Fire detection and emergency disposal system, method and device based on digital twinning

Abstract

This invention discloses a fire detection and emergency response system, comprising a data support layer, a physical sensing layer, a digital twin layer, an intelligent decision-making layer, and an emergency response layer. The data support layer provides support for the storage and management of detection data. The physical sensing layer acquires fire detection data and environmental and equipment status data within enclosed spaces. The digital twin layer constructs a full-element virtual mapping model of the enclosed space, achieving real-time synchronization, data fusion, and visualization between the physical entity and the virtual mapping model. The intelligent decision-making layer, based on the detection data and the status of the virtual mapping model, achieves early fire identification, risk assessment, and generates an emergency response plan. The emergency response layer executes the emergency response plan output by the intelligent decision-making layer. This fire detection and emergency response system enables extremely early and accurate detection, real-time synchronization between virtual and physical data, intelligent dynamic decision-making, and rapid closed-loop response, improving the initiative and effectiveness of fire prevention and control in enclosed spaces.

Description

Technical Field

[0001] This invention relates to the field of fire detection and emergency response technology, specifically to a system and method for early detection and intelligent emergency response to fires in enclosed spaces based on digital twins. Background Technology

[0002] Due to their enclosed spaces, poor ventilation, and complex structures, enclosed spaces are prone to problems such as smoke accumulation, rapid temperature increases, and the spread of toxic gases after a fire breaks out. Furthermore, rescue efforts are difficult, and fires can easily result in significant casualties and property damage. Current fire prevention and control technologies for enclosed spaces mainly focus on two aspects: Fire detection technology: Existing technologies mostly use traditional equipment such as point-type smoke detectors, heat detectors, and infrared flame detectors, which rely on a single physical parameter (such as smoke concentration and ambient temperature) to trigger alarms. This type of technology has obvious limitations: the weak signals generated by fires in the very early stage (smoldering stage) are easily interfered with by the complex environment of a confined space, resulting in alarm delays, usually 3 to 5 minutes; it cannot accurately locate the fire source, and is easily affected by dust, water vapor, etc., resulting in false alarms, with a false alarm rate as high as 5% to 8%.

[0003] Emergency response technology: Existing emergency response relies heavily on manual operation or pre-set fixed procedures, such as manually activating fire extinguishing devices or shutting down ventilation systems after an alarm is triggered. This mode has problems such as delayed response, manual response time is usually ≥10 minutes, and fixed response plans. It is impossible to dynamically adjust strategies according to the stage of fire development, the distribution of people in the confined space, and the operating status of equipment, which can easily lead to secondary disasters caused by untimely response or inappropriate measures.

[0004] The aforementioned existing technologies have the following core defects: ① Low accuracy in very early detection, delayed alarms and high false alarm rate; ② Insufficient integration of virtual and real systems, the digital twin model cannot accurately map the development of the fire; ③ Low level of intelligence in emergency response, lacking a dynamic and adaptive decision-making mechanism; ④ The detection and response processes are separated, failing to form a closed-loop management system for the entire process.

[0005] Digital twin technology enables real-time interaction and collaboration between virtual and real data by constructing virtual mappings of physical entities. It has been widely used in fields such as smart manufacturing and urban management. However, in the field of fire prevention and control in enclosed spaces, existing applications still have shortcomings. Most technical solutions only achieve scene visualization and do not deeply integrate early fire detection data. The synchronization accuracy between virtual models and physical entities is low, with a synchronization error of ≥1m, which cannot support accurate emergency decision-making. It lacks a closed-loop mechanism for the entire process of "detection-analysis-decision-response", and the core value of digital twins has not been fully realized.

[0006] Chinese patent application number 2021114957278 provides a fire monitoring method and system based on artificial intelligence and digital twin technology. This technical solution uses a database of components such as fire scene values ​​from a digital twin system and fire scene data collected by simulated experimental sensors to train an artificial intelligence model for fire situation judgment and monitoring. It is not directly applied to real fire prevention, especially fire detection and handling in confined spaces.

[0007] Therefore, there is an urgent need to build a closed-space fire detection and prevention system that integrates digital twin technology to achieve "very early and accurate detection, real-time synchronization between virtual and real systems, intelligent dynamic decision-making, and rapid closed-loop response," thereby improving the initiative and effectiveness of closed-space fire prevention and control. Summary of the Invention

[0008] The purpose of this invention is to overcome the shortcomings of the prior art and provide a system and method for early detection and intelligent emergency response to fires in enclosed spaces based on digital twins.

