Tunnel fire smoke intelligent control method and emergency disposal system

Abstract

The invention relates to the technical field of tunnel fire smoke accident data processing, and discloses a tunnel fire smoke intelligent control method and an emergency disposal system, and the system comprises a tunnel fire smoke accident identification module, a tunnel fire smoke accident rescue information analysis module, and a tunnel fire smoke accident rescue information feedback module. The tunnel fire smoke accident occurrence type is accurately analyzed by scientifically presetting tunnel fire smoke accident occurrence type image parameters in combination with a deep learning intelligent algorithm and standard tunnel field real-time image data, and the occurrence reason type of the tunnel fire smoke accident is accurately evaluated and generated based on the deep learning intelligent algorithm. And according to the tunnel fire smoke accident occurrence type analysis parameters, the tunnel fire smoke accident rescue scheme is scientifically judged in combination with the intelligent search algorithm and the tunnel fire smoke accident rescue scheme parameters, and personalized customization of the tunnel fire smoke accident specific rescue scheme is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel fire smoke accident data processing, and specifically to an intelligent control method and emergency disposal system for tunnel fire smoke. Background Art

[0002] A tunnel fire refers to a fire incident that occurs in a tunnel. Tunnel fires have the following characteristics: 1. The disaster scope of tunnel fires is large: The space inside the tunnel is enclosed, and the fire is likely to spread, potentially causing a large number of casualties and property losses. 2. The smoke hazard of tunnel fires is significant: The smoke in the tunnel is dense, which can quickly cause asphyxiation and poisoning of people, increasing the difficulty of escape. 3. The heat radiation of tunnel fires is large: The fire releases a large amount of heat energy, which may cause damage to the tunnel structure and burns to people. 4. The fire in the tunnel is difficult to control. At the same time, a large amount of smoke is generated during the occurrence of a tunnel fire. How to scientifically and effectively monitor and process tunnel fire smoke has become an important issue to ensure tunnel safety; the existing tunnel fire smoke monitoring can only simply identify through smoke detectors, and cannot accurately identify the types of tunnel fire smoke accidents and corresponding rescue plans, nor can it accurately notify the tunnel fire smoke accident rescue department according to the types of fire smoke accidents, reducing the efficiency and quality of tunnel fire smoke accident handling.

[0003] The Chinese patent application with the publication number CN117648610A discloses an intelligent smoke removal device with AI smoke recognition function, which is provided with a smoke removal action control unit, a smoke intelligent detection unit, a negative pressure smoke removal module and a foot pedal module; an AI smoke recognition sub-unit is used to receive the data detected by the smoke detector and the air quality sensor, and through machine learning algorithms for identification and classification to determine whether there is smoke and particulate matter. To achieve accurate identification of smoke; however, the above technical solutions cannot scientifically analyze the causes of smoke occurrence. Summary of the Invention

[0004] (I) Technical Problems to be Solved To solve the problem that the existing tunnel fire smoke monitoring can only simply identify through smoke detectors, cannot accurately identify the types of tunnel fire smoke accidents, nor can it accurately notify the tunnel fire smoke accident rescue department according to the types of fire smoke accidents, reducing the efficiency and quality of tunnel fire smoke accident handling, and achieve the purpose of accurately identifying tunnel fire smoke accidents, scientifically analyzing the types of tunnel fire smoke accidents, accurately analyzing the rescue plans for tunnel fire smoke accidents, accurately searching for tunnel fire smoke accident rescue departments, accurately positioning the occurrence location of tunnel fire smoke accidents, and accurately positioning and efficiently feedbacking tunnel fire smoke accident rescue information.

[0005] (II) Technical Solutions The present invention is realized through the following technical solutions: An intelligent control method for tunnel fire smoke, the method comprising the following steps: S1. Collect real-time image data of the tunnel site; S2. Preprocess the tunnel site image data collected according to the real-time image data of the tunnel site, and generate standard real-time image data of the tunnel site; S3. Perform accident occurrence identification processing on the tunnel fire smoke according to the standard real-time image data of the tunnel site and the tunnel fire smoke accident image data, generate tunnel fire smoke accident identification data, and when no accident occurs, directly end the current tunnel fire smoke monitoring operation; S4. When an accident occurs, perform accident occurrence type analysis processing on the tunnel fire smoke based on the standard real-time image data of the tunnel site and the tunnel fire smoke accident occurrence type image data, and generate tunnel fire smoke accident occurrence type analysis data; S5. Perform specific rescue plan analysis processing on the tunnel fire smoke accident according to the tunnel fire smoke accident occurrence type analysis data and the tunnel fire smoke accident rescue plan data, and generate tunnel fire smoke accident rescue plan analysis data; S6. Perform specific rescue execution department identity characteristic information analysis processing on the tunnel fire smoke accident according to the tunnel fire smoke accident rescue plan analysis data and the tunnel fire smoke accident rescue department characteristic data, and generate tunnel fire smoke accident rescue department characteristic identification data; S7. Collect the spatial position coordinate data of the tunnel fire smoke accident occurrence and construct tunnel fire smoke accident rescue data, and perform tunnel fire smoke accident rescue information feedback operation according to the tunnel fire smoke accident rescue data.

[0006] Preferably, the operation steps for collecting the real-time image data of the tunnel site are as follows: S11. Online collect real-time on-site images inside the tunnel through the monitoring cameras installed in the tunnel, and generate a set of real-time image data of the tunnel site .

[0007] Preferably, the operation steps for preprocessing the tunnel site image data collected according to the real-time image data of the tunnel site and generating standard real-time image data of the tunnel site are as follows: S21. Use the adaptive filtering method to perform image data noise reduction preprocessing on the tunnel site real-time image data in the set of tunnel site real-time image data and generate a set of standard real-time image data of the tunnel site , wherein represents the th standard real-time image data of the tunnel site, Represents the maximum value of the number of real-time images of the standard tunnel site.

[0008] Preferably, based on the standard tunnel site real-time image data and the tunnel fire smoke accident image data, perform accident occurrence identification processing on the tunnel fire smoke, generate tunnel fire smoke accident identification data. When no accident has occurred, the operation steps to directly end the current tunnel fire smoke monitoring operation are as follows: S31. Establish a set of tunnel fire smoke accident image data , ; where represents the th tunnel fire smoke accident image data, represents the maximum value of the number of tunnel fire smoke accident images, and the tunnel fire smoke accident image data represents the on-site image data of a fire smoke accident occurring inside the tunnel; S32. Use the iterative deepening search algorithm to match the standard tunnel site real-time image data in the set of standard tunnel site real-time image data with the tunnel fire smoke accident image data in the set of tunnel fire smoke accident image data to perform image feature matching, and generate tunnel fire smoke accident identification data based on the image feature matching result ; When ; When and successfully perform image feature matching, indicating that a fire smoke accident has occurred inside the tunnel, then output the tunnel fire smoke accident identification data as an accident occurred; When and both fail to perform image feature matching, indicating that no fire smoke accident has occurred inside the tunnel, then output the tunnel fire smoke accident identification data as no accident occurred, and at this time, directly end the current tunnel fire smoke monitoring operation.

