulations) to infer the chain risks of equipment failure, personnel violations, and sudden environmental changes. Through spatiotemporal prediction modeling (adapted to the linear construction characteristics of traffic engineering, such as the risk diffusion of tunnel excavation), it predicts the development trend of safety hazards (such as whether slope deformation will lead to collapse), providing intelligent analysis support for early warning and decision-making.

[0017] The cloud-based decision-making layer is used for overall planning of construction safety, outputting precise early warnings and response strategies. Based on the results of the intelligent analysis layer, it uses digital twin scenarios to recreate all elements of traffic engineering construction (3D models of bridges, tunnels, and roadbeds + dynamic data) to intuitively present safety risks. Combined with a graded early warning mechanism (distinguishing between general hazards and major risks), it matches the risk levels of different construction stages of traffic engineering (such as pile foundation construction and bridge deck paving) and outputs targeted early warnings (such as pop-up windows and audible and visual alarms) and response suggestions (such as temporarily closing work areas and activating emergency plans).

[0018] The application layer is used for safety early warning and response, realizing "visualized and traceable" control. Through early warning delivery (pushing risk information to on-site personnel and managers), digital visualization and collaboration (AR display of tunnel hazard locations, BIM model association with risk data), it enables construction teams to respond quickly. With the help of blockchain evidence storage, key safety processes of traffic engineering (hazard handling records, equipment inspection logs) are put on the chain to meet compliance traceability requirements (such as quality supervision department verification, insurance claims).

[0019] The collaborative layer is used to break down management barriers among multiple parties in traffic engineering, improve emergency response efficiency, and build a cross-entity collaborative platform (connecting construction units, supervisors, traffic management departments, and emergency rescue). In the event of sudden risks during traffic engineering construction (such as early warning of bridge support collapse), the emergency command module can quickly connect all parties (allocate rescue forces and close traffic sections); it is also adapted to the cross-regional collaboration needs of linear construction in traffic engineering (such as multi-section linkage in long tunnel construction), ensuring information synchronization and coordinated response.

[0020] The iterative layer is used to enable the system to continuously adapt to the dynamic needs of traffic engineering construction. Through model self-iteration, based on traffic engineering construction data feedback (such as early warning accuracy and accident handling effectiveness), the knowledge graph and prediction model are optimized (to adapt to the risk identification of new equipment and new processes). Combined with scenario-based expansion (covering different traffic engineering types such as bridges, tunnels, and roadbeds), the system capabilities are continuously upgraded with the construction scenarios and technological iterations.

[0021] Furthermore, the perception layer includes personnel status monitoring, mechanical equipment monitoring, environmental and geological monitoring, and management process data;

[0022] The personnel status monitoring adopts a smart safety helmet that integrates millimeter-wave radar, UWB high-precision positioning, and physiological sign sensors (heart rate, body temperature);

[0023] The mechanical equipment monitoring system collects real-time operating data of construction machinery (such as excavators and cranes) through IoT sensors (vibration, oil pressure, temperature) to predict the risk of mechanical failure.

[0024] The environmental and geological monitoring uses fiber optic sensor network + InSAR satellite remote sensing to monitor slope and foundation pit deformation, and combines UAV multispectral inspection to identify surface cracks.

[0025] The management process data is integrated with the BIM system to obtain the construction progress plan and associate risk warning thresholds (such as automatically increasing the deformation monitoring frequency during the deep foundation pit excavation stage).

[0026] Furthermore, the transport layer includes a hybrid network architecture and data security mechanisms;

[0027] The hybrid network architecture adopts a "5G + Beidou + wireless Mesh" integrated networking, which is suitable for linear construction of transportation engineering (such as tunnel excavation and long roadbed construction) and complex terrain (mountainous areas and cross-river operations) scenarios.

[0028] The data security mechanism establishes a system of "transmission encryption + access authorization + behavior auditing".

[0029] Furthermore, the edge computing layer includes real-time feature processing and multi-source data fusion;

[0030] The real-time feature processing is accelerated by an FPGA chip, which extracts key features (abnormal equipment vibration frequency, slope displacement rate) in real time for high-frequency data collected in traffic engineering (such as 100Hz sampling by bridge stress sensors and tunnel deformation monitoring).

[0031] The multi-source data fusion establishes a "spatiotemporal association rule base" for traffic engineering construction, which associates equipment status (such as the pressure of the bridge erecting machine's outriggers), environmental parameters (wind speed, temperature), and process progress (bridge deck paving stage).

