Tunnel safety monitoring, prevention and control method and system based on Internet of Things

Abstract

The invention discloses a tunnel safety monitoring, prevention and control method and system based on the Internet of Things, and relates to the technical field of tunnel safety management and control. Various Internet of Things sensor nodes are deployed in a tunnel in a distributed manner, and corresponding virtual sensor nodes are set in a virtual model; collecting and associating multi-dimensional environment data parameters, structural health data parameters and traffic state data parameters in real time; the method comprises the following steps: pre-constructing a plurality of dangerous emergencies, and determining associated sensor nodes and dynamic data parameters under event evolution; in the real-time monitoring process, event position performance characteristics of the virtual model and sensor dynamic data parameter changes are analyzed, the closeness degree of real-time performance and dangerous performance is quantified, and early warning is carried out according to the closeness degree. According to the invention, a digital twinning technology is utilized to fuse historical accidents and multi-physical field simulation to construct a danger expression library, early accurate identification and early warning of various emergencies such as tunnel fire, collapse, flood, toxic gas leakage and vehicle collision are realized, and the monitoring accuracy and timeliness are effectively improved.

Description

A Tunnel Safety Monitoring and Control Method and System Based on the Internet of Things Technical Field

[0001] This invention relates to the field of tunnel safety management and control technology, and in particular to a tunnel safety monitoring and control method and system based on the Internet of Things. Background Technology

[0002] As a crucial transportation infrastructure, the safe operation of tunnels is directly related to the safety of people's lives and property and economic and social stability. However, existing tunnel safety monitoring systems mainly rely on distributed sensors to monitor single or local parameters such as temperature, smoke, vibration, and water level at thresholds, combined with video surveillance for manual or simple rule-based judgment. Such systems suffer from poor timeliness, low accuracy, and frequent false alarms and missed alarms when facing various sudden dangerous events such as fires, collapses, floods, toxic gas leaks, and vehicle collisions. Specifically, traditional methods struggle to effectively integrate multi-dimensional environmental data, structural health data, and traffic status data, making it impossible to accurately identify and match patterns in the spatiotemporal evolution of disasters. Furthermore, the lack of a hazardous event database based on historical accidents and multiphysics simulations results in insufficient ability to capture early signs of complex emergencies, hindering proactive and tiered accurate early warning systems. Summary of the Invention

[0003] The purpose of this invention is to provide a tunnel safety monitoring and control method and system that can provide relatively accurate early warnings.

[0004] This invention discloses a tunnel safety monitoring and control method based on the Internet of Things (IoT), comprising: step S100, constructing a virtual tunnel structure model based on the tunnel's design structure; step S200, distributing several IoT sensor nodes within the tunnel and setting corresponding virtual sensor nodes within the corresponding virtual tunnel structure model to collect multi-dimensional environmental data parameters, structural health data parameters, and traffic status data parameters, and associating them with the corresponding virtual sensor nodes; step S300, constructing several dangerous emergencies, determining the IoT sensor nodes associated with the dangerous emergencies based on the dangerous emergencies, and determining the dynamic data parameters of the corresponding IoT sensor nodes based on the event details of the dangerous emergencies to obtain the dangerous manifestations of the virtual tunnel structure model, and constructing a dangerous manifestation library based on all dangerous manifestations of the virtual tunnel structure model; step S400, analyzing the event location characteristics and sensor dynamic data parameter changes under the real-time performance of the virtual tunnel structure model, finding matching dangerous manifestations in the dangerous manifestation library, analyzing the proximity of event location characteristics and sensor dynamic data parameter changes to dangerous manifestations, and issuing early warnings based on proximity manifestations.

[0005] In some embodiments disclosed in this invention, the method for constructing several dangerous emergencies includes: step S201, collecting historical accident data parameters and historical monitoring data parameters of the tunnel, wherein the historical accident data parameters include accident type, location of occurrence and consequences of accident, and the historical monitoring data parameters include multidimensional environmental data parameters, structural health data parameters and traffic status data parameters for a period of time before and after the accident; step S202, configuring the virtual tunnel structure model based on the historical accident data parameters and historical monitoring data parameters to obtain the dangerous performance of the virtual tunnel structure model.

[0006] In some embodiments disclosed in this invention, the method for constructing several dangerous emergencies includes: step S203, constructing a virtual simulation model for a virtual tunnel structure model, including setting multiple predefined hazard sources and triggering conditions in the virtual tunnel structure model, wherein the hazard sources include fire, collapse, flood, toxic gas leakage and vehicle collision; step S204, using fluid dynamics simulation, structural mechanics simulation and traffic flow simulation tools, simulating the evolution process of each hazard source under different locations, intensities and propagation conditions, to obtain the dynamic data parameter sequence of virtual sensor nodes and the positional changes of the virtual tunnel structure model during the evolution of the dangerous event; step S205, identifying the definition type, location, evolution result, dynamic data parameter sequence of different virtual sensor nodes and positional changes of the virtual tunnel structure model as dangerous manifestations.