[0009] Preferably, when an accident has occurred, based on the standard tunnel site real-time image data and the tunnel fire smoke accident occurrence type image data, perform accident occurrence type analysis processing on the tunnel fire smoke, and the operation steps to generate tunnel fire smoke accident occurrence type analysis data are as follows: S41. When the tunnel fire smoke accident identification data is an accident occurred, establish a set of tunnel fire smoke accident occurrence type image data , ; where represents the Image data of the occurrence types of tunnel fire smoke accidents Represents the maximum value of the number of occurrence types of tunnel fire smoke accidents. The occurrence types of tunnel fire smoke accidents include tunnel fire smoke accidents caused by electrical equipment failures, vehicle failures, construction operations, natural factors, and human factors. Among them, tunnel fire smoke accidents caused by electrical equipment failures include tunnel fire smoke accidents caused by failures of lighting equipment, ventilation equipment, drainage equipment, and traffic signal equipment in the tunnel; tunnel fire smoke accidents caused by vehicle failures include tunnel fire smoke accidents caused by failures of household vehicles, logistics vehicles, and passenger vehicles in the tunnel; tunnel fire smoke accidents caused by construction operations include tunnel fire smoke accidents caused by welding construction operations, cutting construction operations, and drilling construction operations during tunnel construction and maintenance; tunnel fire smoke accidents caused by natural factors include tunnel fire smoke accidents caused by lightning strikes and climate changes that cause combustibles in the tunnel to burn; tunnel fire smoke accidents caused by human factors include tunnel fire smoke accidents caused by humans discarding lit cigarette butts, setting off fireworks, and burning combustibles in the tunnel. S42. Combine the standard tunnel on-site real-time image data set The standard tunnel on-site real-time image data in it With the image data set of the occurrence types of tunnel fire smoke accidents The image data of the occurrence types of tunnel fire smoke accidents in it Perform image feature matching to search for the image data of the occurrence types of tunnel fire smoke accidents that match the standard tunnel on-site real-time image data And construct the analysis data of the occurrence types of tunnel fire smoke accidents The specific operation steps for generating the analysis data of the occurrence types of tunnel fire smoke accidents are as follows S421. Initialize the algorithm parameters, update the maximum number of iterations T, and the population size N of the accident type search sooty terns S422. Calculate the fitness value of the accident type search sooty terns, that is, calculate the fitness value of the standard tunnel on-site real-time image data In the search space of the image data set of the occurrence types of tunnel fire smoke accidents And the image data of the occurrence types of tunnel fire smoke accidents S423. The accident type search sooty terns perform migration behavior, that is, search in the search space of the image data set of the occurrence types of tunnel fire smoke accidents for the standard tunnel on-site real-time image data The image data of the tunnel fire smoke accident occurrence type that matches , the migration behaviors include conflict avoidance, aggregation, and update; S4231. Conflict avoidance: Update the position in the search space of the tunnel fire smoke accident occurrence type image data set where the accident type search tern does not collide with other accident type search terns. The position update formula is as follows: , where represents the new position of the accident type search tern in the search space of the tunnel fire smoke accident occurrence type image data set after the th iteration, represents the current position of the accident type search tern in the search space of the tunnel fire smoke accident occurrence type image data set before the th iteration, represents the variable coefficient for collision avoidance. The calculation formula of the variable coefficient is as follows: where is the control variable used to adjust ; ; ; S4232. Aggregation: Aggregation means that the accident type search tern approaches the optimal position among adjacent accident type search terns in the search space of the tunnel fire smoke accident occurrence type image data set on the premise of avoiding conflicts, that is, the accident type search tern searches in the search space of the tunnel fire smoke accident occurrence type image data set for the position that best matches the tunnel fire smoke accident occurrence type image data with the standard tunnel site real-time image data . The optimal position calculation formula is as follows: , where represents the moving variable of the different current positions of the accident type search tern individual in the search space of the tunnel fire smoke accident occurrence type image data set that matches the tunnel fire smoke accident occurrence type image data with the standard tunnel site real-time image data to the optimal solution position , ; represents the position of the accident type search tern individual in the search space of the tunnel fire smoke accident occurrence type image data set that best matches the tunnel fire smoke accident occurrence type image data with the standard tunnel site real-time image data . ; ; represents a random variable; S4233. Update: Update refers to updating the position update trajectory of the optimal solution of the type of tunnel fire smoke accident occurrence image data in the search space that is oriented to match the real-time image data of the standard tunnel site of the type of tunnel fire smoke accident occurrence image data , that is, its update trajectory calculation formula is as follows: , , represents in the set of type of tunnel fire smoke accident occurrence image data searching in the search space for the type of tunnel fire smoke accident occurrence image data that matches the real-time image data of the standard tunnel site ; trajectory; S424. The accident type search for Sooty Terns to perform an attack behavior. The attack behavior means that during the migration process, the accident type search for Sooty Terns adjusts the flight altitude or its own speed and attack angle in the search space of the set of type of tunnel fire smoke accident occurrence image data to attack prey with an aerial hovering behavior, that is, searching in the search space of the set of type of tunnel fire smoke accident occurrence image data for the type of tunnel fire smoke accident occurrence image data that matches the real-time image data of the standard tunnel site ; the calculation formula for the aerial hovering behavior is as follows: , , , , , where represents the abscissa of the position of the accident type search for Sooty Terns in the search space of the set of type of tunnel fire smoke accident occurrence image data ; represents the ordinate of the position of the accident type search for Sooty Terns in the search space of the set of type of tunnel fire smoke accident occurrence image data ; represents the vertical coordinate of the position of the accident type search for Sooty Terns in the search space of the set of type of tunnel fire smoke accident occurrence image data ; represents the radius of each helix, represents the angular variable between [0, 2π], and respectively represent the sine value and cosine value of the angular variable ; S425. The accident type search for Sooty Terns updates its position, that is, the accident type search for Sooty Terns in the set of type of tunnel fire smoke accident occurrence image data The position in the search space is updated, and the position update formula is as follows: , where represents the accident type. Search for the updated position of the sooty tern in the image data set of the accident type of tunnel fire flue gas accidents at the th iteration in the search space; S426. Calculate the fitness value, that is, calculate the fitness value of the standard tunnel site real-time image data in the search space of the image data set of the accident type of tunnel fire flue gas accidents and the image data of the accident type of tunnel fire flue gas accidents , and search for the fitness value of the image data of the accident type of tunnel fire flue gas accidents in the global search space that is most matched with the standard tunnel site real-time image data ; S427. When the maximum number of iterations is reached, output the accident type of tunnel fire flue gas accidents that is most matched with the standard tunnel site real-time image data , and construct the accident type analysis data of tunnel fire flue gas accidents . Preferably, according to the accident type analysis data of tunnel fire flue gas accidents and the accident rescue plan data of tunnel fire flue gas accidents, the specific operation steps for analyzing and processing the accident rescue plan of tunnel fire flue gas accidents to generate the accident rescue plan analysis data of tunnel fire flue gas accidents are as follows:

