Fire-fighting evacuation guiding method and system

Abstract

The invention relates to the technical field of fire safety, in particular to a fire evacuation guiding method and system. The method comprises the following steps: obtaining building layout information; performing evacuation path network modeling according to the building layout information to obtain evacuation channel data; performing path congestion assessment according to the evacuation channel data to obtain path congestion data; acquiring fire-fighting sensing data; sensing a fire position in real time based on the fire-fighting sensing data to obtain fire position data; performing fire spreading direction prediction based on the fire position data to obtain fire spreading direction data; identifying a fire hazard area according to the fire spreading direction; building structure heat conduction simulation is carried out according to the fire hazard area to obtain building structure heat conduction data; and identifying an unstable structure based on the heat conduction data of the building structure to obtain unstable structure data. The fire-fighting evacuation safety and evacuation efficiency are improved based on the fire-fighting safety technology.

Description

Technical Field

[0001] The present invention relates to the field of fire safety technology, and in particular to a fire evacuation guidance method and system. Background Art

[0002] Traditional fire evacuation guidance methods and systems typically rely on fixed, pre-set evacuation routes and static signage, lacking the ability to perceive fire dynamics in real time. During a fire, the fire's location, spread direction, smoke density, and structural damage rapidly change. However, traditional systems are often unable to perceive these changes and can only provide guidance based on pre-designed evacuation routes. This can lead to evacuation routes being blocked by dense smoke, high-temperature areas, or structural damage, increasing evacuation risks. They lack real-time analysis of the building's internal structural stability, unable to identify unstable areas of critical load-bearing components such as beams and columns caused by high temperatures. They also fail to account for changes in the actual capacity of the routes, such as severe congestion in localized corridors caused by dense crowds, further reducing evacuation efficiency and safety. While some systems possess sensor data collection capabilities, they lack the ability to integrate and analyze multi-source information. They also lack key processing steps such as fire prediction, structural heat conduction analysis, and dynamic route optimization, making them incapable of providing rapid response and scientific guidance in complex fire environments. They suffer from significant deficiencies such as poor adaptability, low intelligence, and limited practicality. Summary of the Invention

[0003] Based on this, it is necessary for the present invention to provide a fire evacuation guidance method and system to solve at least one of the above technical problems.

[0004] To achieve the above object, a fire evacuation guidance method includes the following steps: Step S1: Acquire building layout information; perform evacuation path network modeling based on the building layout information to obtain evacuation channel data; perform path congestion assessment based on the evacuation channel data to obtain path congestion data; Step S2: Obtain fire sensor data; perceive the fire location in real time based on the fire sensor data to obtain fire location data; predict the fire spread direction based on the fire location data to obtain fire spread direction data; identify the fire hazard area based on the fire spread direction; Step S3: Performing a heat conduction simulation of the building structure according to the fire hazard zone to obtain heat conduction data of the building structure; identifying unstable structures based on the heat conduction data of the building structure to obtain unstable structure data; performing a high-temperature damage prediction based on the unstable structure data to obtain high-temperature damage data of the building structure; Step S4: Optimize the evacuation path based on the building structure high temperature damage data and the path congestion data to obtain evacuation path optimization data; publish fire evacuation guidance information based on the evacuation path optimization data to obtain fire evacuation guidance information release data.

[0005] The present invention realizes the intelligent optimization and real-time update of evacuation guidance paths in complex fire environments through multi-source data fusion and dynamic calculation, and has significant technical advantages. Its beneficial effects are reflected in the following aspects: it can combine building layout information to build an evacuation path network, identify locations prone to congestion such as narrow passages, concentrated obstacles and complex corners, thereby realizing dynamic evaluation of path traffic capacity; by integrating multiple fire sensor data such as smoke, temperature, and flame, it can perceive the location of fire and smoke concentration distribution in real time, and predict the airflow diffusion trend in combination with wind direction and building orientation data, effectively improving the timeliness and accuracy of fire perception; it can scientifically judge the direction of fire spread and the fire spread path based on the fire source intensity and the heat flux calculation of door and window structures, and timely calibrate the dangerous area, Prevent people from accidentally entering high-risk spaces; integrate the heat conduction characteristics of building structures, simulate and calculate the temperature rise effects of key components such as beams and columns, identify potential structural instability units, and determine high-temperature damage areas in advance, which helps to avoid the risk of structural collapse; integrate path congestion data and high-temperature damage location information to construct a dynamic evacuation path reconstruction model to improve path adaptability and the shortest safe passage efficiency; finally, by publishing guidance information on the optimal evacuation path in real time, trapped people are guided to evacuate quickly to safe areas, significantly improving the intelligence level of fire emergency response, the real-time nature of path adjustment, and the safety of personnel evacuation. This method generally breaks through the limitations of traditional evacuation guidance systems that cannot perceive dynamic fire conditions, structural risks, and congestion conditions, and realizes intelligent evacuation guidance under the comprehensive factors of structural safety, traffic capacity, and fire spread, effectively improving the scientificity, adaptability, and practicality of fire emergency systems.

[0006] Preferably, step S1 is specifically as follows: Step S11: Obtain building layout information; Step S12: performing spatial connectivity analysis based on the building layout information to obtain spatial connectivity data; Step S13: identifying a traversable path based on the spatial connectivity relationship data to obtain traversable path data; Step S14: Modeling an evacuation path network based on the traversable path data to obtain evacuation channel data; Step S15: performing path congestion assessment based on the evacuation channel data to obtain path congestion data.

[0007] By accurately extracting building layout information, the present invention can achieve a global understanding of building structures and traffic areas, laying the foundation for the scientific modeling of subsequent evacuation paths; further based on the analysis of spatial connectivity relationships, it can identify the actual accessibility between different rooms, corridors, entrances and exits, and eliminate paths that are inaccessible due to closure, blockage or structural discontinuity, thereby improving the accuracy and practicality of traffic path identification; the accessible path data extracted on this basis not only reflects the spatial topological connection relationship, but also excludes non-valid paths such as dead ends and closed spaces, thereby improving the rationality of network modeling; the constructed evacuation path network model has the advantages of clear structure and real-time update, which is suitable for rapid adjustment under subsequent dynamic changes in fire conditions; in addition, combining information such as the width, corner layout, and obstacle distribution of evacuation channels to carry out path congestion assessment, it can identify potential bottlenecks and high-density risk areas, and help to dynamically avoid areas with dense crowds and narrow spaces, thereby realizing a safer, more efficient, and traffic-adaptive intelligent evacuation path guidance mechanism, effectively making up for the problems of static planning, poor adaptability, and guidance failure of traditional evacuation paths.

[0008] Preferably, step S15 is specifically as follows: Step S151: Calculate the channel width according to the evacuation channel data to obtain channel width data; Step S152: identifying narrow channels based on the channel width data to obtain narrow channel data; Step S153: performing corner area analysis based on the narrow channel data to obtain corner area data; Step S154: performing obstacle area recognition based on the corner area data to obtain obstacle area data; Step S155: performing path congestion assessment based on the obstacle data and the corner area data to obtain path congestion data.

[0009] By accurately calculating the geometric width of the evacuation channel, the present invention can obtain the real scale basic data of each traffic area, providing quantitative support for subsequent bottleneck identification and traffic efficiency evaluation; the narrow channels further identified can serve as important signs of potential congestion areas, facilitating the early formulation of detour strategies or encrypted guidance signs; on this basis, corner area analysis is carried out to help locate areas where the direction of pedestrian flow changes frequently and the turning radius is small. These locations often have the risk of line of sight obstruction, crowd accumulation and movement slowdown; by identifying the obstacle positions associated with the corner areas, a local traffic restriction model can be further constructed to identify bottleneck nodes caused by physical obstruction or space compression; finally, the path congestion assessment is carried out by combining the corner and obstacle factors, which can effectively quantify the dynamic evacuation pressure bearing capacity of each channel in the event of a fire, realize early warning and evacuation optimization of local high-congestion risk areas, significantly improve the practicality, refinement and responsiveness of evacuation path guidance, and break through the limitation of traditional systems that only rely on static path planning and cannot perceive local blockage and dynamic load.

[0010] Preferably, step S2 is specifically as follows: Step S21: Acquire fire sensor data and perform smoke sensor feature extraction to obtain smoke sensor data; calculate real-time smoke concentration based on the smoke sensor data to obtain smoke concentration data; construct a smoke concentration gradient field based on the smoke concentration data; and identify high smoke concentration areas based on the smoke concentration gradient field. Step S22: Identify the building orientation angle according to the building layout information to obtain building orientation angle data; obtain wind direction data; perform wind corridor detection on the building orientation angle data according to the wind direction data to obtain wind corridor data; Step S23: performing maximum wind speed statistics based on the wind corridor data to obtain maximum wind speed data; performing airflow diffusion trend prediction for the high smoke concentration area based on the maximum wind speed data to obtain airflow diffusion trend data; Step S24: Detecting the fire location based on the airflow diffusion trend to obtain fire location data; Step S25: Predict the fire spread direction based on the fire location data to obtain fire spread direction data; and identify the fire hazard area based on the fire spread direction.

