CN-121997510-A - Method, device, equipment and medium for planning rescue path in building fire disaster

CN121997510ACN 121997510 ACN121997510 ACN 121997510ACN-121997510-A

Abstract

The application relates to a method, a device, equipment and a medium for planning a rescue path in a building fire disaster. The method comprises the steps of obtaining an indoor vector map of a target building, fire-fighting resource position information and multi-dimensional fire scene environment data, wherein the multi-dimensional fire scene environment data comprise at least one of fire source thermal imaging distribution data, multi-floor flue gas concentration gradient data and building structure monitoring data, generating an initial static passing network diagram based on the indoor vector map and the fire-fighting resource position information, carrying out dynamic risk quantification assessment on passing costs of all connecting sides in the initial static passing network diagram according to the multi-dimensional fire scene environment data to obtain dynamic risk weight values of all the connecting sides, generating a dynamic risk passing network diagram according to the dynamic risk weight values, and obtaining a building fire internal rescue path based on the dynamic risk passing network diagram. By adopting the method, a safe and efficient rescue path in the building fire disaster can be provided for firefighters.

Inventors

HU YANJIN
WANG CHENGYUAN
Hu Chengshun

Assignees

山东英爱安全技术服务有限公司

Dates

Publication Date: 20260508
Application Date: 20260129

Claims (10)

1. A method for planning a rescue path in a building fire, the method comprising: acquiring an indoor vector map of a target building, fire-fighting resource position information and multidimensional fire scene environment data, wherein the multidimensional fire scene environment data comprises at least one of fire source thermal imaging distribution data, multi-floor flue gas concentration gradient data and building structure monitoring data; generating an initial static traffic network map based on the indoor vector map and the fire control resource position information; Carrying out dynamic risk quantification evaluation on the passing cost of each connecting edge in the initial static passing network diagram according to the multidimensional fire scene environment data to obtain a dynamic risk weight value of each connecting edge; Generating a dynamic risk passing network diagram according to the dynamic risk weight value; And obtaining the rescue path in the building fire disaster based on the dynamic risk passing network diagram.
2. The method of claim 1, wherein the generating a dynamic risk traffic network graph from the dynamic risk weight values comprises: Obtaining an initial network diagram with a risk level label based on the dynamic risk weight value and a preset dynamic risk level mapping rule; According to the thermal imaging distribution data of the fire source and the multi-floor flue gas concentration gradient data, a predicted risk level sequence of each connecting edge on a plurality of continuous time slices in the future in the initial network diagram with the risk level label is obtained; based on the predicted risk level sequence, obtaining a multi-layer dynamic risk network diagram with a time dimension; and carrying out space-time fusion on the multi-layer dynamic risk network diagram and the initial network diagram to obtain the dynamic risk passing network diagram.
3. The method of claim 2, wherein the performing space-time fusion of the multi-layer dynamic risk network map and the initial network map to obtain the dynamic risk traffic network map comprises: According to the network topology of each time slice layer of the multi-layer dynamic risk network diagram and the topological structure of the initial network diagram, performing node space topology mapping alignment of the network topology of the time slice layer and the initial network topology to obtain a time-space aligned network sequence; based on the time-space aligned network sequence and a preset time-space fusion weight function, time-varying risk feature vectors corresponding to the connecting edges in the time-space aligned network sequence are obtained; Obtaining a three-dimensional space-time correlation matrix based on the time-varying risk feature vector, wherein a first dimension of the three-dimensional space-time correlation matrix and a second dimension index identify connecting edges in a space, a third dimension index identifies a time sequence, and matrix element values represent fused risk weights of corresponding edges at corresponding moments; And obtaining the dynamic risk passing network diagram based on the three-dimensional space-time correlation matrix.
4. The method of claim 3, wherein the obtaining a time-varying risk feature vector corresponding to each connecting edge in the spatio-temporal aligned network sequence based on the spatio-temporal aligned network sequence and a preset spatio-temporal fusion weight function includes: the time-varying risk feature vector is calculated using the following formula: Wherein, the Represents a time-varying risk feature vector that, Representing edges At the current moment Is used for the real-time risk observation of (a), Representing edges At future time Is used to predict the risk value of (a), The fused weighting coefficients representing real-time observations and short-term predictions, Represents the time sequence attenuation weight coefficient and satisfies , Representing the final instant of the predicted time window.
5. The method of claim 1, wherein the obtaining a building fire internal rescue path based on the dynamic risk pass network graph comprises: acquiring a firefighter internal attack rescue task instruction, and extracting rescue starting point coordinates and target end point coordinates from the firefighter internal attack rescue task instruction; Based on preset protection parameters and safety criteria of firefighter personal protective equipment, obtaining a multi-dimensional risk constraint condition of a rescue path plan in a building fire disaster, wherein the multi-dimensional risk constraint condition comprises an instantaneous environment risk constraint condition, an accumulated risk threshold constraint condition, an equipment tolerance adaptation constraint condition and a structural safety redundancy constraint condition; And taking the rescue starting point coordinates and the target ending point coordinates as starting and stopping points, and performing iterative search in the dynamic risk passing network diagram according to the multidimensional risk constraint condition to obtain at least one building fire internal rescue path.
6. The method according to any one of claims 1 to 5, wherein the performing, according to the multi-dimensional fire scene environment data, a dynamic risk quantification evaluation on a traffic cost of each connection edge in the initial static traffic network graph to obtain a dynamic risk weight value of each connection edge includes: Based on the thermal imaging distribution data of the fire source and the space coverage area of each connecting edge in the initial static passing network diagram, obtaining a real-time heat radiation flux intensity value in a channel area corresponding to each connecting edge; Obtaining a key toxic gas concentration predicted value of the corresponding channel area of each connecting edge in a future preset time period based on the multi-floor flue gas concentration gradient data; based on the building structure monitoring data, obtaining an instantaneous instability probability value of the associated building member; Based on the real-time heat radiation flux intensity value, the key toxic gas concentration predicted value and the instantaneous instability probability value, obtaining a comprehensive dynamic risk quantized value of each connecting edge; and obtaining the dynamic risk weight value of each connecting edge through a preset dynamic weight mapping function according to the comprehensive dynamic risk quantization value.
7. The method of claim 1, wherein the generating an initial static transit network map based on the indoor vector map and the fire resource location information comprises: based on the building structure parameters of the indoor vector map, extracting channel nodes and a communication relation to construct a basic topology network; Adding a resource attribute label for a corresponding channel node in the basic topology network based on the fire control resource position information; according to the channel nodes, static passing cost initial values are given to edges corresponding to all the communication relations in the basic topology network by combining with preset building passing priority rules; and eliminating invalid connection relations with structural barriers in the basic topology network to obtain an initial static pass network diagram.
8. An interior rescue path planning device for a building fire, the device comprising: The multi-source data acquisition module is used for acquiring an indoor vector map of a target building, fire-fighting resource position information and multi-dimensional fire scene environment data, wherein the multi-dimensional fire scene environment data comprises at least one of fire source thermal imaging distribution data, multi-floor flue gas concentration gradient data and building structure monitoring data; The initial network construction module is used for generating an initial static traffic network diagram based on the indoor vector map and the fire control resource position information; The dynamic risk quantification module is used for carrying out dynamic risk quantification assessment on the passing cost of each connecting edge in the initial static passing network diagram according to the multidimensional fire scene environment data to obtain a dynamic risk weight value of each connecting edge; the dynamic network generation module is used for generating a dynamic risk passing network diagram according to the dynamic risk weight value; And the path planning module is used for obtaining the rescue path in the building fire disaster based on the dynamic risk passing network diagram.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.