[0010] S51. Establish an accident rescue plan data set of tunnel fire flue gas accidents , where represents the accident rescue plan data of tunnel fire flue gas accidents corresponding to the th accident type of tunnel fire flue gas accidents. The accident rescue plan data of tunnel fire flue gas accidents represents the standard specific rescue method information of tunnel fire for different types of tunnel fire flue gas accidents. The standard specific rescue method information of tunnel fire includes rescue equipment, rescue personnel composition, and rescue execution steps; S52. Use the breadth-first search algorithm to match the accident type keywords of the accident type analysis data of tunnel fire flue gas accidents with the accident rescue plan data in the accident rescue plan data set of tunnel fire flue gas accidents to analyze the accident type analysis data of tunnel fire flue gas accidents​​​​ The corresponding tunnel fire smoke accident rescue plan data , and generate tunnel fire smoke accident rescue plan analysis data through data identification .

[0011] Preferably, the operation steps for analyzing and processing the identity characteristic information of the specific rescue execution department of the tunnel fire smoke accident based on the tunnel fire smoke accident rescue plan analysis data and the tunnel fire smoke accident rescue department characteristic data are as follows: S61. Establish a set of tunnel fire smoke accident rescue department characteristic data , where represents the tunnel fire smoke accident rescue department characteristic data corresponding to the execution of the tunnel fire smoke accident rescue plan data . The tunnel fire smoke accident rescue department characteristic data represents the identity characteristic information of the tunnel fire smoke accident rescue department that executes different types of tunnel fire smoke accident rescue plans. The identity characteristic information of the tunnel fire smoke accident rescue department includes the name, address, and communication contact information of the tunnel fire smoke accident rescue department; S62. Use a two-way search algorithm to match the characters of the tunnel fire smoke accident rescue plan between the tunnel fire smoke accident rescue plan analysis data and the tunnel fire smoke accident rescue department characteristic data in the tunnel fire smoke accident rescue department characteristic data set , search for the tunnel fire smoke accident rescue department characteristic data corresponding to the tunnel fire smoke accident rescue plan analysis data , and generate tunnel fire smoke accident rescue department characteristic identification data through data identification . .

[0012] Preferably, the operation steps for collecting the spatial position coordinate data of the tunnel fire smoke accident occurrence location and constructing the tunnel fire smoke accident rescue data, and performing the tunnel fire smoke accident rescue information feedback operation based on the tunnel fire smoke accident rescue data are as follows: S71. Collect the spatial position coordinates of the monitoring camera that captures the real-time on-site image of the tunnel through a position sensor, and generate the spatial position coordinate data of the tunnel fire smoke accident occurrence location . The spatial position coordinate data of the tunnel fire smoke accident occurrence location includes the longitude, latitude, and altitude of the tunnel fire smoke accident occurrence location; S72. The standard tunnel on-site real-time image data set , the tunnel fire smoke accident occurrence type analysis data ​, the analysis data of the rescue plan for tunnel fire smoke accidents , the characteristic identification data of the rescue department for tunnel fire smoke accidents and the spatial location coordinate data of the occurrence of tunnel fire smoke accidents Perform data collection and combination processing, and construct the rescue data for tunnel fire smoke accidents ; S73. Push and feedback the rescue data for tunnel fire smoke accidents to the tunnel fire monitoring platform through the Internet of Things communication network to perform the rescue information feedback operation for tunnel fire smoke accidents.

[0013] An intelligent emergency disposal system for tunnel fire smoke, used to implement the intelligent control method for tunnel fire smoke. The system includes a tunnel fire smoke accident identification module, a tunnel fire smoke accident rescue information analysis module, and a tunnel fire smoke accident rescue information feedback module; The tunnel fire smoke accident identification module includes a tunnel site real-time image acquisition unit, a tunnel site real-time image preprocessing unit, a tunnel fire smoke accident image storage unit, and a tunnel fire smoke accident identification unit; The tunnel site real-time image acquisition unit collects tunnel site real-time image data through a monitoring camera; the tunnel site real-time image preprocessing unit preprocesses the tunnel site image data collected based on the tunnel site real-time image data and generates standard tunnel site real-time image data; the tunnel fire smoke accident image storage unit is used to store tunnel fire smoke accident image data; the tunnel fire smoke accident identification unit performs accident occurrence identification processing for tunnel fire smoke according to the standard tunnel site real-time image data and tunnel fire smoke accident image data, and generates tunnel fire smoke accident identification data; The tunnel fire smoke accident rescue information analysis module includes a tunnel fire smoke accident occurrence type image storage unit, a tunnel fire smoke accident occurrence type analysis unit, a tunnel fire smoke accident rescue plan storage unit, a tunnel fire smoke accident rescue plan formulation unit, a tunnel fire smoke accident rescue department information storage unit, and a tunnel fire smoke accident rescue department identification unit; The tunnel fire smoke accident occurrence type image storage unit is used to store the tunnel fire smoke accident occurrence type image data; the tunnel fire smoke accident occurrence type analysis unit performs analysis and processing of the tunnel fire smoke accident occurrence type based on the standard tunnel site real-time image data combined with the deep learning intelligent algorithm and the tunnel fire smoke accident occurrence type image data, and generates tunnel fire smoke accident occurrence type analysis data; the tunnel fire smoke accident rescue plan storage unit is used to store the tunnel fire smoke accident rescue plan data; the tunnel fire smoke accident rescue plan formulation unit performs analysis and processing of the specific rescue plan for the tunnel fire smoke accident according to the tunnel fire smoke accident occurrence type analysis data and the tunnel fire smoke accident rescue plan data, and generates tunnel fire smoke accident rescue plan analysis data; the tunnel fire smoke accident rescue department information storage unit is used to store the tunnel fire smoke accident rescue department characteristic data; the tunnel fire smoke accident rescue department identification unit performs analysis and processing of the specific rescue execution department identity characteristic information for the tunnel fire smoke accident according to the tunnel fire smoke accident rescue plan analysis data and the tunnel fire smoke accident rescue department characteristic data, and generates tunnel fire smoke accident rescue department characteristic identification data; The tunnel fire smoke accident rescue information feedback module includes a tunnel fire smoke accident occurrence location information collection unit, a tunnel fire smoke accident rescue information construction unit, and a tunnel fire smoke accident rescue information feedback unit; The tunnel fire smoke accident occurrence location information collection unit collects the spatial location coordinate data of the tunnel fire smoke accident through a position sensor; the tunnel fire smoke accident rescue information construction unit is used to construct the tunnel fire smoke accident rescue data; the tunnel fire smoke accident rescue information feedback unit pushes the tunnel fire smoke accident rescue data to the tunnel fire monitoring platform through the Internet of Things communication network to perform the tunnel fire smoke accident rescue information feedback operation.