[0032] Furthermore, the intelligent analysis layer includes knowledge graphs and risk identification and spatiotemporal prediction modeling;

[0033] The knowledge graph and risk identification construct a knowledge graph for traffic engineering construction safety, covering more than 3,000 rules and more than 500 historical accident cases, including the "Technical Specifications for Safety of Highway Engineering Construction".

[0034] The spatiotemporal prediction model integrates the LSTM-Transformer model and trains a "spatial correlation + temporal evolution" prediction model for the spatiotemporal characteristics of linear construction in transportation engineering (such as the diffusion of tunnel deformation along the excavation direction and the accumulation of bridge defects over time).

[0035] Furthermore, the cloud-based decision-making layer includes digital twin scenarios and a tiered early warning mechanism;

[0036] The digital twin scenario builds a 1:1 scale digital twin of traffic engineering construction (including 3D models of bridges, tunnels, and roadbeds), and maps on-site data (equipment GPS trajectory, slope displacement, and steel reinforcement stress) in real time;

[0037] The aforementioned graded early warning mechanism is based on the traffic engineering construction safety risk classification standards (such as the "Regulations on the Supervision and Management of Safety Production of Highway and Waterway Engineering"), and classifies potential hazards into four levels: "blue-yellow-orange-red".

[0038] Furthermore, the application layer includes early warning delivery, digital visualization and collaboration, and blockchain-based evidence storage;

[0039] The early warning system integrates a "multi-terminal + multi-channel" delivery system. For on-site workers, it uses smart safety helmets to broadcast voice messages (such as "K12+300 roadbed section slope displacement exceeds limit, please evacuate").

[0040] The aforementioned digital visualization and collaborative development AR hazard visualization function allows construction workers to "overlay" risk information (such as AR marking of the location of excessive stress on the support structure and associated handling steps) on the tunnel face and bridge operation area through smart glasses;

[0041] The blockchain-based evidence storage is built on a consortium blockchain using Hyperledger Fabric. It stores key safety data for traffic engineering projects (hazard handling records, equipment inspection logs, and emergency drill records) on the blockchain, with the construction unit, supervisor, and quality supervision station serving as consortium nodes.

[0042] Furthermore, the collaboration layer includes a cross-entity collaboration platform and an emergency command module;

[0043] The cross-entity collaborative platform integrates the permissions of multiple departments, including construction units, supervisors, owners, traffic management, and emergency response, through the development of a B / S architecture collaborative platform.

[0044] The emergency command module constructs an emergency command system that combines a digital sand table with contingency plan simulations, and imports emergency plans for traffic engineering construction (such as escape routes for tunnel fires and rescue deployments for bridge collapses).

[0045] Furthermore, the iterative layer includes model self-iteration and scenario-based extension;

[0046] The model self-iteration is used to establish a closed loop of "data feedback - model training - deployment and update";

[0047] The scenario-based extension is used to design a "traffic engineering construction scenario configurator" that pre-sets typical scenario templates such as tunnels, bridges, roadbeds, and interchanges.

[0048] (III) Beneficial Effects

[0049] Compared with the prior art, the present invention provides an intelligent monitoring and early warning system for construction safety hazards in traffic engineering projects, which has the following beneficial effects:

[0050] 1. The intelligent monitoring and early warning system for construction safety hazards in this transportation engineering project constructs a "full-element-full-process" data closed loop through multi-dimensional monitoring at the perception layer (personnel / equipment / environment / process), a secure network at the transmission layer, and real-time feature processing and multi-source fusion at the edge computing layer. This system covers fragmented risk points in the construction scenario (such as equipment failure and personnel violations) and achieves near-end data fusion analysis through AI edge boxes, accurately identifying hazards at the nascent stage (such as the chain risk of sudden environmental changes triggering equipment malfunctions). This solves the problems of "missed detection and delayed detection" in traditional monitoring and significantly improves the accuracy and efficiency of hazard detection.