[0009] In a first aspect, the present invention provides a fire detection and emergency response system, comprising a data support layer, a physical sensing layer, a digital twin layer, an intelligent decision-making layer, and an emergency response layer. The data support layer provides support for the storage and management of detection data. The physical sensing layer is used to acquire fire detection data and environmental and equipment status data within a confined space. The digital twin layer constructs a virtual mapping model of all elements of the confined space, realizing real-time synchronization, data fusion, and visualization of the physical entity and the virtual mapping model. The intelligent decision-making layer realizes early fire identification, risk assessment, and generates an emergency response plan based on the detection data and the status of the virtual mapping model. The emergency response layer executes the emergency response plan output by the intelligent decision-making layer.

[0010] Furthermore, the data support layer further includes a distributed database, a data security unit, and a data interface unit.

[0011] Furthermore, the physical sensing layer further includes an early detection unit, an environment and equipment status unit, and a data transmission unit. The early detection unit consists of a distributed pyroelectric ion sensor, a microelectromechanical system (MEMS) smoke sensor, and an infrared temperature measurement array. The pyroelectric ion sensor collects changes in ion concentration at a frequency ≥20Hz with an accuracy of ±0.1ppb. The MEMS smoke sensor collects the concentration of weak smoke particles with a detection limit ≤0.001mg / m³. The infrared temperature measurement array monitors the regional temperature distribution within a range of -40℃ to 300℃ with an accuracy of ±0.05℃. The environment and equipment status unit consists of a temperature and humidity sensor, an oxygen concentration sensor, a combustible gas detector, and an equipment status monitoring module. It collects environmental parameters of the enclosed space in real time. The environmental parameters include temperature, humidity, oxygen concentration, combustible gas concentration, and the operating status of ventilation equipment and electrical equipment. The operating status includes voltage, current, and start / stop status. The data transmission unit has a data transmission delay of ≤100ms and a reliability of ≥99.99%.

[0012] Furthermore, the digital twin layer further includes a twin model construction unit, a virtual-real synchronization engine, and a visualization display unit. The twin model construction unit constructs a detailed three-dimensional model including a closed space structure, equipment layout, and detection and handling equipment, with a model accuracy of ≤0.1m. The virtual-real synchronization engine receives data from the physical perception layer and performs preprocessing, including data cleaning and noise filtering. The visualization unit displays detection data, fire development status, and emergency response progress in real time.

[0013] Furthermore, the intelligent decision-making layer further includes an early detection unit, a risk assessment unit, and an intelligent decision-making engine. The early detection unit first extracts the ion concentration mutation rate, smoke particle growth rate, and temperature gradient, and then achieves feature fusion through an improved Transformer algorithm. Combined with the environmental baseline of the confined space, the detection threshold is dynamically adjusted. The environmental baseline is dynamically generated through self-learning from the initial 72 hours of data. The accuracy rate of early fire detection is ≥99.7%, and the false alarm rate is ≤0.08%. The risk assessment unit uses the analytic hierarchy process (AHP) to assess the fire risk level and marks the risk areas with different colors in the digital twin model. The intelligent decision engine constructs a decision system that includes a rule base, a case library, and an AI prediction model. The rule base stores the basic handling rules required by fire protection regulations, the case library includes at least 500 similar cases of fire handling in enclosed spaces, and the AI ​​prediction model predicts the development trend of fires based on the LSTM algorithm.

[0014] Furthermore, the emergency response layer 105 further includes an equipment linkage unit, a personnel evacuation guidance unit, and a response effect feedback unit, wherein the equipment linkage unit is linked with the fire-fighting equipment, ventilation system, and electrical equipment in the enclosed space, and the linkage response time is ≤1s; The personnel evacuation guidance unit pushes dynamic evacuation routes through emergency broadcasts, indicator lights, and personnel mobile APPs in the enclosed space. The dynamic evacuation routes are planned in real time based on a digital twin model. The response effect feedback unit collects detection data after emergency response in real time and feeds it back to the intelligent decision-making layer. If the response effect does not meet expectations, it triggers iterative optimization of the solution.