[0014] (III) Beneficial effects The present invention provides a tunnel fire smoke intelligent control method and an emergency disposal system. It has the following beneficial effects: 1. Through the cooperation of the on-site real-time image acquisition unit and the preprocessing unit of the on-site real-time image of the tunnel, the on-site real-time image of the tunnel is dynamically collected online by the monitoring camera, and the data noise reduction algorithm is combined to perform noise reduction processing on the on-site real-time image of the tunnel, improving the acquisition accuracy of the on-site real-time image of the tunnel and the sensitivity of the tunnel fire smoke monitoring; the tunnel fire smoke accident image storage unit and the tunnel fire smoke accident recognition unit cooperate with each other. Based on big data science, the tunnel fire smoke accident image parameters are established, combined with the intelligent search algorithm and the standard on-site real-time image parameters of the tunnel, to perform intelligent and efficient recognition of the tunnel fire smoke accident, realizing intelligent and efficient judgment of the occurrence of the tunnel fire smoke accident and improving the reliability of the tunnel fire smoke accident monitoring.

[0015] 2. Through the cooperation of the tunnel fire smoke accident occurrence type image storage unit and the tunnel fire smoke accident occurrence type analysis unit, the tunnel fire smoke accident occurrence type image parameters are scientifically preset, combined with the deep learning intelligent algorithm and the standard on-site real-time image data of the tunnel, to perform accurate analysis of the tunnel fire smoke accident occurrence type, realizing accurate evaluation and generation of the cause type of the tunnel fire smoke accident based on the deep learning intelligent algorithm; the tunnel fire smoke accident rescue plan formulation unit, based on the tunnel fire smoke accident occurrence type analysis parameters, combines the intelligent search algorithm and the tunnel fire smoke accident rescue plan parameters to scientifically judge the tunnel fire smoke accident rescue plan, realizing personalized customization of the specific rescue plan for the tunnel fire smoke accident and improving the scientific nature of the tunnel fire smoke control and treatment; the tunnel fire smoke accident rescue department identification unit, based on the tunnel fire smoke accident rescue plan analysis parameters, combines the intelligent search algorithm and the tunnel fire smoke accident rescue department characteristic information to perform refined matching of the identity information of the tunnel fire smoke accident rescue execution department, realizing efficient and accurate search for the tunnel fire smoke accident rescue department and improving the efficiency and safety of the tunnel fire smoke control and treatment.

[0016] 3. Through the tunnel fire smoke accident occurrence location information acquisition unit, the spatial location coordinate information of the tunnel fire smoke accident is accurately collected by the position sensor, realizing accurate positioning of the place where the tunnel fire smoke accident occurs; the tunnel fire smoke accident rescue information construction unit and the tunnel fire smoke accident rescue information feedback unit cooperate with each other, accurately collect the on-site real-time image of the tunnel fire smoke accident, the accident occurrence type, the accident rescue plan, the accident rescue department characteristic information and the accident location information, combine numerical processing, scientifically construct the tunnel fire smoke accident rescue information, and timely and accurately feedback it to the tunnel fire monitoring platform through the Internet of Things communication network, realizing intuitive and accurate push of the tunnel fire smoke accident rescue information and improving the response efficiency and quality of the tunnel fire smoke control and treatment. Description of the Drawings

[0017] Figure 1Schematic diagram of modules of an intelligent emergency disposal system for tunnel fire smoke provided by the present invention; Figure 2 Flow chart of an intelligent control method for tunnel fire smoke provided by the present invention. Detailed implementation manners

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0019] Embodiments of the intelligent control method and emergency disposal system for tunnel fire smoke are as follows: Embodiment 1: Please refer to Figure 1 - Figure 2 , an intelligent control method for tunnel fire smoke, the method includes the following steps: S1. Collect real-time image data of the tunnel site; S2. Preprocess the tunnel site image data collected according to the real-time image data of the tunnel site, and generate standard real-time image data of the tunnel site; S3. Perform accident occurrence recognition processing on tunnel fire smoke according to the standard real-time image data of the tunnel site and the accident image data of tunnel fire smoke. Generate accident recognition data of tunnel fire smoke. When no accident occurs, directly end the current tunnel fire smoke monitoring operation; S4. When an accident occurs, perform analysis processing on the occurrence type of the tunnel fire smoke accident based on the standard real-time image data of the tunnel site and the image data of the occurrence type of the tunnel fire smoke accident, and generate analysis data of the occurrence type of the tunnel fire smoke accident; S5. Perform specific rescue plan analysis processing on the tunnel fire smoke accident according to the analysis data of the occurrence type of the tunnel fire smoke accident and the rescue plan data of the tunnel fire smoke accident, and generate analysis data of the rescue plan of the tunnel fire smoke accident; S6. Perform analysis processing on the identity characteristic information of the specific rescue execution department of the tunnel fire smoke accident according to the analysis data of the rescue plan of the tunnel fire smoke accident and the characteristic data of the rescue department of the tunnel fire smoke accident, and generate characteristic recognition data of the rescue department of the tunnel fire smoke accident; S7. Collect the spatial position coordinate data of the occurrence of the tunnel fire smoke accident and construct rescue data for the tunnel fire smoke accident, and perform tunnel fire smoke accident rescue information feedback operation according to the rescue data of the tunnel fire smoke accident.