[0011] The present invention can construct a continuously changing smoke concentration gradient field by extracting smoke sensing features and calculating concentration data, thereby identifying high-risk areas with dense smoke and extremely poor visibility near the fire source, effectively breaking through the problem that traditional systems cannot perceive the details of smoke distribution; combining building orientation information and wind direction data to carry out wind corridor detection, it can identify building passages and connecting openings that are significantly affected by external wind force, and provide key boundary conditions for modeling fire spread trends; performing maximum wind speed statistics based on wind corridors can accurately quantify the impact of airflow velocity on smoke and heat migration paths, and provide highly reliable aerodynamic input parameters for predicting fire spread trends. By simulating the airflow diffusion trend, the indirect perception of the fire location is achieved, and the ability to locate the fire source in visual blind spots and sensor blind spots is improved. The fire spread direction is further predicted based on the fire location data, which can not only determine the evolution trend of the fire toward the inside of the building or in the vertical direction, but also timely identify the channel blockage and spread danger areas. Ultimately, the dynamic identification and spatial distribution of the fire danger zone can be achieved, providing data support for the subsequent structural heat loss risk analysis and evacuation route planning, comprehensively enhancing the system's perception accuracy, response speed and scientific guidance of the spatiotemporal evolution of the fire, and effectively overcoming the lack of adaptability of traditional methods to complex fire situations.

[0012] Preferably, step S25 is specifically as follows: Step S251: Calculating fire source intensity based on fire location data to obtain fire source intensity data; Step S252: performing door and window structure detection based on the fire location data to obtain door and window structure data; Step S253: performing hole analysis based on the door and window structure data to obtain door and window hole data; Step S254: performing heat flux penetration analysis on the door and window hole data based on the fire source intensity data to obtain heat flux penetration data; identifying the fire spread direction based on the heat flux penetration data to obtain fire spread direction data; Step S255: predicting the fire spread path based on the fire spread direction data to obtain fire spread path data; Step S256: Identify fire hazard areas based on the fire spread path data.

[0013] By quantitatively calculating fire source intensity, the present invention can clearly determine the heat release capacity of each fire source point, providing basic heat source parameters for subsequent heat transfer and danger zone identification. Doors and windows are the weakest links in building structures where heat flux migration is most likely to occur. Their structural inspection and hole analysis can not only determine the areas where heat can penetrate, but also assist in identifying potential channels for fire to spread across space. On this basis, integrating fire source intensity data to conduct heat flux penetration analysis can not only simulate the rate and energy distribution of flames and high-temperature gases penetrating the door and window structure, but also accurately restore local spatial temperature rise changes, improving the physical realism of fire propagation direction judgment. Using heat flux penetration data to identify the spread direction can break through the traditional static judgment method that relies on images or temperature sensing, and realize real-time modeling of the dynamic trend of fire. Further development of fire spread path prediction can form a dynamic propagation grid with multiple paths, multiple speeds, and multiple heat levels based on the superposition of multiple factors such as structural spatial distribution, heat conduction boundaries, and wind direction data, significantly improving the ability to predict the evolution trend of fire inside complex buildings. The fire hazard areas finally identified not only have high spatial resolution and dynamic timeliness, but also provide precise boundary support for subsequent evacuation route optimization and structural thermal instability analysis, thus breaking through the problems of insufficient fire spread modeling and rough risk area identification in traditional systems, and realizing truly intelligent dynamic evacuation assistance.

[0014] Preferably, step S3 is specifically as follows: Step S31: Performing a building structure heat conduction simulation based on the fire hazard zone to obtain building structure heat conduction data; Step S32: performing structural beam loosening detection based on the building structure heat conduction data to obtain structural beam loosening data; Step S33: performing structural column embrittlement detection based on the building structure heat conduction data to obtain structural column embrittlement data; Step S34: Integrate the structural beam loosening data and the structural column embrittlement data to obtain unstable structural data; Step S35: Perform high-temperature damage prediction based on the unstable structure data to obtain high-temperature damage data of the building structure.

[0015] By incorporating a heat conduction simulation mechanism into building structures, this method enables dynamic assessment of the response of a building's load-bearing system to fire, effectively addressing the shortcomings of traditional methods in structural safety assessment. By simulating the heat conduction process, the heat propagation path and temperature rise distribution of fire heat within the building's internal structure can be accurately reproduced, providing a physical basis for subsequent identification of thermal responses in key locations. Combining heat conduction data with structural beam loosening detection can identify increased interfacial voids, decreased bond strength, and unbalanced load systems caused by high temperatures, accurately reflecting stability degradation trends. Column embrittlement detection analyzes the effects of high temperatures through dimensions such as longitudinal reinforcement corrosion, stirrup cracking, and concrete envelope failure, dynamically identifying the instability risk of compression components. Fusion analysis of beam loosening and column embrittlement data allows the construction of an instability risk map for the entire structural system, accurately identifying potential risk areas in the event of a fire. Based on this, high-temperature damage prediction not only improves the predictability of structural damage trends during a fire but also assists in determining areas where traffic and stagnation are undesirable. This provides key support for optimizing fire evacuation routes and emergency response deployment, significantly improving evacuation safety and intelligent decision-making.

[0016] Preferably, step S32 is specifically as follows: Step S321: identifying high-temperature areas of structural beams based on the building structure heat conduction data to obtain high-temperature area data of the structural beams; Step S322: performing interface gap detection based on the high temperature area data of the structural beam to obtain interface gap data; Step S323: performing a reinforcement-concrete bond slip simulation based on the interface void data to obtain reinforcement-concrete bond slip data; determining the degree of firmness based on the reinforcement-concrete bond slip data to obtain degree of firmness data; Step S324: performing crack propagation analysis based on the firmness data to obtain crack propagation data; Step S325: performing crack bifurcation identification based on the crack propagation data to obtain crack bifurcation data; Step S326: Evaluate the looseness of the structural beam according to the crack bifurcation data to obtain the looseness data of the structural beam.

[0017] By identifying high-temperature areas in structural beams, this method pinpoints key areas of thermal damage caused by fire heat, providing a location basis for subsequent micro-deterioration analysis. Furthermore, interfacial void detection within these high-temperature areas effectively identifies the risk of bond failure between concrete and steel due to factors such as uneven thermal expansion and material thermal degradation. Simulating reinforced concrete bond slip, combined with interfacial void data, quantifies the extent to which high fire temperatures weaken the interfacial shear force transmission mechanism, further enabling the determination of structural robustness trends. By modeling the strain concentration and crack evolution behavior caused by bond slip, crack propagation paths and rates can be accurately extracted. Based on this, crack bifurcation identification is performed, not only helping to reveal the complex propagation behavior of cracks but also assisting in inferring the stress redistribution process caused by cracks. Ultimately, crack bifurcation characteristics are used to infer the integrity and stability of structural beams after fire, accurately identifying their degree of loosening based on micro-deterioration mechanisms. This provides important technical support for mechanical response assessment and structural safety early warning of building structures under fire, and also provides a solid structural basis for the dynamic updating of fire evacuation routes and traffic safety assessment.

[0018] Preferably, step S33 is specifically as follows: Step S331: identifying high-temperature areas of structural columns based on the building structure heat conduction data to obtain high-temperature area data of the structural columns; Step S332: performing longitudinal reinforcement thermal corrosion analysis based on the high temperature area data of the structural column to obtain longitudinal reinforcement thermal corrosion data; Step S333: detecting stirrup transverse cracks based on the high-temperature area data of the structural column to obtain stirrup transverse crack data; Step S334: performing concrete envelope failure determination based on stirrup transverse crack data to obtain concrete envelope failure data; Step S335: identifying the longitudinal reinforcement thermal corrosion area based on the longitudinal reinforcement thermal corrosion data; identifying the concrete envelope failure area based on the concrete envelope failure data; Step S336: performing an intersection operation of the structural column embrittlement area according to the longitudinal reinforcement thermal corrosion area and the concrete envelope failure area to obtain structural column embrittlement data.

[0019] The present invention identifies and locks the key areas of the structural columns that are most severely heated through high-temperature areas, providing precise targets for subsequent corrosion and crack monitoring. Based on the analysis of thermal corrosion of longitudinal reinforcement in high-temperature areas, it can reveal the material performance degradation and microstructural damage of steel bars induced by temperature, and promptly reflect the trend of reduced load-bearing capacity; at the same time, by detecting transverse cracks in stirrups, it accurately captures the crack damage that appears in the concrete protective layer, providing an important basis for the determination of concrete envelope failure. Combined with the intersection analysis of the longitudinal reinforcement thermal corrosion area and the concrete envelope failure area, the brittle area of ​​the structural column is systematically determined to achieve a comprehensive assessment of the structural integrity and stability. This multi-source data analysis method that integrates heat conduction, material corrosion and crack damage information can not only improve the scientificity and real-time nature of structural safety warnings during fires, but also provide a solid structural risk foundation for fire evacuation guidance, ensure the safety and feasibility of evacuation routes, and effectively reduce the risk of casualties in fires.

[0020] Preferably, step S4 is specifically as follows: Step S41: calibrating an impassable path based on the high-temperature damage data of the building structure to obtain impassable path data; Step S42: Predicting evacuation time based on the path congestion data to obtain road evacuation time data; Step S43: performing detour route analysis based on the road evacuation time data to obtain detour route data; Step S44: Optimizing the shortest evacuation route based on the impassable route data and the detour route data to obtain optimized evacuation route data; Step S45: Publish fire evacuation guidance information based on the evacuation path optimization data to obtain fire evacuation guidance information release data.