[0007] In some embodiments of the present invention, the method for simulating the evolution of various hazard sources under different locations, intensities, and propagation conditions includes: Step S2041, for fire hazard sources, using computational fluid dynamics (CFD) simulation tools combined with a heat transfer model, simulating smoke diffusion, temperature field distribution, and visibility changes under fire source location, combustion intensity, oxygen concentration, and ventilation conditions, obtaining dynamic data parameter sequences of temperature, smoke concentration, and CO concentration for the corresponding virtual sensor nodes; Step S2042, for collapse hazard sources, using finite element analysis (FEA) structural mechanics simulation tools, simulating stress concentration, crack propagation, and local / overall deformation under surrounding rock stress, support structure strength, and geological disturbance conditions, obtaining dynamic data parameter sequences of vibration, strain, and displacement for the corresponding virtual sensor nodes; Step S2043, for... For flood hazards, shallow water equations or multiphase fluid dynamics simulation tools are used to simulate water intrusion, water depth, and flow velocity distribution under rainfall intensity, water source location, and drainage system conditions, obtaining dynamic data parameter sequences of water level, humidity, and flow rate for the corresponding virtual sensor nodes; in step S2044, for toxic gas leak hazards, gas diffusion models combined with fluid dynamics simulation tools are used to simulate the diffusion path and concentration distribution under leak point, gas type, concentration, and ventilation conditions, obtaining dynamic data parameter sequences of toxic gas concentration for the corresponding virtual sensor nodes; in step S2045, for vehicle collision hazards, microscopic traffic flow simulation tools are used to simulate the evolution of traffic congestion and secondary accidents under collision location, vehicle type, speed, and load conditions, obtaining dynamic data parameter sequences of vibration, pressure, and traffic flow for the corresponding virtual sensor nodes.

[0008] In some embodiments disclosed in this invention, the method for analyzing the event location performance characteristics and sensor data parameter performance of a virtual tunnel structure model includes: step S401, setting several location points within the virtual tunnel structure model, and analyzing the dynamic data parameter performance of each virtual sensor node; if the dynamic data parameter performance of a virtual sensor node falls within a first range of interest, then marking the virtual sensor node as an interest point, and associating the marked virtual sensor node with location points within a preset distance range, and identifying the associated location points as interest location points; step S402, associating interest location points that are close to each other to form an interest location point cluster, and recording the changes in the interest location point cluster as event location performance characteristics.

[0009] In some embodiments disclosed in this invention, the method for finding matching hazardous manifestations in a hazardous manifestation library includes: step S403, constructing an indexing mechanism for hazardous manifestations in the hazardous manifestation library, performing feature label analysis on the clusters of location points of interest in real-time events and the dynamic data parameter manifestations, and using the indexing mechanism to compare and analyze the feature labels to determine several hazardous manifestations; wherein, the indexing mechanism includes determining the cluster center point of the cluster of location points of interest in each hazardous manifestation, judging several edge distances between the cluster edge and the center point, calculating the edge distance variance, identifying the cluster center point, edge distance variance, and the sensor type corresponding to the virtual sensor node as a distinguishing label group, and classifying the hazardous manifestations according to the distinguishing label group.

[0010] In some embodiments disclosed in this invention, the method for finding matching dangerous behaviors in the dangerous behavior database includes: step S404, using an indexing mechanism to determine several dangerous behaviors, comparing the cluster of attention location points between each dangerous behavior and a real-time event to determine the equivalent parameters of the event location, and comparing the dynamic data parameter performance between each dangerous behavior and a real-time event to determine the equivalent parameters of the data parameters; step S405, based on the equivalent parameters of the event location and the equivalent parameters of the data parameters, determining the closeness between the real-time event and the dangerous behavior, if the closeness is greater than or equal to a preset value, then selecting the corresponding dangerous behavior for interpretation and warning.

[0011] In some embodiments disclosed in this invention, the method for determining the degree of proximity includes: step S4051, performing frame-by-frame overlap comparison of the cluster of location points of interest between the dangerous manifestation and the real-time event, determining the spatial region jointly mapped by the two, and determining the center point of the spatial region, radiating several detection lines outward from the center point of the region, determining the edge intersection points of each detection line and the edges of different clusters of location points of interest, calculating the edge distance between the edge intersection points, calculating the detection line reference distance from the center point of the region to the first intersection edge, and calculating the distance ratio between the edge distance and the detection line reference distance. Based on the overall performance of the distance ratio corresponding to each detection line, determining the single-frame event position equivalent parameter, and based on the single-frame event of consecutive frames... Step S4052: Align and compare each relative dynamic data parameter sequence of each frame between the hazard manifestation and the real-time event to obtain a dynamic data parameter difference sequence. Analyze the data parameter differences in the dynamic data parameter difference sequence that are greater than or equal to a preset value, and record them as valid data parameter differences. Based on the proportion of valid data parameter differences relative to the parameter differences in the dynamic data parameter difference sequence, determine the single-frame data parameter equivalence parameter. Based on the single-frame data parameter equivalence parameter of consecutive frames, determine the data parameter equivalence parameter between the hazard manifestation and the real-time event. Step S4052: Determine the degree of proximity based on the event location equivalence parameter and the data parameter equivalence parameter.