[0015] Secondly, the present invention provides a fire detection and emergency response method, comprising the following steps: Step 201: System initialization and baseline establishment. After the system starts, the physical perception layer collects the initial environmental data of the enclosed space, the digital twin layer builds a three-dimensional model based on the initial environmental data, and the intelligent decision-making layer generates the environmental baseline and normal operation threshold based on the initial environmental data. Step 202: Very early data acquisition and virtual-real synchronization: The physical sensing layer collects ion concentration, smoke concentration and temperature in real time, and transmits them to the digital twin layer after preprocessing to drive the virtual model to update dynamically and realize the visualization of fire characteristic parameters. Step 203: Fire identification and risk assessment. The intelligent decision-making layer extracts multi-dimensional features from the data collected by the physical perception layer and identifies very early fire signals through a fusion algorithm. If an anomaly is detected, a risk assessment is initiated, and the risk level is determined by combining personnel and equipment data. Step 204: Intelligent decision-making and plan execution. The decision engine combines risk level and fire development prediction to generate dynamic response plans. The emergency response layer automatically links equipment for execution and provides real-time feedback on the response results. Step 205: Determine whether the fire is under control. If the fire is effectively controlled, proceed to step 206; otherwise, return to step 204 to regenerate and execute the response plan. Step 206: Closed-loop optimization and recovery. If the treatment effect meets the standards, the system records the treatment process and restores the system to the monitoring state.

[0016] Thirdly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the fire detection and emergency response method as described in the second aspect.

[0017] Fourthly, the present invention provides a fire detection and emergency response device, comprising: One or more processors; Memory; and One or more computer programs, wherein the one or more computer programs are stored in the memory and configured to be executed by the one or more processors, wherein the processors execute the computer programs to implement the steps of the fire detection and emergency response method as described in the second aspect.

[0018] The fire detection and emergency response system, method and device of the present invention can achieve very early and accurate detection, real-time synchronization of virtual and real data, intelligent dynamic decision-making and rapid closed-loop response, thereby improving the initiative and effectiveness of fire prevention and control in confined spaces. Attached Figure Description

[0019] Figure 1 Fire detection and emergency response system architecture diagram.

[0020] Figure 2 Schematic diagram of the data support layer.

[0021] Figure 3 Schematic diagram of the physical perception layer.

[0022] Figure 4 Schematic diagram of the composition of the digital twin layer.

[0023] Figure 5 : Schematic diagram of the composition of the intelligent decision-making layer.

[0024] Figure 6 Schematic diagram of the emergency response layer.

[0025] Figure 7 Flowchart of fire detection and emergency response methods.

[0026] Figure 8 Schematic diagram of fire detection and emergency response device. Detailed Implementation

[0027] To make the technical problems, technical solutions and beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0028] Please refer to Figure 1 The fire detection and emergency response system architecture diagram of this invention includes a data support layer 101, a physical sensing layer 102, a digital twin layer 103, an intelligent decision-making layer 104, and an emergency response layer 105. The data support layer 101 is responsible for data communication and storage with the other four modules, as well as data security and management support. Further reference... Figure 2The data support layer 101 is further comprised of a distributed database 1011, a data security unit 1012, and a data interface unit 1013. The distributed database 1011 adopts a "MySQL + InfluxDB + Redis" architecture. MySQL stores structured data, including device information and handling logs. InfluxDB stores time-series probe data, and Redis implements high-frequency data caching. The data storage duration is ≥2 years, and it supports fast queries by time, region, and device type.

[0029] The data security unit 1012 uses AES-256 encryption to transmit and store data, and ensures data security through device access authentication, user permission classification, and operation log auditing, meeting the Level 3 compliance requirements of the Information Security Protection 2.0 standard.

[0030] The data interface unit 1013 provides a standardized API interface, which supports connection to enterprise security platforms and fire emergency command center systems to achieve data sharing and collaborative handling.

[0031] refer to Figure 3 A schematic diagram of the physical sensing layer is provided. The physical sensing layer 102 serves as the system's data acquisition terminal, used to acquire early-stage fire characteristic parameters and environmental and equipment status data within a confined space. It further includes: The very early detection unit 1021 consists of a distributed pyroelectric ion sensor, a microelectromechanical system (MEMS) smoke sensor, and an infrared temperature measurement array. It is deployed in key areas of a confined space. The pyroelectric ion sensor collects changes in ion concentration at a frequency ≥20Hz with an accuracy of ±0.1ppb. The MEMS smoke sensor collects the concentration of weak smoke particles with a detection limit ≤0.001mg / m³. The infrared temperature measurement array monitors the regional temperature distribution within a range of -40℃ to 300℃ with an accuracy of ±0.05℃.