[0021] The present invention can effectively eliminate areas with safety hazards during the evacuation process by accurately calibrating inaccessible paths caused by high-temperature damage, ensuring the safety and reliability of the evacuation path. Evacuation time prediction based on path congestion data can dynamically reflect the decline in traffic efficiency caused by dense crowds, helping to identify bottlenecks and congested areas in advance. Detour path analysis can rationally plan alternative paths, avoid personnel detention caused by congested areas, and improve evacuation efficiency. By integrating the information of inaccessible paths and detour paths, the shortest evacuation path is optimized to ensure that the evacuation route is both safe and efficient, minimize evacuation time, and reduce personnel risks. Finally, fire evacuation guidance information is issued based on the optimized evacuation path to achieve real-time scientific guidance of personnel, and improve the intelligence level and practical effect of evacuation response.

[0022] Preferably, this specification also provides a fire evacuation guidance system for executing the fire evacuation guidance method described above, the fire evacuation guidance system comprising: The path congestion assessment module is used to obtain building layout information; perform evacuation path network modeling based on the building layout information to obtain evacuation channel data; perform path congestion assessment based on the evacuation channel data to obtain path congestion data; The fire hazard zone identification module is used to obtain fire sensor data; based on the fire sensor data, it can sense the fire location in real time to obtain fire location data; based on the fire location data, it can predict the fire spread direction to obtain fire spread direction data; and based on the fire spread direction, it can identify the fire hazard zone; The building structure heat conduction simulation module is used to simulate the heat conduction of the building structure according to the fire hazard area to obtain the heat conduction data of the building structure; identify the unstable structure based on the heat conduction data of the building structure to obtain the unstable structure data; and predict high-temperature damage based on the unstable structure data to obtain the high-temperature damage data of the building structure; The fire evacuation guidance information release module is used to optimize the evacuation path based on the high-temperature damage data of the building structure and the path congestion data to obtain the evacuation path optimization data; based on the evacuation path optimization data, the fire evacuation guidance information is released to obtain the fire evacuation guidance information release data.

[0023] A fire evacuation guidance system of the present invention can implement any fire evacuation guidance method of the present invention, and is used to combine the operation and signal transmission medium between various modules to complete the fire evacuation guidance method. The internal modules of the system cooperate with each other to effectively improve the safety and efficiency of fire evacuation. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments thereof made with reference to the following drawings: Figure 1 This is a schematic flow chart of the steps of a fire evacuation guidance method of the present invention; Figure 2 Detailed step flow diagram of step S1 in the present invention; Figure 3 Detailed step flow diagram of step S15 in the present invention; The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0025] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the embodiments described are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative work are within the scope of protection of the present invention.

[0026] In addition, the accompanying drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor and / or microcontroller approaches.

[0027] It should be understood that although the terms "first," "second," and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used solely to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the exemplary embodiments. The term "and / or" as used herein includes any and all combinations of one or more of the listed associated items.

[0028] To achieve this, please refer to Figures 1 to 3 The present invention provides a fire evacuation guidance method, which includes the following steps: Step S1: Acquire building layout information; perform evacuation path network modeling based on the building layout information to obtain evacuation channel data; perform path congestion assessment based on the evacuation channel data to obtain path congestion data; In this embodiment, building layout information is acquired using architectural blueprints, Building Information Modeling (BIM) data, or laser-scanned 3D point cloud data. This information includes key spatial information such as the building's floor plan, floor heights, door and window locations, corridor widths, and stair locations. By converting this building layout information into a vertex and edge structure in a two-dimensional or three-dimensional spatial coordinate system, an internal evacuation path network is constructed. Specifically, each key node, such as a room, corridor, and exit, is mapped as a graph node, and the channels connecting these nodes are mapped as edges. Edge weights are weighted based on channel length and width. Channel widths are based on actual measurements; if unavailable, the minimum channel width of 1.2 meters specified in building codes is used as a default. Edge weights are calculated using the formula: weight = channel length / channel width, reflecting channel capacity. Next, based on the evacuation path network, a graph-based congestion assessment algorithm is used to calculate the capacity and current occupant density of each edge. Crowd density data can be collected through infrared thermal imaging sensors or video surveillance systems deployed within buildings. These data provide real-time statistics on the number of people per unit area. A density threshold of 1.5 people per square meter is set, exceeding which is considered congestion. This data is used to calculate a congestion index for each route, defined as the ratio of the current occupant density to the maximum load density. Finally, a route congestion dataset is constructed based on the congestion index to provide a basis for subsequent evacuation route optimization.

[0029] Step S2: Obtain fire sensor data; perceive the fire location in real time based on the fire sensor data to obtain fire location data; predict the fire spread direction based on the fire location data to obtain fire spread direction data; identify the fire hazard area based on the fire spread direction; In this embodiment, fire sensor data is acquired through multiple smoke and temperature sensors installed at key locations within the building. The smoke sensors utilize light scattering sensors with a response time of less than 5 seconds and adjustable sensitivity. The sensitivity threshold is set to trigger an alarm when the smoke concentration reaches 35% / m. The temperature sensors utilize thermocouples or resistance temperature detectors (RTDs), with a measurement range of -20°C to 600°C and an accuracy of ±1°C. Fire sensor data is aggregated by a data acquisition system and transmitted to a central processing unit. Based on the collected multi-point data, a spatial interpolation algorithm (such as Kriging) is used to construct a spatial distribution map of the fire location within the building. The fire location data represents the center of the fire source and its spread in coordinate format. A physics-based fire spread prediction algorithm then calculates the fire spread direction vector using the fire source location, current temperature, smoke concentration, and building material combustion characteristics (such as burning velocity and heat release rate). This vector is output as an angle and velocity (range: 0° to 360°) with a velocity in meters per minute. Based on the fire spread direction vector, combined with the building wind direction and ventilation system parameters (such as wind speed and wind direction angle), the fire danger area is determined. The area is represented by polygonal coordinate points, and the boundary is determined by the fire spread coverage.

[0030] Step S3: Performing a heat conduction simulation of the building structure according to the fire hazard zone to obtain heat conduction data of the building structure; identifying unstable structures based on the heat conduction data of the building structure to obtain unstable structure data; performing a high-temperature damage prediction based on the unstable structure data to obtain high-temperature damage data of the building structure; In this embodiment, based on the fire hazard area identified in step S2, thermophysical parameters of the building structure's materials, including thermal conductivity, specific heat capacity, and density, are collected. Using thermal conduction finite element analysis software, a three-dimensional mesh model of the building structure and thermal boundary conditions are input to simulate the heat conduction process within the fire zone. Thermal boundary conditions include flame temperature (set to a maximum of 1200°C), ambient temperature (normal temperature 25°C), and convective heat transfer coefficient (generally 10 W / (m²·K)). The calculation time step is set to 1 second, and the total simulation duration is determined based on the duration of the fire. Temperature field distribution data is calculated for each node within the building structure. Based on this temperature field data, pre-set structural instability criteria are applied. For example, a steel beam temperature exceeding 600°C is considered loose, and a concrete column temperature exceeding 400°C accompanied by crack propagation is considered brittle. Combined with structural stress and strain monitoring data, unstable structural areas are identified. Instability data is stored as location coordinates and instability type codes. Subsequently, based on the unstable structural area, a high-temperature damage prediction model was used to calculate the scope and extent of high-temperature damage, taking into account the degradation law of the material's high-temperature mechanical properties and load changes. The degree of damage was expressed as a percentage, with 100% indicating complete damage.

[0031] Step S4: Optimize the evacuation path based on the building structure high temperature damage data and the path congestion data to obtain evacuation path optimization data; publish fire evacuation guidance information based on the evacuation path optimization data to obtain fire evacuation guidance information release data.

[0032] In this embodiment, data on high-temperature damage to building structures and path congestion data are integrated to establish a multi-factor comprehensive evacuation path optimization model. A weighted graph search algorithm is used to eliminate inaccessible path nodes and edges damaged by high temperatures from the evacuation path network. Edge weights are dynamically adjusted based on the congestion index weight. The edge weight calculation formula is adjusted to: Weight = (Channel Length / Channel Width) × Congestion Factor × High-Temperature Factor. The congestion factor is calculated based on real-time occupant density data and ranges from 1 to 5. The high-temperature factor is set based on the degree of structural damage, with the weight of severely damaged areas increased to 10 and the weight of blocked paths set to infinity. Through this dynamic weight adjustment, the shortest and safest evacuation path is calculated. Finally, the optimized evacuation path data is disseminated through various terminals, including the building's internal broadcast system, smartphone push notifications, and fire emergency signs. This guidance information includes detailed information such as a real-time evacuation route map, path safety level, and estimated evacuation time, ensuring that evacuees receive accurate and timely route guidance.

[0033] Preferably, step S1 is specifically as follows: Step S11: Obtain building layout information; In this embodiment, building layout information is acquired using architectural design blueprints, Building Information Modeling (BIM) data, or 3D laser scanning technology. This information includes floor plans of each floor, wall locations, door and window sizes and locations, corridor widths, stairwells, and exit locations. High-precision laser scanning equipment is used to collect point cloud data from the building interior, with a point cloud density of no less than 1,000 points per square meter to ensure accurate capture of spatial details. The point cloud data is filtered to remove noise and converted into a geometric model of the building space. The geometric boundaries of each spatial unit within the building are extracted to form basic data for the spatial topology. The spatial coordinates of the building layout information are represented using a 3D Cartesian coordinate system with millimeter-level accuracy. Based on the boundaries and connectivity of each spatial unit, the specific dimensions of connected elements such as doors, windows, and passageways are annotated. For example, door widths must be no less than 0.9 meters, and corridor widths must be a minimum of 1.2 meters, as per building regulations. Once completed, the building layout information is stored in a structured data format for subsequent analysis.