[0012] In some embodiments disclosed in this invention, the expression for calculating the degree of proximity is: Where J represents the degree of proximity. The first degree of proximity conversion coefficient, The second degree of proximity conversion coefficient, This is the distance ratio analysis function corresponding to the x-th detection line in the i-th frame. Based on the preset interval to which the distance ratio belongs, it outputs the basic value of the corresponding event position equivalent parameter, where X is the total number of detection lines and n is the number of frames participating in the comparison. This is a function for analyzing the percentage difference of the u-th parameter in the i-th frame. Based on the preset interval to which the percentage difference of the parameter belongs, it outputs the basic value of the corresponding data parameter equivalent parameter. Let be the weight coefficient of the virtual sensor node corresponding to the u-th parameter difference, where U is the total number of virtual sensors participating in the comparison.

[0013] In some embodiments disclosed in this invention, an IoT-based tunnel safety monitoring and control system is also disclosed, comprising: a first module for constructing a virtual tunnel structure model based on the tunnel's design structure; a second module for distributively deploying several IoT sensor nodes within the tunnel, setting corresponding virtual sensor nodes within the corresponding virtual tunnel structure model, collecting multi-dimensional environmental data parameters, structural health data parameters, and traffic status data parameters, and associating them with the corresponding virtual sensor nodes; a third module for constructing several dangerous emergencies, determining the IoT sensor nodes associated with the dangerous emergencies based on the dangerous emergencies, determining the dynamic data parameters of the corresponding IoT sensor nodes based on the event details of the dangerous emergencies, obtaining the dangerous manifestations of the virtual tunnel structure model, and constructing a dangerous manifestation library based on all dangerous manifestations of the virtual tunnel structure model; and a fourth module for analyzing the event location characteristics and sensor dynamic data parameter changes under the real-time performance of the virtual tunnel structure model, finding matching dangerous manifestations in the dangerous manifestation library, analyzing the proximity of event location characteristics and sensor dynamic data parameter changes to dangerous manifestations, and issuing warnings based on the proximity manifestations.

[0014] This invention discloses a tunnel safety monitoring and control method and system based on the Internet of Things (IoT), belonging to the field of tunnel safety management technology. It involves the distributed deployment of multiple IoT sensor nodes within the tunnel, with corresponding virtual sensor nodes set in a virtual model. This allows for the real-time collection and correlation of multi-dimensional environmental data parameters, structural health data parameters, and traffic status data parameters. Multiple hazardous emergencies are pre-constructed, and the associated sensor nodes and dynamic data parameters under the evolution of these events are determined. During real-time monitoring, the characteristics of the event location in the virtual model and the changes in the dynamic data parameters of the sensors are analyzed to quantify the degree of proximity between real-time performance and hazardous events, and early warnings are issued accordingly. This invention utilizes digital twin technology to integrate historical accidents and multiphysics simulations to construct a hazardous performance database, enabling early and accurate identification and warning of various emergencies such as tunnel fires, collapses, floods, toxic gas leaks, and vehicle collisions, effectively improving the accuracy and timeliness of monitoring.

[0015] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0016] Figure 1 is a flowchart of a method for monitoring and controlling tunnel safety based on the Internet of Things disclosed in an embodiment of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0018] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings and specific embodiments. It should be understood that the preferred embodiments described herein are only for illustration and explanation of the present invention and should not be construed as limiting the scope of protection of the present invention. Those skilled in the art can make some non-essential improvements and adjustments based on the following content of the present invention. In the present invention, unless otherwise expressly specified and limited, the technical terms used in the present invention should have the ordinary meaning understood by those skilled in the art.

[0019] Example: This invention discloses a tunnel safety monitoring and control method based on the Internet of Things, including: step S100, constructing a virtual tunnel structure model based on the tunnel design structure.

[0020] The core principle of step S100 is to construct a highly realistic virtual tunnel structure model based on the actual design structure of the tunnel using digital twin technology. This model serves as the digital foundation of the entire system, fully replicating the spatial layout, geometric features, and structural attributes of the physical tunnel. It provides a unified virtual environment for subsequent sensor data mapping, hazard simulation, and real-time event analysis. This virtual mirroring ensures a precise correspondence between the physical tunnel and the digital model, achieving spatial unification and visual representation of multi-dimensional data. This avoids the problems of data dispersion and spatial inconsistency in traditional monitoring systems, laying the foundation for the accurate operation of the entire system.

[0021] Step S200: Distribute several types of IoT sensor nodes in the tunnel and set corresponding virtual sensor nodes in the corresponding virtual tunnel structure model to collect multi-dimensional environmental data parameters, structural health data parameters and traffic status data parameters, and associate them with the corresponding virtual sensor nodes.

[0022] The principle behind step S200 lies in establishing a real-time data connection between the physical tunnel and the virtual model. This is achieved by distributing multiple IoT sensor nodes in key areas of the tunnel and correspondingly setting virtual sensor nodes in the virtual model, enabling continuous collection and precise correlation of multi-dimensional environmental data, structural health data, and traffic status data. This step employs a distributed monitoring strategy to form a comprehensive sensor network, ensuring that the collected data is mapped to the corresponding locations in the virtual model in real time. This provides reliable multi-source data support for the construction of a hazard manifestation database and real-time event identification, overcoming the limitations of single-sensor monitoring and achieving complete data fusion and dynamic synchronization within the digital twin environment.