[0051] 2. The intelligent monitoring and early warning system for construction safety hazards in this transportation engineering project relies on cloud-based decision-making layer digital twin modeling and intelligent analysis layer knowledge graph and spatiotemporal prediction to construct a "visualized, predictable, and collaborative" management system. The digital twin scenario dynamically restores all elements of construction, the spatiotemporal prediction model deduces the development trend of risks, and through application layer early warning (tiered early warning, digital visualization) and collaboration layer cross-entity linkage (emergency command, cross-platform collaboration), the safety early warning is upgraded from "single alarm" to a closed loop of "dynamic deduction + precise handling + responsibility traceability (blockchain evidence storage)," significantly improving the systematicness and effectiveness of construction safety risk management. Attached Figure Description

[0052] Figure 1 This is a schematic diagram of the intelligent monitoring and early warning system of the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] Please see Figure 1 A smart monitoring and early warning system for construction safety hazards in traffic engineering projects, comprising a smart monitoring and early warning system, which includes a perception layer, a transmission layer, an edge computing layer, a smart analysis layer, a cloud decision-making layer, an application layer, a collaboration layer, and an iteration layer;

[0055] The perception layer is used to collect construction scene data in all dimensions, covering the core elements of traffic engineering construction safety, monitoring personnel status (fatigue, violation of regulations), mechanical equipment operating parameters (vibration, energy consumption, fault codes), environmental and geological indicators (wind speed, noise, slope displacement, tunnel deformation), and collecting management process data (process progress, safety briefing records), providing the system with a multi-dimensional "data source" for safety hazards.

[0056] The transport layer is used to build a secure and stable construction data transmission channel. It adapts to complex traffic engineering scenarios (such as network differences in tunnels and open-air work areas) through a hybrid network architecture. Combined with data security mechanisms (encryption and access control), it ensures that the multi-source data (video, sensor, process information) collected by the perception layer is reliably transmitted to the edge computing layer and the cloud.

[0057] The edge computing layer is used for near-end real-time processing of construction data, quickly identifying potential hazards, and performing "low-latency" processing on transmission layer data. Through real-time feature processing (extracting abnormal vibration frequencies of equipment and characteristics of personnel violations), combined with multi-source data fusion (associating equipment status + environmental parameters + process progress), potential safety hazards (such as early signs of equipment failure and trends of minor slope deformation) can be identified near the construction site (such as tunnel entrances and bridge work faces), thus gaining response time for subsequent analysis.

[0058] The intelligent analysis layer is used to deeply mine the value of data and build a dual engine of "knowledge + prediction" for construction safety. Relying on knowledge graphs and risk identification, it links with historical accident databases of traffic engineering and standards (such as bridge construction safety regulations) to infer the chain risks of equipment failure, personnel violations, and sudden environmental changes. Through spatiotemporal predictive modeling (adapted to the linear construction characteristics of traffic engineering, such as the risk diffusion of tunnel excavation), it infers the development trend of safety hazards (such as whether slope deformation will lead to collapse), providing intelligent analysis support for early warning and decision-making.

[0059] The cloud-based decision-making layer is used for overall coordination of construction safety, outputting precise early warnings and response strategies. Based on the results of the intelligent analysis layer, it uses digital twin scenarios to recreate all elements of traffic engineering construction (3D models of bridges, tunnels, and roadbeds + dynamic data) to intuitively present safety risks. Combined with a graded early warning mechanism (distinguishing between general hazards and major risks), it matches the risk levels of different construction stages of traffic engineering (such as pile foundation construction and bridge deck paving) and outputs targeted early warnings (such as pop-up windows and audible and visual alarms) and response suggestions (such as temporarily closing work areas and activating emergency plans).

[0060] The application layer is used for implementing safety early warning and response, achieving "visualized and traceable" control. Through early warning delivery (pushing risk information to on-site personnel and managers), digital visualization and collaboration (AR display of tunnel hazard locations, BIM model association with risk data), it enables construction teams to respond quickly. With the help of blockchain evidence storage, key safety processes of traffic engineering (hazard handling records, equipment inspection logs) are put on the chain to meet compliance traceability requirements (such as quality supervision department verification, insurance claims).

[0061] The collaboration layer is used to break down management barriers among multiple parties in transportation engineering, improve emergency response efficiency, and build a cross-entity collaboration platform (connecting construction units, supervision units, traffic management departments, and emergency rescue). In the event of sudden risks during transportation engineering construction (such as early warning of bridge support collapse), the emergency command module can quickly connect all parties (allocate rescue forces and close traffic sections); it is also adapted to the cross-regional collaboration needs of linear construction in transportation engineering (such as multi-section linkage in long tunnel construction), ensuring information synchronization and coordinated response.

[0062] The iteration layer is used to enable the system to continuously adapt to the dynamic needs of traffic engineering construction. Through model self-iteration, based on traffic engineering construction data feedback (such as early warning accuracy and accident handling effectiveness), the knowledge graph and prediction model are optimized (to adapt to the risk identification of new equipment and new processes). Combined with scenario-based expansion (covering different traffic engineering types such as bridges, tunnels, and roadbeds), the system capabilities are continuously upgraded with the construction scenarios and technology iterations.