[0034] Step S12: performing spatial connectivity analysis based on the building layout information to obtain spatial connectivity data; In this embodiment, spatial connectivity analysis is performed based on the building space geometric model and connected element information obtained in step S11. Using the spatial adjacency matrix method from graph theory, each spatial unit in the building (such as a room, corridor, or stairwell) is represented as a node in the graph, and spatial connected elements (doors, passages, etc.) are represented as edges between nodes. The existence of an edge requires a physically accessible connection between two spatial units, and the width of this connection is no less than a preset minimum passable width of 1.2 meters. An adjacency matrix A is used, where A[i, j] = 1 indicates connectivity between spatial units i and j, and A[i, j] = 0 indicates no connectivity. Connectivity is determined based on the size of the connected elements and the presence of obstacles. If a door or passage is blocked, connectivity is set to 0. The adjacency matrix is ​​traversed to generate spatial connectivity data. The data format is a list of nodes and their adjacent nodes, including the width, length, and passable status of the connected paths. The dimensional parameters of all connected paths are recorded in millimeters, and the data structure facilitates subsequent path search and feasibility analysis.

[0035] Step S13: identifying a traversable path based on the spatial connectivity relationship data to obtain traversable path data; In this embodiment, based on the spatial connectivity relationship data generated in step S12, a depth-first search (DFS) or breadth-first search (BFS) algorithm is used to identify all traversable paths from the starting point to the exit. The determination of traversable paths is based on node connectivity and the traffic status of all edges on the path. The width of each edge in the path is not less than 1.2 meters and there are no obstacles blocking it. The path length is calculated by accumulating the weights of each edge. The edge weight uses the actual length of the channel in meters. During the path search process, channels that are blocked, damaged, or in an abnormal state are screened out. All qualified paths are stored as a path set. Each path contains parameters such as the path node sequence, path length, minimum channel width, and traffic status. The path data format is in list form, which is convenient for quick query and subsequent evacuation optimization.

[0036] Step S14: Modeling an evacuation path network based on the traversable path data to obtain evacuation channel data; In this embodiment, a building evacuation path network model is constructed based on the traversable path data identified in step S13. This network uses nodes to represent key points in the building space (room entrances, corridor intersections, exit doors, etc.), and edges to represent traversable paths. Edge attributes include path length (meters), width (meters), and traffic capacity (based on channel width and maximum occupancy density per unit area, with a maximum density of 1.5 people / square meter). The path network model is represented by a weighted undirected graph. Edge weights are calculated based on path length and width. The weight formula is weight = path length / traffic capacity, reflecting the evacuation difficulty of the path. The path network data structure is stored in the form of an adjacency list, containing node coordinates, edge connection information, and attributes. All path data is managed hierarchically by floor to ensure complete path connectivity information in multi-story buildings. Path network data is exported to a standard format (such as GraphML) to facilitate subsequent congestion and path optimization analysis.

[0037] Step S15: performing path congestion assessment based on the evacuation channel data to obtain path congestion data.

[0038] In this embodiment, real-time personnel distribution data in the building is used. This data is provided by an infrared thermal imaging sensor or a video surveillance system, with a resolution of not less than 1 meter and an update frequency of not less than 1 time per second. A people counting algorithm is used to convert the image into the population density on each path segment in the space, with the unit being people per square meter. Combined with the traffic capacity of each edge in the path network (channel width × maximum carrying density of 1.5 people per square meter), the path congestion index is calculated, and the congestion index = actual population density / maximum carrying density. Path segments with a congestion index greater than 1 are considered congested, and the larger the congestion index, the more severe the congestion. The congestion index is attached to the path network edge attributes to generate path congestion data. Congestion data is recorded using timestamps and supports dynamic monitoring. All data is stored in a database, which supports query and real-time update, providing a basis for evacuation path optimization.

[0039] Preferably, step S15 is specifically as follows: Step S151: Calculate the channel width according to the evacuation channel data to obtain channel width data; In this embodiment, based on the geometric data of the building evacuation passage, a two-dimensional space measurement technology is used to calculate the width of each evacuation passage. During the measurement process, the boundary coordinate points of the passage are first extracted, and the coordinates of the boundary points are accurately obtained using millimeter-level three-dimensional laser scanning data or high-definition architectural design drawings. The channel width is defined as the shortest vertical distance between the two side boundaries. The measurement algorithm takes the boundary point set as input, uses the Euclidean distance to calculate the minimum distance point by point, and finally outputs the minimum width of the channel. The channel width parameter unit is millimeter, and the measurement accuracy is controlled at ±5 mm. If the channel is a curved path, the width of each local section is calculated in sections, and the minimum section width is used as the width value of the channel. All width data is stored in a two-dimensional table format, including the channel ID, the coordinates of the start and end nodes, and the corresponding width values. This step requires that the minimum channel width is not less than 0.9 meters specified in the building fire protection code. Channels below this value must be marked as potential narrow areas.

[0040] Step S152: identifying narrow channels based on the channel width data to obtain narrow channel data; In this embodiment, the channel width data obtained in step S151 is used to identify narrow channels, and a threshold screening method is used. A narrow channel is defined as a channel section with a channel width of less than 1.2 meters, and is restricted according to the minimum width standard for personnel evacuation channels in the "Code for Fire Protection Design of Buildings". Each channel width value is compared one by one, and channel sections with a width of <1200 mm are marked as narrow channels. Narrow channel data is recorded in the form of channel ID, start and end nodes, width value and position interval. For curved channels, all continuous intervals with a width of less than 1.2 meters are identified and divided into multiple narrow segments. The data storage format is a structured array to facilitate subsequent analysis and call. This step ensures that all potential congestion risk areas are accurately identified.

[0041] Step S153: performing corner area analysis based on the narrow channel data to obtain corner area data; In this embodiment, the narrow channel data identified in step S152 is used to perform corner area analysis on channel intersections and corner positions. The corner point position is determined by a spatial geometry calculation method, and the corner angle is calculated using the angle between the center lines of two adjacent channels, in degrees. A corner is defined as a node with an angle less than 135 degrees, and is set according to the design standards for fire evacuation channels. The corner area is centered on the center point, and the space within a range of 0.5 meters is extended in all directions as the corner influence area. Laser point clouds or two-dimensional plane coordinates are used to draw the boundaries of the corner area. The corner area data includes corner point coordinates, angle values, influence radius, and adjacent channel IDs. Combined with the channel width data, the channel width change trend at the corner is analyzed, and the flow bottleneck at the corner is marked. The data structure uses JSON format and contains corner attributes and spatial range.

[0042] Step S154: performing obstacle area recognition based on the corner area data to obtain obstacle area data; In this embodiment, based on the spatial range of the corner area, a high-definition indoor video surveillance system and an infrared depth sensor are used to collect data and identify obstacles in the corner area. Obstacle types include fixed furniture, temporary objects, and damaged structures. The video surveillance uses a frame rate of 30 frames per second, an image resolution of 1920×1080, and the infrared depth sensor resolution reaches 1 cm. Combined with image semantic segmentation technology, pixel-level recognition is performed on the images in the corner area, and the obstacle boundaries are marked. The spatial coordinates of the obstacle are calculated using multi-sensor fusion technology, with an error of no more than ±2 cm. The obstacle area data includes obstacle ID, location coordinates, occupied area, and height information, which are stored as point clouds and polygonal boundary descriptions. The specific size parameters of the obstacle are used for subsequent path capacity assessment.

[0043] Step S155: performing path congestion assessment based on the obstacle data and the corner area data to obtain path congestion data.

[0044] In this embodiment, congestion assessment is performed on relevant path segments within the evacuation route network by combining corner area data with obstacle area data. The assessment process is based on the path capacity calculation method. The capacity is calculated using the formula C = W × D, where W is the effective width of the channel (in meters) and D is the maximum allowable occupant density per unit width, set at 1.5 people / square meter. The effective width is calculated by subtracting the width occupied by the obstacle from the channel width. The reduction in effective width due to the obstacle is calculated using two-dimensional spatial projection. Obstacle width data is provided in step S154. The path congestion index is defined as the actual occupant density divided by the capacity. The occupant density is obtained using an on-site infrared counter or video-based people counting system with a count frequency of 1 per second and an accuracy error of no more than ±5%. The congestion index data is stored in association with the path ID and timestamp. A congestion index greater than 1 indicates severe congestion, while a value below 1 indicates unobstructed traffic. The congestion assessment results are used to dynamically adjust evacuation routes to ensure efficient personnel flow. All calculations utilize standard spatial geometry and crowd density theory to ensure accurate and real-time assessments.