[0020] Further, please refer toFigure 1 - Figure 2 The operation steps for collecting real-time image data of the tunnel site are as follows: S11. Online collect real-time on-site images inside the tunnel through the monitoring cameras installed in the tunnel, and generate a set of real-time image data of the tunnel site .

[0021] The operation steps for preprocessing the tunnel site image data collected based on the real-time image data of the tunnel site and generating standard real-time image data of the tunnel site are as follows: S21. Use the adaptive filtering method to perform image data noise reduction preprocessing on the real-time image data of the tunnel site in the set of real-time image data of the tunnel site and generate a set of standard real-time image data of the tunnel site , where represents the th standard real-time image data of the tunnel site, represents the maximum value of the number of standard real-time images of the tunnel site.

[0022] According to the standard real-time image data of the tunnel site and the image data of the tunnel fire smoke accident, perform accident occurrence identification processing on the tunnel fire smoke, generate tunnel fire smoke accident identification data, and when no accident has occurred, directly end the operation steps of this tunnel fire smoke monitoring operation are as follows: S31. Establish a set of image data of the tunnel fire smoke accident , where represents the th image data of the tunnel fire smoke accident, represents the maximum value of the number of tunnel fire smoke accident images, and the image data of the tunnel fire smoke accident represents the on-site image data of the fire smoke accident occurring inside the tunnel; S32. Use the iterative deepening search algorithm to match the standard real-time image data in the set of standard real-time image data of the tunnel site with the image data of the tunnel fire smoke accident in the set of image data of the tunnel fire smoke accident to perform image feature matching, and generate tunnel fire smoke accident identification data based on the image feature matching results ; When and are successfully matched in terms of image features, indicating that a fire smoke accident has occurred inside the tunnel, then output the tunnel fire smoke accident identification data as an accident occurred; When and are None of the image features match successfully, indicating that there is no fire smoke accident inside the tunnel. Then, the tunnel fire smoke accident identification data is output. Since no accident has occurred, this tunnel fire smoke monitoring operation is directly terminated at this time.

[0023] Through the cooperation of the tunnel on-site real-time image acquisition unit and the tunnel on-site real-time image preprocessing unit, the monitoring camera is used to dynamically collect the tunnel on-site real-time images online, and combined with the data denoising algorithm, the tunnel on-site actual situation image denoising processing is carried out to improve the acquisition accuracy of the tunnel on-site actual situation images and the sensitivity of tunnel fire smoke monitoring; the tunnel fire smoke accident image storage unit and the tunnel fire smoke accident identification unit cooperate with each other. Based on big data science, the tunnel fire smoke accident image parameters are established, and combined with the intelligent search algorithm and the standard tunnel on-site real-time image parameters, the intelligent and efficient identification of tunnel fire smoke accidents is carried out to realize the intelligent and efficient judgment of the occurrence of tunnel fire smoke accidents and improve the reliability of tunnel fire smoke accident monitoring.

[0024] Furthermore, please refer to Figure 1 – Figure 2 , when an accident occurs, based on the standard tunnel on-site real-time image data and the tunnel fire smoke accident occurrence type image data, the operation steps for analyzing the occurrence type of the tunnel fire smoke accident and generating the tunnel fire smoke accident occurrence type analysis data are as follows: S41. When the tunnel fire smoke accident identification data indicates that an accident has occurred, a set of tunnel fire smoke accident occurrence type image data is established, ; where represents the th type of tunnel fire smoke accident occurrence type image data, Represents the maximum value of the number of types of tunnel fire smoke accidents. The types of tunnel fire smoke accidents include tunnel fire smoke accidents caused by electrical equipment failures, vehicle failures, construction operations, natural factors, and human factors. Among them, tunnel fire smoke accidents caused by electrical equipment failures include tunnel fire smoke accidents caused by failures of tunnel lighting equipment, ventilation equipment, drainage equipment, and traffic signal equipment. Tunnel fire smoke accidents caused by vehicle failures include tunnel fire smoke accidents caused by failures of household vehicles, logistics vehicles, and passenger vehicles in the tunnel. Tunnel fire smoke accidents caused by construction operations include tunnel fire smoke accidents caused by welding construction operations, cutting construction operations, and drilling construction operations during tunnel construction and maintenance. Tunnel fire smoke accidents caused by natural factors include tunnel fire smoke accidents caused by lightning strikes and climate changes that cause combustibles in the tunnel to burn. Tunnel fire smoke accidents caused by human factors include tunnel fire smoke accidents caused by humans discarding lit cigarette butts, setting off fireworks, and burning combustibles in the tunnel. S42. The set of real-time image data of the standard tunnel site The real-time image data of the standard tunnel site in And the set of image data of the types of tunnel fire smoke accidents The image data of the types of tunnel fire smoke accidents in Perform image feature matching to search for the image data of the types of tunnel fire smoke accidents that match the real-time image data of the standard tunnel site And construct the analysis data of the types of tunnel fire smoke accidents The specific operation steps for generating the analysis data of the types of tunnel fire smoke accidents are as follows: Execute the generation of the analysis data of the types of tunnel fire smoke accidents are as follows: S421. Initialize the algorithm parameters, update the maximum number of iterations T, and the population size N of the accident type search sooty terns. S422. Calculate the fitness value of the accident type search sooty terns, that is, calculate the fitness value of the real-time image data of the standard tunnel site and the image data of the types of tunnel fire smoke accidents in the search space of the set of image data of the types of tunnel fire smoke accidents. In the search space of the set of image data of the types of tunnel fire smoke accidents, calculate the fitness value of the real-time image data of the standard tunnel site And the image data of the types of tunnel fire smoke accidents ; S423. The accident type search sooty terns perform migration behaviors, that is, search for the image data of the types of tunnel fire smoke accidents that match the real-time image data of the standard tunnel site in the search space of the set of image data of the types of tunnel fire smoke accidents. The migration behaviors include conflict avoidance, aggregation, and update. In the search space of the set of image data of the types of tunnel fire smoke accidents, search for the image data of the types of tunnel fire smoke accidents that match the real-time image data of the standard tunnel site And the image data of the types of tunnel fire smoke accidents that match the real-time image data of the standard tunnel site The migration behaviors include conflict avoidance, aggregation, and update. S4231. Conflict Avoidance: In the image data set of tunnel fire smoke accident occurrence types Update the position in the search space where the accident type search tern does not collide with other accident type search terns. The position update formula is as follows: , where represents the new position of the accident type search tern in the image data set of tunnel fire smoke accident occurrence types after the -th iteration, represents the current position of the accident type search tern in the image data set of tunnel fire smoke accident occurrence types before the -th iteration, represents the variable coefficient for collision avoidance. The calculation formula for the variable coefficient is as follows: , where is the control variable used to adjust ; S4232. Aggregation: Aggregation means that the accident type search tern approaches the optimal position among adjacent accident type search terns in the image data set of tunnel fire smoke accident occurrence types while avoiding conflicts. That is, the accident type search tern searches for the position in the image data set of tunnel fire smoke accident occurrence types that is most matched with the standard tunnel on-site real-time image data among the tunnel fire smoke accident occurrence type image data. The calculation formula for the optimal position is as follows: , where represents the moving variable from the different current positions of the accident type search tern individual in the image data set of tunnel fire smoke accident occurrence types that are matched with the standard tunnel on-site real-time image data to the optimal solution position ; represents the position of the accident type search tern individual in the image data set of tunnel fire smoke accident occurrence types that is most matched with the standard tunnel on-site real-time image data among the tunnel fire smoke accident occurrence type image data ; represents the random variable; S4233. Update: Update means in the image data set of tunnel fire smoke accident occurrence types The position update trajectory of the optimal solution of the image data of the tunnel fire smoke accident type that matches the orientation in the search space and the real-time image data of the standard tunnel site The image data of the tunnel fire smoke accident type that matches is as follows: That is, its update trajectory calculation formula is as follows: , represents searching in the set of image data of the tunnel fire smoke accident type in the search space for the image data of the tunnel fire smoke accident type that matches the real-time image data of the standard tunnel site ; The trajectory; S424. The accident type search for the sooty tern to perform an attack behavior. The attack behavior means that during the migration process, the accident type search for the sooty tern adjusts its flight altitude or its own speed and attack angle in the search space of the set of image data of the tunnel fire smoke accident type to attack the prey with a hovering behavior in the air. That is, searching in the search space of the set of image data of the tunnel fire smoke accident type for the image data of the tunnel fire smoke accident type that matches the real-time image data of the standard tunnel site ; The image data of the tunnel fire smoke accident type that matches The hovering behavior in the air calculation formula is as follows: , , , where represents the abscissa of the position of the accident type search for the sooty tern in the search space of the set of image data of the tunnel fire smoke accident type ; represents the ordinate of the position of the accident type search for the sooty tern in the search space of the set of image data of the tunnel fire smoke accident type ; represents the vertical coordinate of the position of the accident type search for the sooty tern in the search space of the set of image data of the tunnel fire smoke accident type ; represents the radius of each helix, represents the angle variable between [0, 2π], and respectively represent the sine value and cosine value of the angle variable ; S425. The accident type search for the sooty tern updates its position, that is, the position of the accident type search for the sooty tern in the search space of the set of image data of the tunnel fire smoke accident type is updated. The position update formula is as follows: , where represents the accident type search for the sooty tern at the After the ith iteration, update the position in the search space; S426. Calculate the fitness value, that is, calculate the fitness value of the standard tunnel site real-time image data in the search space of the tunnel fire smoke accident occurrence type image data set and the tunnel fire smoke accident occurrence type image data , and search for the global tunnel fire smoke accident occurrence type image data set in the search space, and find the fitness value of the tunnel fire smoke accident occurrence type image data that is most matched with the standard tunnel site real-time image data in the search space; S427. When the maximum number of iterations is reached, output the tunnel fire smoke accident occurrence type that is most matched with the standard tunnel site real-time image data , and construct the tunnel fire smoke accident occurrence type analysis data .