[0063] In the case implementation, the perception layer includes personnel status monitoring, mechanical equipment monitoring, environmental and geological monitoring, and management process data;

[0064] Personnel status monitoring utilizes a smart safety helmet that integrates millimeter-wave radar, UWB high-precision positioning, and physiological characteristic sensors (heart rate, body temperature);

[0065] Among them, by real-time monitoring of workers' location, fatigue status and sudden accidents (such as falls), combined with voiceprint recognition technology to detect violations (such as voice alarms when protective equipment is not worn);

[0066] Mechanical equipment monitoring uses IoT sensors (vibration, oil pressure, temperature) to collect real-time operating data of construction machinery (such as excavators and cranes) and predict the risk of mechanical failure.

[0067] Among them, by using Beidou high-precision positioning and RFID tags to build a mechanical "digital fingerprint", the movement trajectory and collision risk of the equipment are tracked, and edge computing nodes are deployed to realize real-time anti-collision warning for collaborative operation of mechanical groups;

[0068] Environmental and geological monitoring employs fiber optic sensor networks and InSAR satellite remote sensing to monitor slope and foundation pit deformation, combined with UAV multispectral inspection to identify surface cracks;

[0069] Among them, weather stations (wind speed, precipitation) and dust / noise sensors are deployed to dynamically assess the impact of extreme weather on construction.

[0070] Management process data is integrated with the BIM system to obtain construction progress plans and associated with risk warning thresholds (such as automatically increasing the deformation monitoring frequency during the deep foundation pit excavation stage);

[0071] This includes integrating management data such as electronic fences and work permits to identify illegal construction activities (such as entering high-risk areas without a permit).

[0072] In the case implementation, the transport layer includes a hybrid network architecture and data security mechanisms;

[0073] The hybrid network architecture adopts a 5G + Beidou + wireless Mesh integrated network, which is suitable for linear construction of traffic engineering (such as tunnel excavation and long roadbed construction) and complex terrain (mountainous areas and cross-river operations).

[0074] Among them, 5G ensures the transmission of high-bandwidth equipment data (such as high-definition video and BIM models), Beidou supplements GNSS positioning data and timing, and wireless mesh addresses network blind spots (such as relay in tunnels), so as to achieve “no dead angle and low latency” transmission of data in the entire construction scenario.

[0075] The data security mechanism establishes a system of "transmission encryption + access authorization + behavior auditing";

[0076] Among these measures, the construction data (personnel location, equipment parameters, geological monitoring) is encrypted and transmitted using national cryptographic algorithms; data access permissions are dynamically allocated based on role permissions (construction worker / supervisor / manager); and the entire process of data uploading, downloading, and modification is audited to prevent the risk of leakage or tampering of confidential data in transportation engineering (such as design parameters of super-large bridges).

[0077] In the case implementation, the edge computing layer includes real-time feature processing and multi-source data fusion;

[0078] Real-time feature processing is achieved by deploying FPGA acceleration chips to extract key features (abnormal equipment vibration frequency, slope displacement rate) in real time for high-frequency data acquisition in traffic engineering (such as 100Hz sampling by bridge stress sensors and tunnel deformation monitoring).

[0079] Among them, the "feature screening-compression-preprocessing" pipeline retains the core information for hazard identification (such as equipment failure features that are free from environmental noise interference), reduces cloud computing pressure, and ensures "millisecond-level" response to hazards such as tunnel collapses and bridge support instability.

[0080] Multi-source data fusion establishes a "spatiotemporal correlation rule base" for traffic engineering construction, linking equipment status (such as the pressure of the bridge erecting machine's outriggers), environmental parameters (wind speed, temperature), and process progress (bridge deck paving stage);

[0081] Among them, by integrating multi-dimensional data through the DS evidence theory, potential hazards in complex scenarios can be identified (such as the probability of failure of the hydraulic system of the bridge erecting machine under high temperature environment, and whether to trigger a work stoppage warning in combination with the paving progress).

[0082] In the case implementation, the intelligent analysis layer includes knowledge graph and risk identification and spatiotemporal prediction modeling;

[0083] Knowledge graph and risk identification construct a knowledge graph for traffic engineering construction safety, covering more than 3,000 rules and 500+ historical accident cases, including the "Technical Specifications for Safety of Highway Engineering Construction".