[0045] Preferably, step S2 is specifically as follows: Step S21: Acquire fire sensor data and perform smoke sensor feature extraction to obtain smoke sensor data; calculate real-time smoke concentration based on the smoke sensor data to obtain smoke concentration data; construct a smoke concentration gradient field based on the smoke concentration data; and identify high smoke concentration areas based on the smoke concentration gradient field. In this example, raw signal data is first collected from multiple smoke sensors installed within the building. The sensors are photoelectric smoke detectors with a sampling frequency of 1 Hz and a resolution of 0.1 ppm smoke concentration. The collected electrical signals are filtered using a bandpass filter to remove noise components with frequencies below 0.1 Hz and above 10 Hz, improving signal quality. A feature extraction algorithm is used to extract smoke concentration-related features, including instantaneous concentration value, short-term mean (with a 30-second time window), and concentration change rate. The real-time smoke concentration is converted from the sensor voltage signal to actual concentration value in ppm using a calibration curve, with a calibration error of ±2%. A two-dimensional smoke concentration gradient field is constructed based on the multi-point concentration data. The gradient is calculated using the finite difference method, calculating the concentration difference between adjacent sensors divided by the distance, in units of ppm / m. A concentration gradient threshold of 5 ppm / m is set. Areas exceeding this threshold are classified as high smoke concentration areas. Area boundaries are determined by gradient continuity, with coordinates expressed in the building plane coordinate system. This process is repeated every 10 seconds, with an update frequency of every 10 seconds.

[0046] Step S22: Identify the building orientation angle according to the building layout information to obtain building orientation angle data; obtain wind direction data; perform wind corridor detection on the building orientation angle data according to the wind direction data to obtain wind corridor data; In this embodiment, the coordinate points (x1, y1) (x2, y2) at both ends of the building opening boundary are extracted according to the building layout drawing to calculate the building orientation angle. It is defined as the angle between the boundary and the north direction (positive direction of the Y axis), using the formula: ; in It is a two-parameter inverse tangent function used to solve the angle between the vector from point (x1, y1) to point (x2, y2) and the positive direction of the X-axis. The result is multiplied by 180 / π to convert radians to degrees, and the angle is normalized to 0 to 360 degrees modulo 360. When measuring the orientation angle on the spot, a laser scanner or total station is used to locate the two ends of the main entrance of the building, with an accuracy of ±0.5 degrees. Then, the wind direction data is obtained through the meteorological sensor. ,The update frequency is once per minute, and the wind direction angle is defined as the angle of the wind source direction, in degrees. and By comparison, the angle is calculated as: ); when When the wind direction is less than 30, it is determined that the current wind direction is consistent with the direction of the building opening, forming a wind corridor. At this time, the wind corridor data is recorded, including the wind direction angle , wind speed Vw in m / s), building opening coordinates (x1, y1), (x2, y2), opening width W= , detection timestamp t, wind corridor validity (Boolean value 1 indicates valid, 0 indicates invalid), and all parameters are recorded in a two-dimensional coordinate system for subsequent airflow diffusion trend analysis and fire spread direction identification.

[0047] Step S23: performing maximum wind speed statistics based on the wind corridor data to obtain maximum wind speed data; performing airflow diffusion trend prediction for the high smoke concentration area based on the maximum wind speed data to obtain airflow diffusion trend data; In this embodiment, historical and real-time wind speed records are extracted from wind corridor data. The data is sourced from ultrasonic wind speed sensors installed around the building, with a sampling frequency of 1 Hz and a measurement accuracy of ±0.1 m / s. Based on the collected wind speed data, the maximum wind speed value within the previous 30 minutes is calculated as the maximum wind speed data, expressed in meters per second. This maximum wind speed value, combined with the building's wind corridor geometric parameters (channel width and opening area), is used to calculate the airflow diffusion trend. A simplified transmission formula based on wind speed and smoke diffusion coefficient is used: diffusion distance = wind speed × diffusion time × diffusion coefficient, with the diffusion coefficient taken as 0.1 m² / s. The default diffusion time is set to 5 minutes. The calculated airflow diffusion trend data represents the airflow direction and range in vector form, with the vector coordinates relative to the building's plane coordinate system.

[0048] Step S24: Detecting the fire location based on the airflow diffusion trend to obtain fire location data; In this embodiment, the fire location is determined through spatial overlay analysis, combining the high smoke concentration area constructed in step S21 and the airflow diffusion trend vector calculated in step S23. The specific implementation involves drawing the boundaries of the smoke concentration hotspot on the building plan and extending them in the direction of airflow diffusion. The extension length is calculated based on the maximum wind speed and diffusion time. The fire location is the overlap between the point of maximum smoke concentration and its extension in the direction of airflow diffusion, with coordinate accuracy controlled to ±1 meter. The fire location data format is a set of two-dimensional coordinate points with a timestamp and concentration value.

[0049] Step S25: Predict the fire spread direction based on the fire location data to obtain fire spread direction data; and identify the fire hazard area based on the fire spread direction.

[0050] In this embodiment, spatial vector analysis is used to calculate the fire spread direction based on the fire location determined in step S24. The fire spread direction is determined by combining the airflow diffusion trend vector and the smoke concentration gradient direction, using a vector weighted average method, with weights of 70% of the airflow velocity and 30% of the concentration gradient magnitude. The spread direction is expressed as an angle, relative to true north in the building plane coordinate system, ranging from 0° to 360° with an accuracy of ±2 degrees. Based on the spread direction, a certain distance (presumably 10 meters) is extended along that direction to identify the fire danger zone, which represents the potential range of fire spread. The danger zone is stored as polygonal coordinates, with its boundaries consisting of the centerline of the spread direction and a 5-meter extension to its left and right.

[0051] Preferably, step S25 is specifically as follows: Step S251: Calculating fire source intensity based on fire location data to obtain fire source intensity data; In this embodiment, based on the fire location data, we first acquire multi-source sensor data, such as temperature and smoke density, around that location. Sensor types include thermocouples and photoelectric smoke detectors. The temperature measurement range is 0°C to 1200°C with an accuracy of ±2°C, and the smoke density resolution is 0.1ppm. The fire source intensity is calculated using this sensor data, specifically using the radiant heat flux calculation method, using the formula Q = σ × ε × T 4 , where Q is the radiative heat flux (W / m²) and σ is the Stefan-Boltzmann constant ε is the surface emissivity, set to 0.85, and T is the fire source temperature (Kelvin), which is converted from thermocouple data. The calculated fire source intensity data is in W / m² and is positioned in conjunction with spatial coordinates with an accuracy of ±1 meter.

[0052] Step S252: performing door and window structure detection based on the fire location data to obtain door and window structure data; In this example, architectural design drawings and on-site laser scanning data were used to inspect door and window structures near the fire location. A laser scanner captured three-dimensional point cloud data of the doors and windows, achieving a resolution of 1mm. Based on this point cloud data, geometric feature extraction methods were used to identify the size, shape, and material of the doors and windows. The extracted dimensional parameters included width, height, and thickness, with a measurement accuracy of ±2mm. The door and window structural data was stored as a set of structural geometric parameters, including door and window type (e.g., sliding door, swing door), material (steel, aluminum, wood), and surface treatment information. This data was then linked to the spatial location of the fire through coordinate mapping.

[0053] Step S253: performing hole analysis based on the door and window structure data to obtain door and window hole data; In this embodiment, hole analysis is performed based on the door and window structure data. A hole is defined as a gap or damaged area in the door and window components that is larger than 10 mm. Abnormal areas on the door and window surfaces are identified by performing difference analysis on the laser scanning point cloud. Using image processing technology, the point cloud projection image of the door and window area is binarized, and the threshold is set to an area with a point cloud density below 0.6 points / mm², which is considered a hole. The geometric parameters of the hole are extracted, including area, position, and shape. The minimum area threshold is 10 cm², and the shape is determined by a boundary detection algorithm. The hole data is stored in the form of polygon vertex coordinates and aperture size.

[0054] Step S254: performing heat flux penetration analysis on the door and window hole data based on the fire source intensity data to obtain heat flux penetration data; identifying the fire spread direction based on the heat flux penetration data to obtain fire spread direction data; In this example, heat flux penetration is calculated using a coupled heat conduction and convection analysis method based on fire source intensity data and door and window aperture data. The heat transfer equation Q = k × A × ΔT / d is used, where Q is the penetrating heat flux (W), k is the material thermal conductivity (approximately 50 W / m·K for steel and 0.13 W / m·K for wood), A is the aperture area, ΔT is the difference between the fire source temperature and the indoor temperature (using 50°C as the standard indoor temperature), and d is the door and window thickness (in meters). Using these parameters, the heat flux penetration intensity at each aperture is calculated to obtain heat flux penetration data. Based on the spatial distribution, aperture areas with heat flux exceeding 1000 W / m² are marked as areas of concern. The direction of fire spread is identified based on the spatial gradient and direction of the heat flux penetration data. The spread direction is represented as a two-dimensional vector pointing toward the area of ​​maximum heat flux penetration.

[0055] Step S255: predicting the fire spread path based on the fire spread direction data to obtain fire spread path data; In this embodiment, the path extension method is used to predict the fire spread path based on the fire spread direction data. Starting from the fire source location, the path points of the fire spread in the next 5 minutes are calculated in sequence along the spread direction vector with a step length of 0.5 meters. The path diffusion rate is calculated based on the fire source intensity and heat flux data, with the average diffusion speed set to 0.2m / s and the step time interval set to 1 second. The path point coordinate set forms the fire spread path data. By performing a buffer analysis on the spatial distribution of the path points, a buffer radius of 1 meter is set to form a diffusion area. The diffusion path data is stored in the form of timestamps and spatial coordinates.

[0056] Step S256: Identify fire hazard areas based on the fire spread path data.

[0057] In this example, a spatial overlay analysis is performed based on fire spread path data and building floor plans to identify fire danger zones. The danger zone is defined as a buffer zone along the fire spread path, with a radius of 1.5 meters to cover the potential impact area of ​​fire spread. The danger zone is represented by a polygonal coordinate set, with its boundaries formed by connecting path points. This zone data is combined with building structural information to exclude non-combustible or obstructed areas. Danger zone data is updated in real time, with an interval of 30 seconds, to ensure responsiveness to dynamic changes in the fire situation.