[0023] Step S300: Construct several dangerous emergencies, and based on the dangerous emergencies, determine the IoT sensor nodes associated with the dangerous emergencies, and based on the event details of the dangerous emergencies, determine the dynamic data parameters of the corresponding IoT sensor nodes to obtain the dangerous manifestations of the virtual tunnel structure model, and based on all the dangerous manifestations of the virtual tunnel structure model, construct a dangerous manifestation library.

[0024] Step S300 works by pre-constructing virtual evolution processes of typical hazardous emergencies through various methods, generating corresponding hazard manifestations, and compiling them into a standardized library. Specifically, this includes reproducing real-world event manifestations using historical accident data, and simulating the development of different hazard sources under various locations and conditions in a virtual model using multiphysics simulation tools. This process acquires dynamic data changes from associated sensor nodes and spatial performance characteristics of the model, defining these characteristics as hazard manifestations. The core of this step is to establish a comprehensive manifestation library covering multiple disaster types, providing a reliable benchmark for real-time monitoring, enabling a shift from passive monitoring to proactive prediction, and enhancing the system's pattern recognition capabilities for complex emergencies.

[0025] Step S400: Analyze the event location performance characteristics and sensor dynamic data parameter changes under the real-time performance of the virtual tunnel structure model, find the matching dangerous performance in the dangerous performance database, analyze the similarity between the event location performance characteristics and sensor dynamic data parameter changes and the dangerous performance, and issue an early warning based on the similarity performance.

[0026] Step S400 operates on the principle of real-time dynamic analysis and similarity matching. Through in-depth analysis of the event location characteristics and sensor dynamic data changes in the virtual model, similar manifestations are quickly filtered from a hazard manifestation database. Subsequently, through geometric comparison of location features and difference analysis of data sequences, the degree of similarity between the real-time event and the hazard manifestations in the database is quantified, triggering corresponding early warnings. This step integrates spatial location features with multi-dimensional temporal data features, enabling accurate identification of early signs of disasters, significantly improving the accuracy and timeliness of early warnings, and reducing false alarms and missed alarms.

[0027] In some embodiments disclosed in this invention, the method for constructing several dangerous emergencies includes: step S201, collecting tunnel historical accident data parameters and historical monitoring data parameters, wherein the historical accident data parameters include accident type, location of occurrence and consequences of accident, and the historical monitoring data parameters include multidimensional environmental data parameters, structural health data parameters and traffic status data parameters for a period of time before and after the accident.

[0028] Step S202: Based on historical accident data parameters and historical monitoring data parameters, configure the virtual tunnel structure model to obtain the dangerous performance of the virtual tunnel structure model.

[0029] In some embodiments disclosed in this invention, the method for constructing several dangerous emergencies includes: step S203, constructing a virtual simulation model for a virtual tunnel structure model, including setting a variety of predefined hazard sources and triggering conditions in the virtual tunnel structure model, wherein the hazard sources include fire, collapse, flood, toxic gas leakage and vehicle collision.

[0030] Step S204: Using fluid dynamics simulation, structural mechanics simulation, and traffic flow simulation tools, the evolution process of each hazard source under different locations, intensities, and propagation conditions is simulated to obtain the dynamic data parameter sequence of virtual sensor nodes during the evolution of the hazard event and the positional changes of the virtual tunnel structure model.

[0031] Step 205: Define the type of hazard source, its location, evolution results, dynamic data parameter sequences of different virtual sensor nodes, and positional changes of the virtual tunnel structure model as hazard manifestations.

[0032] In some embodiments disclosed in this invention, the method for simulating the evolution of various hazard sources under different locations, intensities, and propagation conditions includes: step S2041, for fire hazard sources, using computational fluid dynamics (CFD) simulation tools combined with heat transfer models to simulate smoke diffusion, temperature field distribution, and visibility changes under fire source location, combustion intensity, oxygen concentration, and ventilation conditions, to obtain dynamic data parameter sequences of temperature, smoke concentration, and CO concentration for the corresponding virtual sensor nodes.

[0033] The core principle of step S2041 is to use computational fluid dynamics (CFD) simulation tools combined with a heat transfer model to perform high-precision simulations of the evolution of fire hazards under different fire source locations, combustion intensities, oxygen concentrations, and ventilation conditions. The focus is on reproducing the dynamic processes of smoke diffusion, temperature field distribution, and visibility changes. This method utilizes fluid dynamics and thermodynamics principles to simulate the spatiotemporal evolution of a fire from ignition to spread, generating dynamic data parameter sequences such as temperature, smoke concentration, and CO concentration for corresponding virtual sensor nodes. This provides a standardized change pattern of multi-dimensional sensor data under fire scenarios for the hazard manifestation library, supporting the system's accurate identification of early fire signs.

[0034] Step S2042: For the collapse hazard source, use the finite element analysis (FEA) structural mechanics simulation tool to simulate stress concentration, crack propagation and local / overall deformation under the conditions of surrounding rock stress, support structure strength and geological disturbance, and obtain the vibration, strain and displacement dynamic data parameter sequence of the corresponding virtual sensor node.