[0025] According to the tunnel fire smoke accident occurrence type analysis data and the tunnel fire smoke accident rescue plan data, the specific operation steps for analyzing and processing the tunnel fire smoke accident rescue plan are as follows: S51. Establish a tunnel fire smoke accident rescue plan data set , where represents the tunnel fire smoke accident rescue plan data corresponding to the th type of tunnel fire smoke accident occurrence type. The tunnel fire smoke accident rescue plan data represents the standard specific rescue method information for tunnel fires set for different types of tunnel fire smoke accidents. The standard specific rescue method information for tunnel fires includes rescue equipment, rescue personnel composition, and rescue execution steps; S52. Use the breadth-first search algorithm to perform keyword matching of the tunnel fire smoke accident occurrence type between the tunnel fire smoke accident occurrence type analysis data and the tunnel fire smoke accident rescue plan data in the tunnel fire smoke accident rescue plan data set . Analyze the tunnel fire smoke accident rescue plan data corresponding to the tunnel fire smoke accident occurrence type analysis data , and generate the tunnel fire smoke accident rescue plan analysis data through data identification. .

[0026] ​Analyze and process the identity characteristic information of the specific rescue execution department for tunnel fire smoke accidents based on the analysis data of the tunnel fire smoke accident rescue plan and the characteristic data of the tunnel fire smoke accident rescue department, and the operation steps for generating the tunnel fire smoke accident rescue department characteristic identification data are as follows: S61. Establish a set of tunnel fire smoke accident rescue department characteristic data , where represents the data of the tunnel fire smoke accident rescue plan implementation corresponding tunnel fire smoke accident rescue department characteristic data. The tunnel fire smoke accident rescue department characteristic data represents the identity characteristic information of the tunnel fire smoke accident rescue department implementing different types of tunnel fire smoke accident rescue plans. The identity characteristic information of the tunnel fire smoke accident rescue department includes the name, address, and communication contact information of the tunnel fire smoke accident rescue department; S62. Use a two-way search algorithm to match the tunnel fire smoke accident rescue plan analysis data with the tunnel fire smoke accident rescue department characteristic data in the tunnel fire smoke accident rescue department characteristic data set for character matching of the tunnel fire smoke accident rescue plan, search for the tunnel fire smoke accident rescue department characteristic data corresponding to the tunnel fire smoke accident rescue plan analysis data , and generate tunnel fire smoke accident rescue department characteristic identification data through data identification .

[0027] Through the cooperation of the tunnel fire smoke accident occurrence type image storage unit and the tunnel fire smoke accident occurrence type analysis unit, scientifically preset the tunnel fire smoke accident occurrence type image parameters, combine deep learning intelligent algorithms with standard tunnel on-site real-time image data for accurate analysis of the tunnel fire smoke accident occurrence type, and realize the accurate assessment and generation of the cause type of the tunnel fire smoke accident based on deep learning intelligent algorithms; the tunnel fire smoke accident rescue plan formulation unit scientifically judges the tunnel fire smoke accident rescue plan according to the tunnel fire smoke accident occurrence type analysis parameters combined with the intelligent search algorithm and the tunnel fire smoke accident rescue plan parameters, realizes the personalized customization of the specific rescue plan for the tunnel fire smoke accident, and improves the scientific nature of the tunnel fire smoke control and treatment; the tunnel fire smoke accident rescue department identification unit finely matches the identity information of the tunnel fire smoke accident rescue execution department according to the tunnel fire smoke accident rescue plan analysis parameters combined with the intelligent search algorithm and the tunnel fire smoke accident rescue department characteristic information, realizes the efficient and accurate search for the tunnel fire smoke accident rescue department, and improves the efficiency and safety of the tunnel fire smoke control and treatment.