[0032] Environment and equipment status unit 1022: includes temperature and humidity sensor, oxygen concentration sensor, combustible gas detector and equipment status monitoring module, which collects environmental parameters of the enclosed space in real time. The environmental parameters include temperature, humidity and oxygen concentration, combustible gas concentration and the operating status of ventilation equipment and electrical equipment. The operating status includes voltage, current and start / stop status.

[0033] Data transmission unit 1023: adopts wired + wireless dual backup transmission mode. Wired transmission achieves large-capacity data transmission through industrial Ethernet, and wireless transmission achieves data coverage in remote areas through 5G industrial modules. The data transmission latency of the data transmission unit 1023 is ≤100ms and the reliability is ≥99.99%.

[0034] refer to Figure 4A schematic diagram of the digital twin layer is provided. The digital twin layer 103 constructs a full-element virtual mapping model of a closed space, realizing real-time synchronization, data fusion, and visualization of physical entities and virtual models. It further includes: The twin model building unit 1031 is based on BIM+GIS+point cloud scanning technology to build a three-dimensional detailed model that includes the structure of the enclosed space, equipment layout, and detection and treatment equipment. The model accuracy is ≤0.1m and supports scene scaling, roaming, and perspective switching.

[0035] The virtual-real synchronization engine 1032 adopts a synchronization mechanism of edge computing + cloud collaboration. The edge receives physical perception layer data in real time and performs preprocessing, including data cleaning and noise filtering. The cloud uses Kalman filtering algorithm to achieve data fusion and drive the virtual model to update dynamically. For example, in the very early stage of a fire, the corresponding area in the virtual model displays temperature changes with thermal imaging color gradation and synchronously marks abnormal points of ion concentration and smoke concentration.

[0036] The visualization unit 1033 is used to develop visualization interfaces for web, large screen and mobile terminals. It supports the real-time display of detection data, fire development status and emergency response progress using digital twin models as carriers. It provides data drill-down function and allows users to view detailed parameters and historical data of sensors or equipment in the virtual model.

[0037] refer to Figure 5 A schematic diagram of the intelligent decision-making layer is provided. The intelligent decision-making layer 104, based on detection data and the state of the twin model, enables very early fire identification, risk assessment, and emergency response plan generation. It further includes: The very early detection unit 1041 constructs a multi-dimensional feature fusion + dynamic threshold detection model. First, it extracts 15 core features such as ion concentration mutation rate, smoke particle growth rate, and temperature gradient. Then, it achieves feature fusion through an improved Transformer algorithm. Combined with the environmental baseline of the confined space, the detection threshold is dynamically adjusted. The environmental baseline is dynamically generated by self-learning from the initial 72 hours of data. The accuracy of very early fire detection is ≥99.7%, and the false alarm rate is ≤0.08%.

[0038] Risk assessment unit 1042, based on the results of early fire identification, combined with real-time location data of personnel in confined spaces, equipment value, and distribution of combustible materials, uses the analytic hierarchy process to assess the fire risk level and marks risk areas with different colors in the digital twin model.

[0039] The intelligent decision engine 1043 constructs a decision-making system that includes a rule base, a case base, and an AI prediction model. The rule base stores the basic handling rules required by fire protection regulations; the case base contains at least 500 similar cases of fire handling in enclosed spaces; the AI ​​prediction model predicts the fire development trend based on the LSTM algorithm, such as the spread range and temperature change curve within 10 minutes. The system combines the risk level and the prediction results to automatically generate dynamic handling plans, such as initiating local smoke extraction and cooling for Level 1 risk (very early stage) and initiating full-area fire extinguishing and personnel evacuation for Level 4 risk (fire spread).

[0040] refer to Figure 6 The diagram illustrates the composition of the emergency response layer. The emergency response layer 105 executes the response plan output by the intelligent decision-making layer 104, realizing automated and precise emergency response to fires in enclosed spaces. The emergency response layer 105 further includes an equipment linkage unit 1051, a personnel evacuation guidance unit 1052, and a response effect feedback unit 1053. The equipment linkage unit 1051 is linked with the fire-fighting equipment, ventilation system, and electrical equipment in the enclosed space, supporting automatic control, such as activating the gas extinguishing device in the corresponding area, closing the ventilation valves near the ignition point, and cutting off the power supply to the faulty electrical equipment. Its linkage response time is ≤1 second.