[0058] Preferably, step S3 is specifically as follows: Step S31: Performing a building structure heat conduction simulation based on the fire hazard zone to obtain building structure heat conduction data; In this embodiment, the thermal physical properties of the building structure are obtained, such as the thermal conductivity of concrete is 1.5W / (m·K), the thermal conductivity of steel is 50W / (m·K), the density and specific heat capacity are 2400kg / m³ and 880J / (kg·K), respectively. According to the three-dimensional geometric dimensions of the building structure, a heat conduction discrete grid is established, and the grid unit size is set to 0.1 meters to ensure a balance between simulation accuracy and computational efficiency. Boundary conditions are applied to the boundaries of the fire hazard area, the temperature of the fire area is set to 800°C, and the ambient temperature is set to 20°C. The steady-state heat conduction equation is used for numerical solution, the time step is set to 1 second, and the simulation time is set to 300 seconds based on the actual fire situation. The output building structure heat conduction data includes the temperature distribution and heat flux density of each grid unit, and the data accuracy is controlled at ±2°C.

[0059] Step S32: performing structural beam loosening detection based on the building structure heat conduction data to obtain structural beam loosening data; In this embodiment, the temperature change of the structural beam area is analyzed in detail based on the heat conduction data of the building structure. By extracting the temperature time series curve of the structural beam, it is determined whether its temperature exceeds the critical loosening temperature of the material. The critical temperature is determined according to the structural design specifications. The critical loosening temperature of the concrete beam is 300°C, and that of the steel beam is 500°C. The temperature gradient change rate threshold of 0.5°C / s is used to detect the presence of a rapid temperature rise area in the beam body. Combined with the local thermal stress calculation of the beam body, the principle of thermal expansion and contraction is used to infer the loosening risk of the structural beam. The coordinates of the loosening area and the degree of loosening (the loosening coefficient ranges from 0 to 1, and more than 0.7 is defined as severe loosening) are recorded as structural beam loosening data.

[0060] Step S33: performing structural column embrittlement detection based on the building structure heat conduction data to obtain structural column embrittlement data; In this embodiment, based on the heat conduction data of the building structure, the temperature field of the structural column area is extracted, and the temperature distribution and gradient are analyzed. The embrittlement of the structural column is determined based on the concrete embrittlement temperature threshold, which is generally set at 450°C. By measuring the surface and internal temperature of the column, the area with a high temperature exceeding 450°C is identified, and the internal cement matrix water loss rate is calculated in combination with the temperature hysteresis effect of the column. The cement water loss rate threshold is set to 10%, and areas exceeding this threshold are identified as brittle areas. Combined with the temperature field data, the empirical curve of the change of elastic modulus with temperature is used to calculate the percentage of local elastic modulus decrease. If the decrease exceeds 40%, the area is confirmed to be a brittle area. The brittleness data includes specific spatial coordinates and brittleness level (mild, moderate, severe).

[0061] Step S34: Integrate the structural beam loosening data and the structural column embrittlement data to obtain unstable structural data; In this embodiment, the structural beam loosening data obtained in step S32 and the structural column brittleness data obtained in step S33 are spatially fused. The two types of data are mapped using a three-dimensional spatial coordinate system, and spatial superposition analysis techniques are used to identify the intersecting damaged areas of beams and columns. The unstable structural region is defined as the overlapping area where the beam loosening coefficient is greater than 0.7 and the column brittleness level is moderate or above. This regional data includes a set of spatial coordinates and a comprehensive instability index. The instability index comprehensively considers the loosening coefficient and brittleness level and is calculated using a weighted method with weights of 0.6 and 0.4, respectively. The resulting unstable structural data is used for subsequent risk assessment.

[0062] Step S35: Perform high-temperature damage prediction based on the unstable structure data to obtain high-temperature damage data of the building structure.

[0063] In this embodiment, high-temperature damage prediction is performed for each unstable area based on the unstable structure data. The damage progress is calculated based on the high-temperature mechanical property attenuation curve of the structural material, combined with the current temperature and time exposure parameters. The damage prediction adopts the cumulative damage theory and defines the damage factor D=∫(t0 to t)(f(T(t))dt), where f(T) is the temperature-dependent damage rate function. The temperature threshold is set to 300°C, and the damage rate increases significantly above this temperature. The damage factor is calculated by time integration, and when D≥1 is set, it is considered as structural damage. The damage factor is dynamically updated in combination with the time step of the heat conduction simulation. The final output is the high-temperature damage data of the building structure, including the spatial coordinates of the damaged area, the degree of damage (D value) and the expected damage time. The data accuracy is controlled at 1 second time series resolution and 0.1 meter spatial resolution.

[0064] Preferably, step S32 is specifically as follows: Step S321: identifying high-temperature areas of structural beams based on the building structure heat conduction data to obtain high-temperature area data of the structural beams; In this example, 3D temperature data of the structural beams was extracted. A temperature threshold of 450°C was used as the high-temperature threshold, marking areas above 450°C as high-temperature zones. A segmentation algorithm was used to cluster continuous high-temperature point clouds, generating independent spatial volumetric data for these high-temperature zones. Clustering was performed using the density-based clustering algorithm (DBSCAN), with clustering parameters set to a neighborhood radius of ε = 0.2 meters and a minimum number of neighbor points (MinPts) of 5 to ensure spatial continuity and physical consistency within the high-temperature zones. The output high-temperature zone data for the structural beams includes the 3D coordinate range and volume of each high-temperature zone.

[0065] Step S322: performing interface gap detection based on the high temperature area data of the structural beam to obtain interface gap data; In this embodiment, ultrasonic detection technology is used, and ultrasonic sensors with a frequency of 50kHz are used to arrange sensing points along the longitudinal and cross-sectional surfaces of the beam. The echo signal received by each sensing point is analyzed in the time domain and spectrum to extract the interface reflection wave characteristics. The interface gap thickness is calculated by comparing the normal interface reflection time with the actual detection time. The threshold is set to 2mm, and a gap thickness greater than this value is defined as the presence of a gap. The spatial interpolation method uses Kriging interpolation to reconstruct the three-dimensional gap thickness field of the detection point data to form interface gap data, including gap location and thickness distribution.

[0066] Step S323: performing a reinforcement-concrete bond slip simulation based on the interface void data to obtain reinforcement-concrete bond slip data; determining the degree of firmness based on the reinforcement-concrete bond slip data to obtain degree of firmness data; In this embodiment, the finite element method is used to establish the bonding interface model between steel bars and concrete. The thickness of the interface element is set based on the gap thickness data of step S322, and the unit size is 0.01 meters. The bonding-slip model adopts a bilinear model based on the bonding-slip constitutive relationship, in which the initial bonding strength τ_max is 5MPa, the slip strain range is 0-0.1mm, and the residual strength τ_res is 1MPa. Loading boundary conditions are applied to simulate the relative slip displacement of steel bars and concrete under shear load. The bond slip data of steel bars and concrete is obtained by calculating the bond stress distribution and relative slip. Based on the slip threshold of 0.05mm, the degree of bond strength of steel bars and concrete is determined. A slip less than the threshold is defined as firm, and a slip exceeding the threshold is defined as loose.

[0067] Step S324: performing crack propagation analysis based on the firmness data to obtain crack propagation data; In this example, a mechanical crack propagation criterion is used to calculate the stress intensity factor (K_I) at the crack tip based on linear elastic fracture mechanics. Input parameters include the initial crack length (converted based on the maximum thickness of the interfacial void; the initial crack length is generally 10 times the void thickness), the material elastic modulus E, which is taken as 30 GPa, and the fracture toughness K_IC, which is taken as 1.2 MPa·m^0.5. A thermal stress field caused by a temperature gradient in the beam is applied to calculate the stress intensity factor at the crack tip. The Paris law is used to predict the crack propagation rate: da / dN = C(ΔK)^m, where the constant C is 1×10^-10, m is 3, and ΔK is the stress intensity factor range. Crack propagation length and direction are obtained through numerical integration, and crack propagation data is output, including the crack length change and propagation path coordinates.

[0068] Step S325: performing crack bifurcation identification based on the crack propagation data to obtain crack bifurcation data; In this example, three-dimensional image processing technology is used to analyze the curvature of the crack path and calculate the rate of change of curvature along the crack propagation path. A curvature threshold of 0.01 mm^-1 is set, and regions exceeding this threshold are identified as crack bifurcations. Combining the geometric characteristics of the crack path and the propagation rate, the spatial coordinates and bifurcation angles of all crack bifurcations are extracted. The bifurcation data is output as a set of node coordinates and a distribution of bifurcation angles.

[0069] Step S326: Evaluate the looseness of the structural beam according to the crack bifurcation data to obtain the looseness data of the structural beam.

[0070] In this embodiment, the degree of looseness of the structural beam is comprehensively assessed based on the crack bifurcation data. A weighted scoring method is used to include crack length, number of cracks, and bifurcation angle in the looseness assessment index. The crack length weight is 0.4, the crack number weight is 0.3, and the bifurcation angle weight is 0.3. The crack length adopts the actual extended total length, the number of cracks is the number of bifurcation nodes, and the bifurcation angle calculates the average bifurcation angle. The scoring range is 0-1, and a score greater than 0.6 is determined to be severely loose. The looseness data includes a looseness scoring value and a corresponding spatial position. The looseness degree result output is used for subsequent structural safety assessment and fire evacuation path adjustment.