[0028] Further, please refer toFigure 1 - Figure 2 When collecting the spatial position coordinate data of a tunnel fire smoke accident and constructing the rescue data for the tunnel fire smoke accident, the operating steps for performing the rescue information feedback operation of the tunnel fire smoke accident based on the rescue data for the tunnel fire smoke accident are as follows: S71. Collect the spatial position coordinates of the monitoring camera that captures the real-time on-site images of the tunnel through a position sensor, and generate the spatial position coordinate data of the tunnel fire smoke accident The spatial position coordinate data of the tunnel fire smoke accident includes the longitude, latitude, and altitude of the location where the tunnel fire smoke accident occurs; S72. Perform data collection and combination processing on the standard tunnel on-site real-time image data set , the analysis data of the type of tunnel fire smoke accident , the analysis data of the rescue plan for the tunnel fire smoke accident , the characteristic identification data of the rescue department for the tunnel fire smoke accident and the spatial position coordinate data of the tunnel fire smoke accident to construct the rescue data for the tunnel fire smoke accident ; S73. Push and feedback the rescue data for the tunnel fire smoke accident to the tunnel fire monitoring platform through the Internet of Things communication network to perform the rescue information feedback operation of the tunnel fire smoke accident.

[0029] Through the tunnel fire smoke accident location information collection unit, use a position sensor to accurately collect the spatial position coordinate information of the tunnel fire smoke accident, and achieve precise positioning of the location where the tunnel fire smoke accident occurs; the tunnel fire smoke accident rescue information construction unit and the tunnel fire smoke accident rescue information feedback unit cooperate with each other to accurately collect the on-site actual situation images of the tunnel fire smoke accident, the type of accident, the accident rescue plan, the characteristic information of the accident rescue department, and the accident location information, combine numerical processing, scientifically construct the rescue information for the tunnel fire smoke accident, and timely and accurately feedback it to the tunnel fire monitoring platform through the Internet of Things communication network, realizing the intuitive and precise push of the rescue information for the tunnel fire smoke accident, and improving the response efficiency and quality of the tunnel fire smoke control and treatment.

[0030] Example 2: Please refer to Figure 1 - Figure 2 , a tunnel fire smoke intelligent emergency disposal system for implementing a tunnel fire smoke intelligent control method. The system includes a tunnel fire smoke accident identification module, a tunnel fire smoke accident rescue information analysis module, and a tunnel fire smoke accident rescue information feedback module; The tunnel fire smoke accident identification module includes a tunnel on-site real-time image acquisition unit, a tunnel on-site real-time image preprocessing unit, a tunnel fire smoke accident image storage unit, and a tunnel fire smoke accident identification unit; The tunnel on-site real-time image acquisition unit collects tunnel on-site real-time image data through a monitoring camera; the tunnel on-site real-time image preprocessing unit preprocesses the tunnel on-site image data collected based on the tunnel on-site real-time image data and generates standard tunnel on-site real-time image data; the tunnel fire smoke accident image storage unit is used to store tunnel fire smoke accident image data; the tunnel fire smoke accident identification unit performs accident occurrence identification processing on tunnel fire smoke according to the standard tunnel on-site real-time image data and the tunnel fire smoke accident image data, and generates tunnel fire smoke accident identification data; The tunnel fire smoke accident rescue information analysis module includes a tunnel fire smoke accident occurrence type image storage unit, a tunnel fire smoke accident occurrence type analysis unit, a tunnel fire smoke accident rescue plan storage unit, a tunnel fire smoke accident rescue plan formulation unit, a tunnel fire smoke accident rescue department information storage unit, and a tunnel fire smoke accident rescue department identification unit; The tunnel fire smoke accident occurrence type image storage unit is used to store tunnel fire smoke accident occurrence type image data; the tunnel fire smoke accident occurrence type analysis unit performs occurrence type analysis processing on tunnel fire smoke accidents based on the standard tunnel on-site real-time image data combined with deep learning intelligent algorithms and the tunnel fire smoke accident occurrence type image data, and generates tunnel fire smoke accident occurrence type analysis data; the tunnel fire smoke accident rescue plan storage unit is used to store tunnel fire smoke accident rescue plan data; the tunnel fire smoke accident rescue plan formulation unit performs specific rescue plan analysis processing on tunnel fire smoke accidents according to the tunnel fire smoke accident occurrence type analysis data and the tunnel fire smoke accident rescue plan data, and generates tunnel fire smoke accident rescue plan analysis data; the tunnel fire smoke accident rescue department information storage unit is used to store tunnel fire smoke accident rescue department characteristic data; the tunnel fire smoke accident rescue department identification unit performs specific rescue execution department identity characteristic information analysis processing on tunnel fire smoke accidents according to the tunnel fire smoke accident rescue plan analysis data and the tunnel fire smoke accident rescue department characteristic data, and generates tunnel fire smoke accident rescue department characteristic identification data; The tunnel fire smoke accident rescue information feedback module includes a tunnel fire smoke accident occurrence location information acquisition unit, a tunnel fire smoke accident rescue information construction unit, and a tunnel fire smoke accident rescue information feedback unit; The information collection unit for the occurrence location of tunnel fire smoke accidents collects the spatial location coordinate data of tunnel fire smoke accidents through position sensors; the rescue information construction unit for tunnel fire smoke accidents is used to construct the rescue data for tunnel fire smoke accidents; the rescue information feedback unit for tunnel fire smoke accidents pushes the rescue data for tunnel fire smoke accidents to the tunnel fire monitoring platform through the Internet of Things communication network to perform the rescue information feedback operation for tunnel fire smoke accidents.