[0041] The personnel evacuation guidance unit 1052 pushes dynamic evacuation routes through emergency broadcasts, indicator lights, and personnel mobile APP in the enclosed space. The dynamic evacuation routes are planned in real time based on digital twin models to avoid fire spread areas. Combined with UWB positioning data, it monitors the progress of personnel evacuation in real time and provides accurate reminders to stranded personnel.

[0042] The emergency response feedback unit 1053 collects detection data after emergency response in real time, such as the temperature drop and changes in smoke concentration, and further feeds it back to the intelligent decision layer 104. If the response effect does not meet expectations, such as the temperature not dropping by 10°C within 10 minutes, the solution is iteratively optimized.

[0043] refer to Figure 7 A flowchart of a fire detection and emergency response method, comprising the following steps: Step 201: System initialization and baseline establishment: After the system starts, the physical perception layer 102 collects the initial environmental data of the enclosed space, such as the environmental data within 72 hours. The digital twin layer 103 completes the construction of a three-dimensional model based on the environmental data. The intelligent decision layer 104 generates the environmental baseline and normal operation threshold based on the initial data. Step 202: Very early data acquisition and virtual-real synchronization: The physical sensing layer 102 collects parameters such as ion concentration, smoke concentration, and temperature in real time, and transmits them to the digital twin layer 103 after preprocessing to drive the virtual model to update dynamically and realize the visualization of fire characteristic parameters. Step 203: Fire Identification and Risk Assessment: The intelligent decision layer 104 extracts multi-dimensional features from the collected data, identifies very early fire signals through a fusion algorithm, and initiates a risk assessment if an anomaly is detected, combining personnel and equipment data to determine the risk level. Step 204: Intelligent Decision-Making and Solution Execution: The decision engine combines risk level and fire development prediction to generate a dynamic response plan, which is automatically executed by the emergency response layer 105 and the response effect is fed back in real time. Step 205: Determine whether the fire is under control. If the fire is effectively controlled, proceed to step 206; otherwise, return to step 204 to regenerate and execute the response plan. Step 206: Closed-loop optimization and recovery: If the treatment effect meets the standards, the system records the treatment process and restores the system to the monitoring state.

[0044] refer to Figure 8 A schematic diagram of the fire detection and emergency response device is shown. The fire detection and emergency response device of the present invention further includes one or more memories 20 and one or more processors 30, wherein the one or more computer programs are stored in the memories 20 and configured to be executed by the one or more processors 30, and the processors 30 implement the steps of the fire detection and emergency response method when executing the computer programs.

[0045] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0046] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A fire detection and emergency response system, characterized in that, It includes a data support layer, a physical sensing layer, a digital twin layer, an intelligent decision-making layer, and an emergency response layer. The data support layer provides support for the storage and management of detection data. The physical sensing layer is used to acquire fire detection data and environmental and equipment status data in enclosed spaces. The digital twin layer constructs a virtual mapping model of all elements in the enclosed space, realizing real-time synchronization, data fusion, and visualization of physical entities and virtual mapping models. The intelligent decision-making layer realizes early fire identification, risk assessment, and generates emergency response plans based on the detection data and the status of the virtual mapping model. The emergency response layer executes the emergency response plans output by the intelligent decision-making layer.

2. The fire detection and emergency response system according to claim 1, characterized in that, The data support layer further includes a distributed database, a data security unit, and a data interface unit.

3. The fire detection and emergency response system according to claim 1, characterized in that, The physical sensing layer further includes an early detection unit, an environment and equipment status unit, and a data transmission unit. The early detection unit consists of a distributed pyroelectric ion sensor, a microelectromechanical system (MEMS) smoke sensor, and an infrared temperature measurement array. The pyroelectric ion sensor collects changes in ion concentration at a frequency ≥20Hz with an accuracy of ±0.1ppb. The MEMS smoke sensor collects the concentration of weak smoke particles with a detection limit ≤0.001mg / m³. The infrared temperature measurement array monitors the regional temperature distribution within a range of -40℃ to 300℃ with an accuracy of ±0.05℃. The environment and equipment status unit consists of a temperature and humidity sensor, an oxygen concentration sensor, a combustible gas detector, and an equipment status monitoring module. It collects environmental parameters of the enclosed space in real time. The environmental parameters include temperature, humidity, oxygen concentration, combustible gas concentration, and the operating status of ventilation equipment and electrical equipment. The operating status includes voltage, current, and start / stop status. The data transmission unit has a data transmission delay of ≤100ms and a reliability of ≥99.99%.