[0084] Among them, by associating the entity relationships of "personnel-equipment-environment-process" (such as "tunnel excavation team-rock drill-face gas concentration-blasting process"), and through SPARQL reasoning queries, hidden risks such as illegal operations (such as blasting when gas exceeds the standard) and operating equipment with defects (such as abnormal temperature of the shield machine main bearing but not stopping the machine) can be identified.

[0085] Spatiotemporal prediction modeling integrates the LSTM-Transformer model and trains a "spatial correlation + temporal evolution" prediction model for the spatiotemporal characteristics of linear construction in transportation engineering (such as the diffusion of tunnel deformation along the excavation direction and the accumulation of bridge defects over time).

[0086] Among them, by inputting the tunnel surrounding rock deformation monitoring sequence and bridge stress-load historical data, the system predicts the development trend of risks such as collapse and bearing detachment within 24 hours, and outputs a three-dimensional early warning of "risk location-time-affect range", which is adapted to the long-cycle and linearly extended construction characteristics of traffic engineering.

[0087] In the implementation of this case, the cloud-based decision-making layer includes digital twin scenarios and a tiered early warning mechanism;

[0088] A 1:1 scale digital twin of a transportation engineering construction project (including 3D models of bridges, tunnels, and roadbeds) is built, and real-time data (equipment GPS trajectory, slope displacement, and steel reinforcement stress) is mapped.

[0089] Among them, WebGL rendering technology enables construction managers to view risk points in an "immersive" way (such as clicking on the tunnel model to view the location of gas exceeding the limit at the tunnel face and details of related equipment failures), and provides a visual platform for multi-department collaborative decision-making (such as remote consultation between the owner, supervisor, and construction party on the risks of the closure of the super bridge);

[0090] The graded early warning mechanism is based on the classification standards for safety risks in transportation engineering construction (such as the "Measures for Supervision and Management of Safety Production in Highway and Waterway Engineering"), and classifies potential hazards into four levels: "blue-yellow-orange-red".

[0091] Among them, by linking early warning response plans (such as blue warning triggering "equipment inspection + SMS reminder", red warning activating "work stoppage + emergency rescue linkage"), for major risks such as tunnel collapse and bridge hanging basket falling, historical handling plans (such as "XX super bridge hanging basket falling emergency process") are automatically called up, realizing standardized output of "risk level - response measures - plan matching".

[0092] In the implementation of the case, the application layer includes early warning delivery, digital visualization and collaboration, and blockchain evidence storage;

[0093] The early warning system integrates a "multi-terminal + multi-channel" delivery system. For on-site workers, it uses smart safety helmets to broadcast voice messages (such as "K12+300 roadbed section slope displacement exceeds limit, please evacuate").

[0094] Among these measures, the system pushes pop-up windows to managers' mobile apps and warning dashboards to the BIM management platform; and for cross-regional collaborative units (such as remote supervision and emergency command centers), it triggers video conferences and email warnings to ensure that "risk information reaches key personnel within 10 seconds" in dispersed work areas such as tunnels and bridges.

[0095] Digital visualization and collaborative development of AR hazard visualization function: Construction workers can use smart glasses to "overlay" risk information on the tunnel face and bridge operation area (such as AR marking the location of the support stress exceeding the limit and related handling steps);

[0096] Among them, digital twin scenarios support multi-section collaboration (such as simultaneously viewing roadbed settlement risks in multiple construction zones of a highway and adjusting construction plans accordingly);

[0097] The blockchain-based evidence storage is built on a consortium blockchain using Hyperledger Fabric. It stores key safety data for traffic engineering (hazard handling records, equipment inspection logs, and emergency drill records) on the blockchain, with construction units, supervisors, and quality supervision stations serving as consortium nodes.

[0098] Among these features, by ensuring that the data is "tamper-proof and traceable," it is possible to quickly access full-cycle safety management data (such as blockchain-based evidence of 100 inspection records during the construction of a major bridge pile foundation) when dealing with compliance reviews (such as the completion and acceptance of transportation projects).

[0099] In the implementation of the case, the collaboration layer includes a cross-entity collaboration platform and an emergency command module;

[0100] The cross-entity collaboration platform integrates the permissions of multiple departments, including construction units, supervisors, owners, traffic management, and emergency response, through the development of a B / S architecture collaboration platform.