[0071] Preferably, step S33 is specifically as follows: Step S331: identifying high-temperature areas of structural columns based on the building structure heat conduction data to obtain high-temperature area data of the structural columns; In this embodiment, a temperature threshold determination method is used to identify high-temperature areas in structural columns based on the three-dimensional distribution of column temperatures in building structure heat conduction data. The temperature threshold is set at 500°C, and areas exceeding this temperature are defined as high-temperature areas. By extracting the column temperature field data, the columns are discretized using a three-dimensional voxel partitioning method, with a voxel side length of 0.1 meter. The temperature of each voxel unit is determined, and voxels exceeding the threshold are classified as high-temperature areas. Using a three-dimensional connected domain analysis algorithm, continuous high-temperature voxels are clustered, and the spatial location and volume parameters of each high-temperature area are output to form the high-temperature area data of the structural columns.

[0072] Step S332: performing longitudinal reinforcement thermal corrosion analysis based on the high temperature area data of the structural column to obtain longitudinal reinforcement thermal corrosion data; In this example, an electrochemical corrosion rate tester, combined with the principle of high-temperature accelerated corrosion, was used to calculate the corrosion rate of column reinforcement in a high-temperature environment using heat conduction data. The corrosion rate calculation formula is R = k*exp(-Ea / (RT)), where R is the corrosion rate (mm / year), k is an experimental constant set at 2.5 × 10^-3 mm / year, Ea is the activation energy (60 kJ / mol), R is the gas constant (8.314 J / (mol·K), and T is the average temperature (K) in the high-temperature region of the column. A calculated corrosion rate exceeding 0.05 mm / year is defined as a severe corrosion zone. The corrosion rate is mapped to spatial coordinates to generate longitudinal reinforcement hot corrosion data.

[0073] Step S333: detecting stirrup transverse cracks based on the high-temperature area data of the structural column to obtain stirrup transverse crack data; In this example, high-definition optical imaging equipment and crack detection technology were used to detect transverse cracks in structural column stirrups. The optical image resolution was maintained at no less than 0.1 mm / pixel, covering all exposed stirrup surfaces in high-temperature areas. The image was processed using an edge detection algorithm (such as the Canny operator) to extract crack outlines. A threshold method was then used to determine crack width. Cracks exceeding 0.3 mm were labeled as transverse cracks. Combined with image geometric correction technology, the spatial distribution and length of the cracks were determined to generate data on transverse stirrup cracks.

[0074] Step S334: performing concrete envelope failure determination based on stirrup transverse crack data to obtain concrete envelope failure data; In this embodiment, the spatial distribution of cracks in the column stirrup area is extracted from 3D crack data collected by sensors or detection equipment. High-resolution 3D scanning technology (such as laser scanning or structured light scanning) is used to obtain crack length and width information. For each continuous crack segment, its projected length along the stirrup direction is calculated, and the lengths of all continuous crack segments are summed to obtain the cumulative crack length. The ratio of the cumulative crack length to the corresponding stirrup length is calculated, with a threshold of 30% set. If this ratio exceeds this threshold, concrete envelope failure is determined. Crack width is measured using an image processing algorithm, with a threshold of 0.3 mm. All crack widths must exceed this threshold to be counted towards the cumulative length. The spatial locations of cracks identified as failures are mapped using 3D coordinates onto a 3D model of the building structure, generating spatial boundary data for the concrete envelope failure area. This spatial boundary data includes the volume, surface shape, and specific location coordinates of the failure area, and is output as a concrete envelope failure data file in a point cloud or triangular mesh data structure.

[0075] Step S335: identifying the longitudinal reinforcement thermal corrosion area based on the longitudinal reinforcement thermal corrosion data; identifying the concrete envelope failure area based on the concrete envelope failure data; In this embodiment, the longitudinal reinforcement thermal corrosion data is provided by corrosion sensors or material test results, the corrosion rate unit is mm / year, and the sampling interval is 0.05 meters. By spatially interpolating the sensor data, a three-dimensional distribution field of the corrosion rate is constructed. The area with a corrosion rate greater than 0.05 mm / year is defined as a high-risk corrosion area, that is, the longitudinal reinforcement thermal corrosion area. The concrete envelope failure area directly uses the spatial boundary data output by step S334. Using a spatial gridding processing method, the longitudinal reinforcement thermal corrosion area and the concrete envelope failure area are respectively mapped into a three-dimensional binary matrix with a spatial resolution of 0.1 meters. The element value in the matrix is ​​1, indicating that there is corrosion or envelope failure at that location, and 0 indicates no corrosion. The spatial coordinate system is based on the global coordinates of the building structure to ensure that the two regional data are aligned in the same reference system.

[0076] Step S336: performing an intersection operation of the structural column embrittlement area according to the longitudinal reinforcement thermal corrosion area and the concrete envelope failure area to obtain structural column embrittlement data.

[0077] In this embodiment, an intersection operation is performed on the binary spatial matrices corresponding to the longitudinal reinforcement thermal corrosion region and the concrete envelope failure region. An element-by-element logical AND operation (AND) is used. Only when the corresponding elements in both matrices are 1, the resulting matrix element is set to 1, indicating that the spatial unit belongs to both the corrosion region and the envelope failure region. The intersection matrix undergoes three-dimensional morphological processing to remove isolated noise points. Connected domain analysis is used to extract continuous embrittled regions, and their spatial location, total volume, and boundary surfaces are calculated. This boundary information is expressed as a triangular mesh or envelope box to facilitate subsequent structural safety assessments. The output structural column embrittlement data file contains a list of the spatial coordinates of the intersection region, its volume (cubic meters), vertex information of the boundary polygons, and associated attribute data, which can be used to guide post-fire structural reinforcement and evacuation route adjustments.

[0078] Preferably, step S4 is specifically as follows: Step S41: calibrating an impassable path based on the high-temperature damage data of the building structure to obtain impassable path data; In this embodiment, a path identification technique based on three-dimensional spatial gridding is used to discretize the building's interior space into voxel units with a side length of 0.2 meters. By combining the spatial coordinates and temperature threshold information (a threshold of 600°C or above is considered severe damage) from the high-temperature damage data, high-temperature damage areas within the corresponding voxel units are identified. Based on these areas, nodes and edges in the evacuation channel network are marked as impassable, generating impassable path data. Impassable paths include channel sections that have been deemed closed due to structural damage or where high-temperature areas are deemed too dangerous. This data is represented as a graph structure of nodes and edges. Nodes include spatial coordinates and path status, while edges include connection relationships and pass weights (the weight is set to infinity when impassable).

[0079] Step S42: Predicting evacuation time based on the path congestion data to obtain road evacuation time data; In this embodiment, the current pedestrian density (unit: people / m²) of each path segment in the path congestion data is used as an input parameter and combined with a standard human evacuation speed model to calculate the actual evacuation speed. The evacuation speed V (m / s) is calculated according to the following formula: V = V0 × (1-k × ρ), where V0 is the free walking speed, set to 1.2 m / s, k is the congestion coefficient, set to 0.3, and ρ is the path congestion density. The required evacuation time T = L / V is calculated for each path segment, taking into account the path segment length. By accumulating the path segment time, evacuation time data for different paths is obtained. This process is based on real-time congestion data and is updated at a frequency of 1 second to ensure the timeliness of the prediction.

[0080] Step S43: performing detour route analysis based on the road evacuation time data to obtain detour route data; In this embodiment, a weighted directed graph is constructed based on evacuation time data, with nodes representing intersections or forks within the building and edge weights representing evacuation times. The Dijkstra algorithm is used to search for paths within the graph structure, eliminating inaccessible nodes and edges, and calculating the shortest evacuation time path from the starting point to the safe exit. To account for detour paths, a maximum allowable detour time threshold is set at 1.5 times the shortest starting time. All paths with path times within this threshold are identified as valid detour paths. The output detour path data includes a sequence of path nodes and the corresponding cumulative time, formatted as a list structure.

[0081] Step S44: Optimizing the shortest evacuation route based on the impassable route data and the detour route data to obtain optimized evacuation route data; In this embodiment, inaccessible path nodes and edges are assigned infinite weights to eliminate their influence. Combining all valid paths in the detour path data, a multi-source path optimization algorithm is employed to calculate the optimal evacuation path for all occupants within the building to the exit, taking into account path length, evacuation time, and path safety factors. The heuristic function used in the path optimization process is Euclidean distance-weighted evacuation time, with a heuristic weight of 0.7. The resulting evacuation path optimization data includes the path node sequence corresponding to each starting point, the estimated evacuation time, and the path safety level (ranging from 1 to 5 based on the high-temperature or congested areas traversed by the path).

[0082] Step S45: Publish fire evacuation guidance information based on the evacuation path optimization data to obtain fire evacuation guidance information release data.

[0083] In this embodiment, information dissemination is achieved through digital display screens, intelligent lighting systems, and mobile terminal push technology. Evacuation route optimization data is converted into guidance signals. Combined with the building's lighting layout, the lighting and flashing of indicator arrow lights are dynamically controlled. Digital screens corresponding to path nodes display the current evacuation status and recommended directions in real time. Mobile terminal push information, including route maps, remaining evacuation time, and safety warnings, is transmitted in real time via the building's wireless LAN. The published data format includes path node coordinates, guidance light status (on / off / flashing), text prompt content, and an update frequency set to 1 second to ensure the real-time and accuracy of guidance information.