[0031] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent control method for tunnel fire smoke, characterized in that, The method includes the following steps: S1. Collect real-time image data of the tunnel site; S2. Preprocess the tunnel site image data collected based on the real-time image data of the tunnel site, and generate standard real-time image data of the tunnel site; S3. Perform accident occurrence identification processing of tunnel fire smoke based on the standard real-time image data of the tunnel site and the tunnel fire smoke accident image data, generate tunnel fire smoke accident identification data, and directly end the current tunnel fire smoke monitoring operation when no accident occurs; S4. When an accident occurs, perform accident type analysis processing of the tunnel fire smoke accident based on the standard real-time image data of the tunnel site and the tunnel fire smoke accident occurrence type image data, and generate tunnel fire smoke accident occurrence type analysis data; S5. Perform specific rescue plan analysis processing of the tunnel fire smoke accident based on the tunnel fire smoke accident occurrence type analysis data and the tunnel fire smoke accident rescue plan data, and generate tunnel fire smoke accident rescue plan analysis data; S6. Perform specific rescue execution department identity characteristic information analysis processing of the tunnel fire smoke accident based on the tunnel fire smoke accident rescue plan analysis data and the tunnel fire smoke accident rescue department characteristic data, and generate tunnel fire smoke accident rescue department characteristic identification data; S7. Collect the spatial position coordinate data of the tunnel fire smoke accident occurrence and construct tunnel fire smoke accident rescue data, and perform tunnel fire smoke accident rescue information feedback operations based on the tunnel fire smoke accident rescue data.

2. The intelligent control method for tunnel fire smoke according to claim 1, wherein: The S1 includes the following steps: S11. Online collect real-time on-site images inside the tunnel through the monitoring cameras installed in the tunnel, and generate a set of real-time tunnel on-site image data .

3. The intelligent control method for tunnel fire smoke according to claim 2, wherein: The S2 includes the following steps: S21. Use the adaptive filtering method to perform image data noise reduction preprocessing on the real-time image data set of the tunnel site in the real-time image data of the tunnel site described above, and generate a standard real-time image data set of the tunnel site , ; where represents the th standard real-time image data of the tunnel site, represents the maximum value of the number of standard real-time images of the tunnel site.

4. A method for intelligent control of tunnel fire smoke according to claim 3, characterized in that: The S3 includes the following steps: S31. Establish an image data set of tunnel fire smoke accidents , ; where represents the th image data of tunnel fire smoke accidents, represents the maximum value of the number of tunnel fire smoke accident images; S32. Use the iterative deepening search algorithm to described in and the described in perform image feature matching, and generate tunnel fire smoke accident recognition data based on the image feature matching results ; When and successfully match the image features, the tunnel fire smoke accident identification data is for an accident; When and no successful matches are found for the image features, the tunnel fire smoke accident identification data indicates that no accident has occurred, and at this time, the current tunnel fire smoke monitoring operation is directly terminated.

5. The intelligent control method for tunnel fire smoke according to claim 4, characterized in that: The S4 includes the following steps: S41. When the is an accident, establish an image data set of the accident occurrence type of tunnel fire smoke , ; where represents the image data of the th accident occurrence type of tunnel fire smoke, represents the maximum value of the number of accident occurrence types of tunnel fire smoke; S42. Compare the described in with the described in to perform image feature matching, search for the tunnel fire smoke accident occurrence type that matches the matched , and construct the analysis data of the tunnel fire smoke accident occurrence type . The specific operation steps for generating the analysis data of the tunnel fire smoke accident occurrence type are as follows: S421. Initialize the algorithm parameters, update the maximum number of iterations T, and the number of accident type search sooty tern populations N; S422. Calculate the fitness value of the accident type search for the Chinese Crested Tern, that is, calculate the fitness value of the in the search space and the ; S423, accident type search for sooty terns migrating, that is, in the Search the search space for Matching the ,Migration behaviors include conflict avoidance, aggregation, and renewal; S4231. Conflict avoidance: Update the position in the search space where the accident type search tern does not collide with other accident type search terns; S4232. Aggregation: Aggregation means that the accident type search for Sooty Terns moves closer to the optimal position among the adjacent accident type search for Sooty Terns in the search space while avoiding conflicts, that is, the accident type search for Sooty Terns searches for the position in the search space that best matches the ; position S4233. Update: An update refers to updating the position update trajectory of the optimal solution in the search space that is oriented to match the ; ; ; S424. The accident type searches for Sooty Terns to perform an attack behavior. The attack behavior means that during the migration process, the accident type searches for Sooty Terns to adjust the flight altitude or its own speed and attack angle in the search space to attack prey with an aerial hovering behavior, that is, to search for the in the search space that matches the and the ; S425. Search for the updated position of the sooty tern by accident type, that is, search for the position of the sooty tern in the search space for updating; S426. Calculate the fitness value, that is, calculate the fitness value of the in the search space and the , and search for the global in the search space that is most matched with the and the fitness value of the ; S427. When the maximum number of iterations is satisfied, output the most matching corresponding tunnel fire smoke accident occurrence type, and construct tunnel fire smoke accident occurrence type analysis data .

6. The intelligent control method for tunnel fire smoke according to claim 5, characterized in that: The S5 includes the following steps: S51. Establish a data set of rescue plans for tunnel fire smoke accidents , where represents the rescue plan data for tunnel fire smoke accidents corresponding to the th type of tunnel fire smoke accident S52. Use the breadth-first search algorithm to with the described in the perform keyword matching for the types of tunnel fire smoke accident occurrences, and analyze the corresponding , and generate tunnel fire smoke accident rescue plan analysis data through data identification .

7. The intelligent control method for tunnel fire smoke according to claim 6, characterized in that: The S6 includes the following steps: S61. Establish a characteristic data set of tunnel fire smoke accident rescue departments , where represents the characteristic data of the tunnel fire smoke accident rescue department corresponding to the execution of the ; S62. Use a bidirectional search algorithm to with the described in perform character matching for the tunnel fire smoke accident rescue plan, and search for the corresponding , and generate tunnel fire smoke accident rescue department feature recognition data .

8. The intelligent control method for tunnel fire smoke according to claim 7, wherein: The S7 includes the following steps: S71. Collect the spatial position coordinates of the monitoring camera that captures the real-time on-site images of the tunnel through the position sensor, and generate the spatial position coordinate data of the tunnel fire smoke accident ; S72. Combine the , the , the , the , and the for data collection and combination processing, and construct tunnel fire smoke accident rescue data ; S73. Push the feedback through the Internet of Things communication network to the tunnel fire monitoring platform to perform the feedback operation of tunnel fire smoke accident rescue information.

9. An intelligent emergency disposal system for tunnel fire smoke, which is used to implement the intelligent control method for tunnel fire smoke according to any one of claims 1-8, and is characterized in that: The system includes a tunnel fire smoke accident identification module, a tunnel fire smoke accident rescue information analysis module, and a tunnel fire smoke accident rescue information feedback module.

Images