[0035] Step S2042 utilizes finite element analysis (FEA) structural mechanics simulation tools to perform mechanical simulations of the evolution of collapse hazard sources under different surrounding rock stresses, support structure strengths, and geological disturbances, focusing on the gradual development of stress concentration, crack propagation, and local or overall deformation. Based on the principles of structural mechanics and materials mechanics, this method discretizes the tunnel structure into a finite element mesh, simulates the deformation and failure process under load, and generates dynamic data parameter sequences such as vibration, strain, and displacement for corresponding virtual sensor nodes. This constructs a multi-stage mechanical performance characteristic database of collapse events, enhancing the system's ability to predict structural hazards.

[0036] Step S2043: For flood hazard sources, use shallow water equations or multiphase fluid dynamics simulation tools to simulate water intrusion, water depth and velocity distribution under rainfall intensity, water source location and drainage system conditions, and obtain the dynamic data parameter sequence of water level, humidity and flow rate of the corresponding virtual sensor nodes.

[0037] The core principle of step S2043 is to use shallow water equations or multiphase fluid dynamics simulation tools to perform fluid dynamics simulations of the intrusion process of flood hazard sources under different rainfall intensities, water source locations, and drainage system conditions. The simulation focuses on the dynamic changes in water intrusion, water depth, and flow velocity distribution. This method utilizes hydrodynamic principles, considering the effects of gravity, friction, and topography, to simulate the entire process of flooding from its inlet to its spread. It generates dynamic data parameter sequences such as water level, humidity, and flow rate for corresponding virtual sensor nodes, thereby providing the hazard performance database with hydrological characteristic change patterns under flood scenarios and supporting the system's timely early warning of flood risks.

[0038] Step S2044: For the hazard source of toxic gas leakage, the gas diffusion model is used in combination with fluid dynamics simulation tools to simulate the diffusion path and concentration distribution under the leakage point, gas type, concentration and ventilation conditions, so as to obtain the dynamic data parameter sequence of toxic gas concentration for the corresponding virtual sensor node.

[0039] Step S2044 utilizes a gas diffusion model combined with fluid dynamics simulation tools to simulate the diffusion process of toxic gas leak hazards under different leak points, gas types, initial concentrations, and ventilation conditions, tracking the spatiotemporal evolution of diffusion paths and concentration distributions. Based on convection-diffusion equations and turbulence models, this method simulates the transport and dilution process of gas within the confined space of a tunnel, generating dynamic data parameter sequences of toxic gas concentrations for corresponding virtual sensor nodes. This provides a foundation for constructing a hazard manifestation database of concentration gradients and propagation characteristics of gas leak events, offering a basis for the system's refined identification of chemical risks.

[0040] Step S2045: For vehicle collision hazards, use microscopic traffic flow simulation tools to simulate the evolution of traffic congestion and secondary accidents under collision location, vehicle type, speed and load conditions, and obtain the vibration, pressure and traffic flow dynamic data parameter sequences of the corresponding virtual sensor nodes.

[0041] The core principle of step S2045 is to use microscopic traffic flow simulation tools to perform traffic dynamics simulations on the evolution of vehicle collision hazards under different collision locations, vehicle types, speeds, and load conditions, focusing on the formation of traffic congestion and the chain reaction of secondary accidents. This method, based on car-following models and collision dynamics principles, simulates the interaction between individual vehicle behavior and group traffic flow, generating dynamic data parameter sequences such as vibration, pressure, and traffic flow for corresponding virtual sensor nodes. This provides traffic disturbance patterns under collision events to the hazard manifestation database, supporting early intervention and warning of accident chain reactions by the system.

[0042] In some embodiments disclosed in this invention, the method for analyzing the event location performance characteristics and sensor data parameter performance of a virtual tunnel structure model includes: step S401, setting several location points within the virtual tunnel structure model, and analyzing the dynamic data parameter performance of each virtual sensor node; if the dynamic data parameter performance of a virtual sensor node falls within a first range of interest, then marking the virtual sensor node as a point of interest, and associating the marked virtual sensor node with location points within a preset distance range, and identifying the associated location points as points of interest.

[0043] Step S402: Associate the location points of interest that are close to each other to form a cluster of location points of interest, and record the changes in the cluster of location points of interest as the event location performance feature.

[0044] In some embodiments disclosed in this invention, the method for finding matching dangerous behaviors in the dangerous behavior database includes: step S403, constructing an indexing mechanism for dangerous behaviors in the dangerous behavior database, performing feature label analysis on the cluster of location points of interest and dynamic data parameter performance of real-time events, and using the indexing mechanism to compare and analyze the feature labels to determine several dangerous behaviors.

[0045] The indexing mechanism includes determining the cluster center point of the cluster of location points of concern in each hazard manifestation, judging several edge distances of the cluster edge relative to the center point, calculating the edge distance variance, identifying the cluster center point, edge distance variance, and the sensor type corresponding to the virtual sensor node as a distinguishing label group, and classifying the hazard manifestation according to the distinguishing label group.