4. The fire detection and emergency response system according to claim 1, characterized in that, The digital twin layer further includes a twin model construction unit, a virtual-real synchronization engine, and a visualization display unit. The twin model construction unit constructs a detailed three-dimensional model that includes a closed space structure, equipment layout, and detection and disposal equipment, with a model accuracy of ≤0.1m. The virtual-real synchronization engine receives data from the physical perception layer and performs preprocessing, including data cleaning and noise filtering. The visualization unit displays detection data, fire development status, and emergency response progress in real time.

5. The fire detection and emergency response system according to claim 1, characterized in that, The intelligent decision-making layer further includes an early detection unit, a risk assessment unit, and an intelligent decision-making engine. The early detection unit first extracts the ion concentration mutation rate, smoke particle growth rate, and temperature gradient. Then, it achieves feature fusion through an improved Transformer algorithm and dynamically adjusts the detection threshold based on the environmental baseline of the confined space. The environmental baseline is dynamically generated through self-learning from the initial 72 hours of data. The accuracy rate of early fire detection is ≥99.7%, and the false alarm rate is ≤0.08%. The risk assessment unit uses the analytic hierarchy process (AHP) to assess the fire risk level and marks the risk areas with different colors in the digital twin model. The intelligent decision engine constructs a decision system that includes a rule base, a case library, and an AI prediction model. The rule base stores the basic handling rules required by fire protection regulations, the case library includes at least 500 similar cases of fire handling in enclosed spaces, and the AI ​​prediction model predicts the development trend of fires based on the LSTM algorithm.

6. The fire detection and emergency response system according to claim 1, characterized in that, The emergency response layer 105 further includes an equipment linkage unit, a personnel evacuation guidance unit, and a response effect feedback unit, wherein the equipment linkage unit is linked with the fire-fighting equipment, ventilation system, and electrical equipment in the enclosed space, and the linkage response time is ≤1s. The personnel evacuation guidance unit pushes dynamic evacuation routes through emergency broadcasts, indicator lights, and personnel mobile APPs in the enclosed space. The dynamic evacuation routes are planned in real time based on a digital twin model. The response effect feedback unit collects detection data after emergency response in real time and feeds it back to the intelligent decision-making layer. If the response effect does not meet expectations, it triggers iterative optimization of the solution.

7. A method for fire detection and emergency response, characterized in that, Includes the following steps: Step 201: System initialization and baseline establishment. After the system starts, the physical perception layer collects the initial environmental data of the enclosed space, the digital twin layer builds a three-dimensional model based on the initial environmental data, and the intelligent decision-making layer generates the environmental baseline and normal operation threshold based on the initial environmental data. Step 202: Very early data acquisition and virtual-real synchronization: The physical sensing layer collects ion concentration, smoke concentration and temperature in real time, and transmits them to the digital twin layer after preprocessing to drive the virtual model to update dynamically and realize the visualization of fire characteristic parameters. Step 203: Fire identification and risk assessment. The intelligent decision-making layer extracts multi-dimensional features from the data collected by the physical perception layer and identifies very early fire signals through a fusion algorithm. If an anomaly is detected, a risk assessment is initiated, and the risk level is determined by combining personnel and equipment data. Step 204: Intelligent decision-making and plan execution. The decision engine combines risk level and fire development prediction to generate dynamic response plans. The emergency response layer automatically links equipment for execution and provides real-time feedback on the response results. Step 205: Determine whether the fire is under control. If the fire is effectively controlled, proceed to step 206; otherwise, return to step 204 to regenerate and execute the response plan. Step 206: Closed-loop optimization and recovery. If the treatment effect meets the standards, the system records the treatment process and restores the system to the monitoring state.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the fire detection and emergency response method as described in claim 7.

9. A fire detection and emergency response device, comprising: One or more processors; Memory; as well as One or more computer programs, wherein the one or more computer programs are stored in the memory and configured to be executed by the one or more processors, characterized in that the processors, when executing the computer programs, implement the steps of the fire detection and emergency response method as described in claim 7.

Images