Description

Method, device, equipment and medium for planning rescue path in building fire disaster Technical Field The invention belongs to the technical field of fire rescue path planning, and particularly relates to a method, a device, equipment and a medium for planning a rescue path in a building fire. Background With the development of urban construction and building industrialization technology, the number of high-rise, super-high-rise and large-scale complex buildings is continuously increased, and the buildings have the characteristics of complex structure, dense functions and large personnel flow, so that a fire disaster response mode based on fixed fire fighting facilities and conventional emergency schemes appears. In the traditional technology, fire rescue work is carried out by presetting evacuation channel identification, static escape route planning and manual command and dispatch, and part of emergency systems can output fixed evacuation paths in combination with building plan to assist trapped people to evacuate fire scene. However, the existing fire disaster coping mode and emergency system have the core defects that firstly, path planning logic takes a fire source as a core target, is only suitable for evacuation of trapped people, cannot meet the rescue path requirements that firefighters need to actively go deep into a fire scene and go forward by facing fire, secondly, path planning data depend on a static building drawing, do not combine dynamic parameters such as fire spreading, smoke spreading, structural collapse risks and the like in real time of the fire scene, and thirdly, lack of path adaptability adjustment mechanisms aiming at extreme environments such as dense smoke, high temperature, low visibility and the like, so that a planned path is easy to fail due to sudden change of the fire scene environment, rescue time is delayed, and operation risks of firefighters are greatly increased. Disclosure of Invention Accordingly, it is necessary to provide a method for planning a rescue path in a building fire in order to solve the above-mentioned problems. In a first aspect, the present application provides a method for planning a rescue path in a building fire, including: Acquiring an indoor vector map of a target building, fire-fighting resource position information and multi-dimensional fire scene environment data, wherein the multi-dimensional fire scene environment data comprises at least one of fire source thermal imaging distribution data, multi-floor flue gas concentration gradient data and building structure monitoring data; Generating an initial static traffic network map based on the indoor vector map and the fire control resource position information; Carrying out dynamic risk quantification evaluation on the passing cost of each connecting edge in the initial static passing network diagram according to the multidimensional fire scene environment data to obtain dynamic risk weight values of each connecting edge; generating a dynamic risk passing network diagram according to the dynamic risk weight value; And obtaining the rescue path in the building fire disaster based on the dynamic risk passing network diagram. In one embodiment, generating a dynamic risk passing network graph according to the dynamic risk weight value includes: obtaining an initial network diagram with a risk level label based on the dynamic risk weight value and a preset dynamic risk level mapping rule; according to the thermal imaging distribution data of the fire source and the multi-floor flue gas concentration gradient data, a predicted risk level sequence of each connecting edge on a plurality of continuous time slices in the future in an initial network diagram with a risk level label is obtained; obtaining a multi-layer dynamic risk network diagram with a time dimension based on the predicted risk level sequence; And carrying out space-time fusion on the multi-layer dynamic risk network diagram and the initial network diagram to obtain a dynamic risk passing network diagram. In one embodiment, performing space-time fusion on the multi-layer dynamic risk network graph and the initial network graph to obtain a dynamic risk passing network graph, including: According to the network topology of each time slice layer of the multi-layer dynamic risk network diagram and the topological structure of the initial network diagram, performing node space topology mapping alignment of the network topology of the time slice layer and the initial network topology to obtain a time-space aligned network sequence; Based on the time-space aligned network sequence and a preset time-space fusion weight function, time-varying risk feature vectors corresponding to all connecting edges in the time-space aligned network sequence are obtained; Obtaining a three-dimensional space-time correlation matrix based on the time-varying risk feature vector, wherein a first dimension of the three-dimensional space-time correlation matrix and