[0046] In some embodiments disclosed in this invention, the method for finding matching dangerous behaviors in the dangerous behavior database includes: step S404, after determining several dangerous behaviors using an indexing mechanism, comparing the cluster of attention location points between each dangerous behavior and a real-time event to determine the equivalent parameters of the event location, and comparing the dynamic data parameter performance between each dangerous behavior and a real-time event to determine the equivalent parameters of the data parameters.

[0047] Step S405: Based on the equivalent parameters of event location and data parameters, determine the proximity between the real-time event and the dangerous manifestation. If the proximity is greater than or equal to a preset value, select the corresponding dangerous manifestation for interpretation and warning.

[0048] In some embodiments disclosed in this invention, the method for determining the degree of proximity includes: step S4051, performing frame-by-frame overlap comparison of the cluster of location points of interest between the dangerous manifestation and the real-time event, determining the spatial region that is jointly mapped by the two, and determining the center point of the spatial region, radiating several detection lines outward from the center point of the region, determining the edge intersection point of each detection line and the edge of different clusters of location points of interest, calculating the edge distance between the edge intersection points, calculating the detection line reference distance from the center point of the region to the first intersection edge, and calculating the distance ratio between the edge distance and the detection line reference distance, determining the single-frame event position equivalent parameter based on the overall performance of the distance ratio corresponding to each detection line, and determining the event position equivalent parameter between the dangerous manifestation and the real-time event based on the single-frame event position equivalent parameter of consecutive frames.

[0049] Step S4052: Align and compare each relative dynamic data parameter sequence of each frame between the hazard manifestation and the real-time event to obtain a dynamic data parameter difference sequence. Analyze the data parameter differences in the dynamic data parameter difference sequence that are greater than or equal to a preset value and record them as valid data parameter differences. Based on the proportion of valid data parameter differences relative to the parameter differences in the dynamic data parameter difference sequence, determine the equivalent parameters of single-frame data parameters. Based on the equivalent parameters of single-frame data parameters of consecutive frames, determine the equivalent parameters of data parameters between the hazard manifestation and the real-time event.

[0050] Step S4052: Determine the degree of proximity based on the equivalent parameters of event location and data parameters.

[0051] In some embodiments disclosed in this invention, the expression for calculating the degree of proximity is: Where J represents the degree of proximity. The first degree of proximity conversion coefficient, The second degree of proximity conversion coefficient, This is the distance ratio analysis function corresponding to the x-th detection line in the i-th frame. Based on the preset interval to which the distance ratio belongs, it outputs the basic value of the corresponding event position equivalent parameter, where X is the total number of detection lines and n is the number of frames participating in the comparison. This is a function for analyzing the percentage difference of the u-th parameter in the i-th frame. Based on the preset interval to which the percentage difference of the parameter belongs, it outputs the basic value of the corresponding data parameter equivalent parameter. Let be the weight coefficient of the virtual sensor node corresponding to the u-th parameter difference, where U is the total number of virtual sensors participating in the comparison.

[0052] In some embodiments disclosed in this invention, an Internet of Things-based tunnel safety monitoring and control system is also disclosed, comprising: a first module for constructing a virtual tunnel structure model based on the tunnel's design structure.

[0053] The second module is used to deploy several types of IoT sensor nodes in a distributed manner within the tunnel, and to set up corresponding virtual sensor nodes in the corresponding virtual tunnel structure model to collect multi-dimensional environmental data parameters, structural health data parameters, and traffic status data parameters, and associate them with the corresponding virtual sensor nodes.

[0054] The third module is used to construct several dangerous emergencies, and based on the dangerous emergencies, to determine the IoT sensor nodes associated with the dangerous emergencies, and based on the event details of the dangerous emergencies, to determine the dynamic data parameters of the corresponding IoT sensor nodes, to obtain the dangerous manifestations of the virtual tunnel structure model, and to construct a dangerous manifestation library based on all the dangerous manifestations of the virtual tunnel structure model.

[0055] The fourth module is used to analyze the event location characteristics and sensor dynamic data parameter changes in the real-time performance of the virtual tunnel structure model. It identifies matching hazardous behaviors in the hazardous behavior library, analyzes the similarity between the event location characteristics and sensor dynamic data parameter changes and the hazardous behaviors, and issues warnings based on the similarity.

[0056] This invention discloses a tunnel safety monitoring and control method and system based on the Internet of Things (IoT), belonging to the field of tunnel safety management technology. It involves the distributed deployment of multiple IoT sensor nodes within the tunnel, with corresponding virtual sensor nodes set in a virtual model. This allows for the real-time collection and correlation of multi-dimensional environmental data parameters, structural health data parameters, and traffic status data parameters. Multiple hazardous emergencies are pre-constructed, and the associated sensor nodes and dynamic data parameters under the evolution of these events are determined. During real-time monitoring, the characteristics of the event location in the virtual model and the changes in the dynamic data parameters of the sensors are analyzed to quantify the degree of proximity between real-time performance and hazardous events, and early warnings are issued accordingly. This invention utilizes digital twin technology to integrate historical accidents and multiphysics simulations to construct a hazardous performance database, enabling early and accurate identification and warning of various emergencies such as tunnel fires, collapses, floods, toxic gas leaks, and vehicle collisions, effectively improving the accuracy and timeliness of monitoring.