[0101] Among them, a "Construction Safety Collaboration Workbench" is set up to support multiple parties to annotate the rectification plan for hidden dangers online (such as the supervisor marking the defects in the laying of the tunnel waterproof membrane, and the construction party uploading the comparison picture before and after rectification). For cross-regional transportation engineering projects (such as cross-provincial expressways), the data interface of the provincial supervision platform is opened up.

[0102] The emergency command module constructs an emergency command system that combines a "digital sand table" with contingency plan simulations, and imports emergency plans for transportation engineering construction (such as escape routes for tunnel fires and rescue deployments for bridge collapses).

[0103] Among these features, digital twins are used to simulate the evolution of disasters (such as the spread of smoke in tunnel fires and the escape routes of people), and GIS maps are used to coordinate rescue resources (the locations of nearby fire trucks and ambulances). The system automatically generates the optimal solution for "rescue forces-supplies-routes" (such as prioritizing the dispatch of water rescue boats in the event of a major bridge accident), ensuring that the emergency response is "precise and efficient".

[0104] In the implementation of the case, the iteration layer includes model self-iteration and scenario-based expansion;

[0105] Model self-iteration is used to establish a closed loop of "data feedback - model training - deployment and update";

[0106] This involves collecting real-world data from traffic engineering construction scenarios (such as cases of rock deformation in newly built tunnels and new parameters for intelligent bridge monitoring) and automatically labeling potential hazard samples (through manual review and verification using a historical accident database). The intelligent analysis layer model is then updated based on transfer learning (e.g., a risk prediction model adapted to new construction methods).

[0107] The scenario-based extension is used to design a "traffic engineering construction scenario configurator", which pre-sets typical scenario templates such as tunnels, bridges, roadbeds, and interchanges;

[0108] Among them, through "parameterized configuration of scenarios (such as inputting bridge span, tunnel length, and roadbed width)," it automatically adapts to different project requirements (such as safety monitoring of construction of super-large bridges and management of mountain tunnel groups), supports rapid expansion of "access to intelligent connected construction equipment (such as status monitoring of unmanned pavers)" and "adaptation to new infrastructure scenarios (such as collaborative construction and operation of smart highways)," and ensures that the system covers the construction safety requirements of all types and stages of transportation engineering.

[0109] In summary, this intelligent monitoring and early warning system for construction safety hazards in transportation engineering projects constructs a "full-element, full-process" data closed loop through multi-dimensional monitoring at the perception layer (personnel / equipment / environment / process), a secure network at the transmission layer, and real-time feature processing and multi-source fusion at the edge computing layer. This system covers fragmented risk points in the construction scenario (such as equipment failure and personnel violations) and achieves near-end data fusion analysis through AI edge boxes, accurately identifying hazards at the nascent stage (such as the chain risk of sudden environmental changes triggering equipment malfunctions). This solves the problems of "missed detection and delayed detection" in traditional monitoring, significantly improving the accuracy and efficiency of hazard detection.

[0110] Furthermore, relying on cloud-based decision-making layer digital twin modeling and intelligent analysis layer knowledge graph and spatiotemporal prediction, a "visualized-predictable-collaborative" management and control system is constructed. The digital twin scenario dynamically restores all elements of construction, the spatiotemporal prediction model deduces the development trend of risks, and through application layer early warning (tiered early warning, digital visualization) and collaboration layer cross-entity linkage (emergency command, cross-platform collaboration), the safety early warning is upgraded from "single alarm" to a closed loop of "dynamic deduction + precise handling + responsibility traceability (blockchain evidence storage)," which significantly improves the systematicness and effectiveness of construction safety risk management and control.