[0084] Preferably, this specification also provides a fire evacuation guidance system for executing the fire evacuation guidance method described above, the fire evacuation guidance system comprising: The path congestion assessment module is used to obtain building layout information; perform evacuation path network modeling based on the building layout information to obtain evacuation channel data; perform path congestion assessment based on the evacuation channel data to obtain path congestion data; The fire hazard zone identification module is used to obtain fire sensor data; based on the fire sensor data, it can sense the fire location in real time to obtain fire location data; based on the fire location data, it can predict the fire spread direction to obtain fire spread direction data; and based on the fire spread direction, it can identify the fire hazard zone; The building structure heat conduction simulation module is used to simulate the heat conduction of the building structure according to the fire hazard area to obtain the heat conduction data of the building structure; identify the unstable structure based on the heat conduction data of the building structure to obtain the unstable structure data; and predict high-temperature damage based on the unstable structure data to obtain the high-temperature damage data of the building structure; The fire evacuation guidance information release module is used to optimize the evacuation path based on the high-temperature damage data of the building structure and the path congestion data to obtain the evacuation path optimization data; based on the evacuation path optimization data, the fire evacuation guidance information is released to obtain the fire evacuation guidance information release data.

[0085] The present invention is therefore intended to be illustrative and non-restrictive in all respects, with the scope of the invention being defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the application documents are intended to be embraced therein.

[0086] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. A fire evacuation guidance method, characterized in that: The following steps are involved: Step S1: Acquire building layout information; perform evacuation path network modeling based on the building layout information to obtain evacuation channel data; Conduct path congestion assessment based on evacuation channel data to obtain path congestion data; Step S2: Obtain fire sensor data; perceive the fire location in real time based on the fire sensor data to obtain fire location data; predict the fire spread direction based on the fire location data to obtain fire spread direction data; identify the fire hazard area based on the fire spread direction; Step S3: Performing a building structure heat conduction simulation based on the fire hazard zone to obtain building structure heat conduction data; Identify unstable structures based on building structure heat conduction data to obtain unstable structure data; predict high-temperature damage based on unstable structure data to obtain high-temperature damage data for building structures; Step S4: Optimizing the evacuation path based on the building structure high-temperature damage data and the path congestion data to obtain evacuation path optimization data; Fire evacuation guidance information is released based on the evacuation path optimization data to obtain fire evacuation guidance information release data.

2. The fire evacuation guidance method according to claim 1, characterized in that: Step S1 is specifically as follows: Step S11: Obtain building layout information; Step S12: performing spatial connectivity analysis based on the building layout information to obtain spatial connectivity data; Step S13: identifying a traversable path based on the spatial connectivity relationship data to obtain traversable path data; Step S14: Modeling an evacuation path network based on the traversable path data to obtain evacuation channel data; Step S15: performing path congestion assessment based on the evacuation channel data to obtain path congestion data.

3. The fire evacuation guidance method according to claim 2, characterized in that: Step S15 is specifically as follows: Step S151: Calculate the channel width according to the evacuation channel data to obtain channel width data; Step S152: identifying narrow channels based on the channel width data to obtain narrow channel data; Step S153: performing corner area analysis based on the narrow channel data to obtain corner area data; Step S154: performing obstacle area recognition based on the corner area data to obtain obstacle area data; Step S155: performing path congestion assessment based on the obstacle data and the corner area data to obtain path congestion data.

4. The fire evacuation guidance method according to claim 1, characterized in that: Step S2 is specifically as follows: Step S21: Obtain fire sensor data and perform smoke sensor feature extraction to obtain smoke sensor data; calculate real-time smoke concentration based on the smoke sensor data to obtain smoke concentration data; Construct a smoke concentration gradient field based on smoke concentration data; identify high smoke concentration areas based on the smoke concentration gradient field; Step S22: Identify the building orientation angle according to the building layout information to obtain building orientation angle data; obtain wind direction data; perform wind corridor detection on the building orientation angle data according to the wind direction data to obtain wind corridor data; Step S23: performing maximum wind speed statistics based on the wind corridor data to obtain maximum wind speed data; performing airflow diffusion trend prediction for the high smoke concentration area based on the maximum wind speed data to obtain airflow diffusion trend data; Step S24: Detecting the fire location based on the airflow diffusion trend to obtain fire location data; Step S25: Predict the fire spread direction based on the fire location data to obtain fire spread direction data; and identify the fire hazard area based on the fire spread direction.

5. The fire evacuation guidance method according to claim 4, characterized in that: Step S25 is specifically as follows: Step S251: Calculating fire source intensity based on fire location data to obtain fire source intensity data; Step S252: performing door and window structure detection based on the fire location data to obtain door and window structure data; Step S253: performing hole analysis based on the door and window structure data to obtain door and window hole data; Step S254: performing heat flux penetration analysis on the door and window hole data according to the fire source intensity data to obtain heat flux penetration data; Identify the fire spread direction based on the heat flux penetration data and obtain the fire spread direction data; Step S255: predicting the fire spread path based on the fire spread direction data to obtain fire spread path data; Step S256: Identify fire hazard areas based on the fire spread path data.

6. The fire evacuation guidance method according to claim 1, characterized in that: Step S3 is specifically as follows: Step S31: Performing a building structure heat conduction simulation based on the fire hazard zone to obtain building structure heat conduction data; Step S32: performing structural beam loosening detection based on the building structure heat conduction data to obtain structural beam loosening data; Step S33: performing structural column embrittlement detection based on the building structure heat conduction data to obtain structural column embrittlement data; Step S34: Integrate the structural beam loosening data and the structural column embrittlement data to obtain unstable structural data; Step S35: Perform high-temperature damage prediction based on the unstable structure data to obtain high-temperature damage data of the building structure.

7. The fire evacuation guidance method according to claim 6, characterized in that: Step S32 is specifically as follows: Step S321: identifying high-temperature areas of structural beams based on the building structure heat conduction data to obtain high-temperature area data of the structural beams; Step S322: performing interface gap detection based on the high temperature area data of the structural beam to obtain interface gap data; Step S323: performing a reinforcement-concrete bond slip simulation based on the interface void data to obtain reinforcement-concrete bond slip data; Determine the firmness based on the reinforced concrete bond slip data to obtain firmness data; Step S324: performing crack propagation analysis based on the firmness data to obtain crack propagation data; Step S325: performing crack bifurcation identification based on the crack propagation data to obtain crack bifurcation data; Step S326: Evaluate the looseness of the structural beam according to the crack bifurcation data to obtain the looseness data of the structural beam.

8. The fire evacuation guidance method according to claim 6, characterized in that: Step S33 is specifically as follows: Step S331: identifying high-temperature areas of structural columns based on the building structure heat conduction data to obtain high-temperature area data of the structural columns; Step S332: performing longitudinal reinforcement thermal corrosion analysis based on the high temperature area data of the structural column to obtain longitudinal reinforcement thermal corrosion data; Step S333: detecting stirrup transverse cracks based on the high-temperature area data of the structural column to obtain stirrup transverse crack data; Step S334: performing concrete envelope failure determination based on stirrup transverse crack data to obtain concrete envelope failure data; Step S335: identifying the longitudinal reinforcement hot corrosion area according to the longitudinal reinforcement hot corrosion data; Identify concrete envelope failure areas based on concrete envelope failure data; Step S336: performing an intersection operation of the structural column embrittlement area according to the longitudinal reinforcement thermal corrosion area and the concrete envelope failure area to obtain structural column embrittlement data.

9. The fire evacuation guidance method according to claim 1, characterized in that: Step S4 is specifically as follows: Step S41: calibrating an impassable path based on the high-temperature damage data of the building structure to obtain impassable path data; Step S42: Predicting evacuation time based on the path congestion data to obtain road evacuation time data; Step S43: performing detour route analysis based on the road evacuation time data to obtain detour route data; Step S44: Optimizing the shortest evacuation route based on the impassable route data and the detour route data to obtain optimized evacuation route data; Step S45: Publish fire evacuation guidance information based on the evacuation path optimization data to obtain fire evacuation guidance information release data.

10. A fire evacuation guidance system, characterized in that: For executing the fire evacuation guidance method according to claim 1, the fire evacuation guidance system comprises: The path congestion assessment module is used to obtain building layout information; perform evacuation path network modeling based on the building layout information to obtain evacuation channel data; perform path congestion assessment based on the evacuation channel data to obtain path congestion data; The fire hazard zone identification module is used to obtain fire sensor data; based on the fire sensor data, it can sense the fire location in real time to obtain fire location data; based on the fire location data, it can predict the fire spread direction to obtain fire spread direction data; and based on the fire spread direction, it can identify the fire hazard zone; The building structure heat conduction simulation module is used to simulate the heat conduction of the building structure according to the fire hazard area to obtain the heat conduction data of the building structure; identify the unstable structure based on the heat conduction data of the building structure to obtain the unstable structure data; and predict high-temperature damage based on the unstable structure data to obtain the high-temperature damage data of the building structure; The fire evacuation guidance information release module is used to optimize the evacuation path based on the high-temperature damage data of the building structure and the path congestion data to obtain the evacuation path optimization data; based on the evacuation path optimization data, the fire evacuation guidance information is released to obtain the fire evacuation guidance information release data.

Images