[0057] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented in hardware or by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0058] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. 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 still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A tunnel safety monitoring and control method based on the Internet of Things, characterized in that, include: Step S100: Construct a virtual tunnel structure model based on the tunnel's design structure; Step S200: Distribute several types of IoT sensor nodes within the tunnel and set corresponding virtual sensor nodes within the corresponding virtual tunnel structure model. Collect multi-dimensional environmental data parameters, structural health data parameters, and traffic status data parameters, and associate them with the corresponding virtual sensor nodes. Step S300: Construct several dangerous emergencies and, based on these emergencies, determine the IoT sensor nodes associated with them. Based on the event details of the emergencies, determine the dynamic data parameters of the corresponding IoT sensor nodes to obtain the dangerous manifestations of the virtual tunnel structure model. Based on all the dangerous manifestations of the virtual tunnel structure model, construct a dangerous manifestation library. Step S400: Analyze the event location characteristics and sensor dynamic data parameter changes under the real-time performance of the virtual tunnel structure model. Find matching dangerous manifestations in the dangerous manifestation library and analyze the proximity of the event location characteristics and sensor dynamic data parameter changes to the dangerous manifestations. Based on the proximity, issue an early warning.

2. The tunnel safety monitoring and control method based on the Internet of Things according to claim 1, characterized in that, The method for constructing several dangerous emergencies includes: step S201, collecting historical accident data parameters and historical monitoring data parameters of the tunnel, wherein the historical accident data parameters include accident type, location of occurrence and consequences of accident, and the historical monitoring data parameters include multi-dimensional environmental data parameters, structural health data parameters and traffic status data parameters for a period of time before and after the accident; step S202, configuring the virtual tunnel structure model based on the historical accident data parameters and historical monitoring data parameters to obtain the dangerous performance of the virtual tunnel structure model.

3. The tunnel safety monitoring and air defense method based on the Internet of Things according to claim 1, characterized in that, The method for constructing several hazardous emergencies includes: Step S203, constructing a virtual simulation model for a virtual tunnel structure model, including setting multiple predefined hazard sources and triggering conditions in the virtual tunnel structure model, wherein the hazard sources include fire, collapse, flood, toxic gas leakage and vehicle collision; Step S204, using fluid dynamics simulation, structural mechanics simulation and traffic flow simulation tools, simulating the evolution process of each hazard source under different locations, intensities and propagation conditions, to obtain the dynamic data parameter sequence of virtual sensor nodes and the positional changes of the virtual tunnel structure model during the evolution of the hazardous event; Step 205, identifying the definition type, location, evolution result, dynamic data parameter sequence of different virtual sensor nodes and positional changes of the virtual tunnel structure model as hazardous manifestations.

4. The tunnel safety monitoring method based on the Internet of Things according to claim 3, characterized in that, The method for simulating the evolution of various hazard sources under different locations, intensities, and propagation conditions includes: Step S2041, for fire hazard sources, using computational fluid dynamics (CFD) simulation tools combined with a heat transfer model, simulating smoke diffusion, temperature field distribution, and visibility changes under fire source location, combustion intensity, oxygen concentration, and ventilation conditions, obtaining dynamic data parameter sequences of temperature, smoke concentration, and CO concentration for the corresponding virtual sensor nodes; Step S2042, for collapse hazard sources, using finite element analysis (FEA) structural mechanics simulation tools, simulating stress concentration, crack propagation, and local / overall deformation under surrounding rock stress, support structure strength, and geological disturbance conditions, obtaining dynamic data parameter sequences of vibration, strain, and displacement for the corresponding virtual sensor nodes; Step S2043, for flood hazard sources, ... Using shallow water equations or multiphase fluid dynamics simulation tools, the intrusion, water depth, and velocity distribution under rainfall intensity, water source location, and drainage system conditions are simulated to obtain the dynamic data parameter sequence of water level, humidity, and flow rate for the corresponding virtual sensor nodes; in step S2044, for toxic gas leakage hazards, the diffusion path and concentration distribution under leakage point, gas type, concentration, and ventilation conditions are simulated using a gas diffusion model combined with fluid dynamics simulation tools to obtain the dynamic data parameter sequence of toxic gas concentration for the corresponding virtual sensor nodes; in step S2045, for vehicle collision hazards, the evolution of traffic congestion and secondary accidents under collision location, vehicle type, speed, and load conditions is simulated using microscopic traffic flow simulation tools to obtain the dynamic data parameter sequence of vibration, pressure, and traffic flow for the corresponding virtual sensor nodes.

5. The tunnel safety monitoring and control method based on the Internet of Things according to claim 1, characterized in that, The method for analyzing the event location performance characteristics and sensor data parameter performance of a virtual tunnel structure model includes: Step S401, setting several location points within the virtual tunnel structure model, and analyzing the dynamic data parameter performance of each virtual sensor node. If the dynamic data parameter performance of a virtual sensor node falls within the first range of interest, the virtual sensor node is marked for interest, and the marked virtual sensor node is associated with the location points within a preset distance range, and the associated location points are identified as the location points of interest; Step S402, the location points of interest that are close to each other are associated to form a cluster of location points of interest, and the changes in the cluster of location points of interest are recorded as event location performance characteristics.