[0111] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0112] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An intelligent monitoring and early warning system for construction safety hazards in traffic engineering projects, comprising an intelligent monitoring and early warning system, characterized in that: The intelligent monitoring and early warning system includes a perception layer, a transmission layer, an edge computing layer, an intelligent analysis layer, a cloud decision-making layer, an application layer, a collaboration layer, and an iteration layer. The perception layer is used to collect construction scene data in all dimensions, covering the core elements of traffic engineering construction safety, monitoring personnel status (fatigue, violation of regulations), mechanical equipment operating parameters (vibration, energy consumption, fault codes), environmental and geological indicators (wind speed, noise, slope displacement, tunnel deformation), and collecting management process data (process progress, safety briefing records), providing the system with a multi-dimensional "data source" for safety hazards. The transmission layer is used to build a secure and stable construction data transmission channel. It adapts to complex traffic engineering scenarios (such as network differences in tunnels and open-air work areas) through a hybrid network architecture. Combined with data security mechanisms (encryption and access control), it ensures that the multi-source data (video, sensor, process information) collected by the perception layer is reliably transmitted to the edge computing layer and the cloud. The edge computing layer is used for near-end real-time processing of construction data, quickly identifying potential hazards, performing "low-latency" processing on transmission layer data, and through real-time feature processing (extracting abnormal vibration frequencies of equipment and characteristics of personnel violations), combined with multi-source data fusion (associating equipment status + environmental parameters + process progress), it can identify potential safety hazards (such as early signs of equipment failure and trends of minor slope deformation) at the construction front line (such as tunnel entrances and bridge work faces), thus gaining response time for subsequent analysis. The intelligent analysis layer is used to deeply mine the value of data and build a dual engine of "knowledge + prediction" for construction safety. Relying on knowledge graphs and risk identification, it links with historical accident databases of traffic engineering and standards (such as bridge construction safety regulations) to infer the chain risks of equipment failure, personnel violations, and sudden environmental changes. Through spatiotemporal prediction modeling (adapted to the linear construction characteristics of traffic engineering, such as the risk diffusion of tunnel excavation), it infers the development trend of safety hazards (such as whether slope deformation will lead to collapse), providing intelligent analysis support for early warning and decision-making. The cloud-based decision-making layer is used for overall planning of construction safety, outputting precise early warnings and response strategies. Based on the results of the intelligent analysis layer, it uses digital twin scenarios to recreate all elements of traffic engineering construction (3D models of bridges, tunnels, and roadbeds + dynamic data) to intuitively present safety risks. Combined with a graded early warning mechanism (distinguishing between general hazards and major risks), it matches the risk levels of different construction stages of traffic engineering (such as pile foundation construction and bridge deck paving) and outputs targeted early warnings (such as pop-up windows and audible and visual alarms) and response suggestions (such as temporarily closing work areas and activating emergency plans). The application layer is used for safety early warning and response, realizing "visualized and traceable" control. Through early warning delivery (pushing risk information to on-site personnel and managers), digital visualization and collaboration (AR display of tunnel hazard locations, BIM model association with risk data), it enables construction teams to respond quickly. With the help of blockchain evidence storage, key safety processes of traffic engineering (hazard handling records, equipment inspection logs) are recorded on the chain to meet compliance traceability requirements (such as quality supervision department verification, insurance claims). The collaborative layer is used to break down management barriers among multiple parties in traffic engineering, improve emergency response efficiency, and build a cross-entity collaborative platform (connecting construction units, supervisors, traffic management departments, and emergency rescue). In the event of sudden risks during traffic engineering construction (such as early warning of bridge support collapse), the emergency command module can quickly connect all parties (allocate rescue forces and close traffic sections); it is also adapted to the cross-regional collaboration needs of linear construction in traffic engineering (such as multi-section linkage in long tunnel construction), ensuring information synchronization and coordinated response. The iterative layer is used to enable the system to continuously adapt to the dynamic needs of traffic engineering construction. Through model self-iteration, based on traffic engineering construction data feedback (such as early warning accuracy and accident handling effectiveness), the knowledge graph and prediction model are optimized (to adapt to the risk identification of new equipment and new processes). Combined with scenario-based expansion (covering different traffic engineering types such as bridges, tunnels, and roadbeds), the system capabilities are continuously upgraded with the construction scenarios and technological iterations.

2. The intelligent monitoring and early warning system for construction safety hazards in traffic engineering projects according to claim 1, characterized in that: The perception layer includes personnel status monitoring, mechanical equipment monitoring, environmental and geological monitoring, and management process data; The personnel status monitoring adopts a smart safety helmet that integrates millimeter-wave radar, UWB high-precision positioning, and physiological sign sensors (heart rate, body temperature); The mechanical equipment monitoring system collects real-time operating data of construction machinery (such as excavators and cranes) through IoT sensors (vibration, oil pressure, temperature) to predict the risk of mechanical failure. The environmental and geological monitoring uses fiber optic sensor network + InSAR satellite remote sensing to monitor slope and foundation pit deformation, and combines UAV multispectral inspection to identify surface cracks. The management process data is integrated with the BIM system to obtain the construction progress plan and associate risk warning thresholds (such as automatically increasing the deformation monitoring frequency during the deep foundation pit excavation stage).

3. The intelligent monitoring and early warning system for construction safety hazards in transportation engineering projects according to claim 1, characterized in that: The transport layer includes a hybrid network architecture and data security mechanisms; The hybrid network architecture adopts a "5G + Beidou + wireless Mesh" integrated networking, which is suitable for linear construction of transportation engineering (such as tunnel excavation and long roadbed construction) and complex terrain (mountainous areas and cross-river operations) scenarios. The data security mechanism establishes a system of "transmission encryption + access authorization + behavior auditing".

4. The intelligent monitoring and early warning system for construction safety hazards in traffic engineering projects according to claim 1, characterized in that: The edge computing layer includes real-time feature processing and multi-source data fusion; The real-time feature processing is accelerated by an FPGA chip, which extracts key features (abnormal equipment vibration frequency, slope displacement rate) in real time for high-frequency data collected in traffic engineering (such as 100Hz sampling by bridge stress sensors and tunnel deformation monitoring). The multi-source data fusion establishes a "spatiotemporal association rule base" for traffic engineering construction, which associates equipment status (such as the pressure of the bridge erecting machine's outriggers), environmental parameters (wind speed, temperature), and process progress (bridge deck paving stage).

5. The intelligent monitoring and early warning system for construction safety hazards in traffic engineering projects according to claim 1, characterized in that: The intelligent analysis layer includes knowledge graph and risk identification, and spatiotemporal prediction modeling; The knowledge graph and risk identification construct a knowledge graph for traffic engineering construction safety, covering more than 3,000 rules and more than 500 historical accident cases, including the "Technical Specifications for Safety of Highway Engineering Construction". The spatiotemporal prediction model integrates the LSTM-Transformer model and trains a "spatial correlation + temporal evolution" prediction model to address the spatiotemporal characteristics of linear construction in transportation engineering (such as the diffusion of tunnel deformation along the excavation direction and the accumulation of bridge defects over time).

6. The intelligent monitoring and early warning system for construction safety hazards in traffic engineering projects according to claim 1, characterized in that: The cloud-based decision-making layer includes digital twin scenarios and a tiered early warning mechanism; The digital twin scenario builds a 1:1 scale digital twin of traffic engineering construction (including 3D models of bridges, tunnels, and roadbeds), and maps on-site data (equipment GPS trajectory, slope displacement, and steel reinforcement stress) in real time; The aforementioned graded early warning mechanism is based on the traffic engineering construction safety risk classification standards (such as the "Regulations on the Supervision and Management of Safety Production of Highway and Waterway Engineering"), and classifies hidden dangers into four levels: "blue-yellow-orange-red".

7. The intelligent monitoring and early warning system for construction safety hazards in traffic engineering projects according to claim 1, characterized in that: The application layer includes early warning delivery, digital visualization and collaboration, and blockchain evidence storage; The early warning system integrates a "multi-terminal + multi-channel" delivery system. For on-site workers, it uses voice broadcasts via smart safety helmets (e.g., "Slope displacement exceeds limits at K12+300 roadbed section, please evacuate"). The aforementioned digital visualization and collaborative development AR hazard visualization function allows construction workers to "overlay" risk information (such as AR marking of the location of excessive stress on the support structure and associated handling steps) on the tunnel face and bridge operation area through smart glasses; The blockchain-based evidence storage is built on a consortium blockchain using Hyperledger Fabric. It stores key safety data for traffic engineering projects (hazard handling records, equipment inspection logs, and emergency drill records) on the blockchain, with the construction unit, supervisor, and quality supervision station serving as consortium nodes.

8. The intelligent monitoring and early warning system for construction safety hazards in traffic engineering projects according to claim 1, characterized in that: The collaboration layer includes a cross-entity collaboration platform and an emergency command module; The cross-entity collaborative platform integrates the permissions of multiple departments, including construction units, supervisors, owners, traffic management, and emergency response, through the development of a B / S architecture collaborative platform. The emergency command module constructs an emergency command system that combines a "digital sand table" with "emergency plan simulation" and imports emergency plans for traffic engineering construction (such as escape routes for tunnel fires and rescue deployments for bridge collapses).

9. The intelligent monitoring and early warning system for construction safety hazards in traffic engineering projects according to claim 1, characterized in that: The iterative layer includes model self-iteration and scenario-based extension; The model self-iteration is used to establish a closed loop of "data feedback - model training - deployment and update"; The scenario-based extension is used to design a "traffic engineering construction scenario configurator" that pre-sets typical scenario templates such as tunnels, bridges, roadbeds, and interchanges.

Images