6. The tunnel safety monitoring and control method based on the Internet of Things according to claim 5, characterized in that, The method for finding matching hazardous manifestations in the hazardous manifestation database includes: Step S403, constructing an indexing mechanism for hazardous manifestations in the hazardous manifestation database, performing feature label analysis on the clusters of location points of interest in real-time events and the dynamic data parameter manifestations, and using the indexing mechanism to compare and analyze the feature labels to identify several hazardous manifestations; wherein, the indexing mechanism includes determining the cluster center point of the cluster of location points of interest in each hazardous manifestation, judging several edge distances of the cluster edge relative to the center point, calculating the edge distance variance, identifying the cluster center point, edge distance variance, and the sensor type corresponding to the virtual sensor node as a distinguishing label group, and classifying the hazardous manifestations according to the distinguishing label group.

7. The tunnel safety monitoring and control method based on the Internet of Things according to claim 5, characterized in that, The method for finding matching hazardous manifestations in the hazardous manifestation database includes: Step S404, using an indexing mechanism to identify several hazardous manifestations, comparing the cluster of attention location points between each hazardous manifestation and the real-time event to determine the equivalent parameters of the event location, and comparing the dynamic data parameter manifestations between each hazardous manifestation and the real-time event to determine the equivalent parameters of the data parameters; Step S405, based on the equivalent parameters of the event location and the equivalent parameters of the data parameters, determining the closeness between the real-time event and the hazardous manifestation, if the closeness is greater than or equal to a preset value, then selecting the corresponding hazardous manifestation for interpretation and warning.

8. The tunnel safety monitoring and control method based on the Internet of Things according to claim 7, characterized in that, The method for determining the degree of proximity includes: Step S4051, performing frame-by-frame overlap comparison of the cluster of location points of interest between the hazard manifestation and the real-time event, determining the spatial region jointly mapped by the two, and determining the center point of the spatial region. Several detection lines are radiated outward from the center point, and the edge intersection points of each detection line and the edges of different clusters of location points of interest are determined. The distance between edge intersection points is calculated, and the reference distance of the detection line from the center point of the region to the first intersecting edge is calculated. The ratio of the distance between edges to the reference distance of the detection line is calculated. Based on the overall performance of the distance ratio corresponding to each detection line, the equivalent parameter of the single-frame event position is determined, and the equivalent parameter of the single-frame event position is determined based on the single-frame event position of consecutive frames. Step S4052: Align and compare each relative dynamic data parameter sequence of each frame between the hazard manifestation and the real-time event to obtain a dynamic data parameter difference sequence. Analyze the data parameter differences in the dynamic data parameter difference sequence that are greater than or equal to a preset value, and record them as valid data parameter differences. Based on the proportion of valid data parameter differences relative to the parameter differences in the dynamic data parameter difference sequence, determine the single-frame data parameter equivalent parameters. Based on the single-frame data parameter equivalent parameters of consecutive frames, determine the data parameter equivalent parameters between the hazard manifestation and the real-time event. Step S4052: Determine the degree of proximity based on the event location equivalent parameters and the data parameter equivalent parameters.

9. A tunnel safety monitoring and control method based on the Internet of Things according to claim 8, characterized in that, The expression for calculating the degree of proximity is: Where J represents the degree of proximity. The first degree of proximity conversion coefficient, The second degree of proximity conversion coefficient, This is the distance ratio analysis function corresponding to the x-th detection line in the i-th frame. Based on the preset interval to which the distance ratio belongs, it outputs the basic value of the corresponding event position equivalent parameter, where X is the total number of detection lines and n is the number of frames participating in the comparison. This is a function for analyzing the percentage difference of the u-th parameter in the i-th frame. Based on the preset interval to which the percentage difference of the parameter belongs, it outputs the basic value of the corresponding data parameter equivalent parameter. Let be the weight coefficient of the virtual sensor node corresponding to the u-th parameter difference, where U is the total number of virtual sensors participating in the comparison.

10. A tunnel safety monitoring and control system based on the Internet of Things, characterized in that, The tunnel safety monitoring and control method for executing any one of claims 1-9 comprises: a first module for constructing a virtual tunnel structure model based on the tunnel's design structure; a second module for distributively deploying several types of IoT sensor nodes within the tunnel, setting corresponding virtual sensor nodes within the corresponding virtual tunnel structure model, collecting multi-dimensional environmental data parameters, structural health data parameters, and traffic status data parameters, and associating them with the corresponding virtual sensor nodes; a third module for constructing several dangerous emergencies, determining the IoT sensor nodes associated with the dangerous emergencies based on the dangerous emergencies, determining the dynamic data parameters of the corresponding IoT sensor nodes based on the event details of the dangerous emergencies, obtaining the dangerous manifestations of the virtual tunnel structure model, and constructing a dangerous manifestation library based on all dangerous manifestations of the virtual tunnel structure model; and a fourth module for analyzing the event location characteristics and sensor dynamic data parameter changes under the real-time performance of the virtual tunnel structure model, finding matching dangerous manifestations in the dangerous manifestation library, analyzing the proximity of event location characteristics and sensor dynamic data parameter changes to dangerous manifestations, and issuing early warnings based on the proximity manifestations.

Images