CN-121997606-A - Scene digital twinning-based fire emergency simulation method and system

CN121997606ACN 121997606 ACN121997606 ACN 121997606ACN-121997606-A

Abstract

The invention relates to the technical field of fire simulation, and discloses a scene digital twinning-based fire emergency simulation method and a scene digital twinning-based fire emergency simulation system. The method comprises the steps of obtaining space topology data of a target building through a three-dimensional laser scanning technology, combining temperature gradient distribution data collected by a thermal imaging sensor to construct a geometric model and a thermodynamic property model of a digital twin body of the building, extracting pyrolysis kinetic parameters of different materials according to fire source diffusion mode data in a historical fire case library to construct a fire spreading probability matrix, inputting real-time meteorological monitoring data into the thermodynamic property model to calculate a dynamic coupling relation between a temperature field and a smoke concentration field in the building, generating a fire scene evolution path set based on the fire spreading probability matrix and the dynamic coupling relation, and outputting a traffic capacity attenuation curve of a key escape node. The method realizes the accurate simulation and dynamic evolution prediction of the fire scene, and provides scientific support for fire emergency decision and personnel evacuation planning.

Inventors

TIAN HUA
GUO SUYAN
LI HAIXIA
FAN SUMIN

Assignees

仁安消防设备科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260130

Claims (10)

1. The scene digital twinning-based fire emergency simulation method is characterized by comprising the following steps of: acquiring space topology data of a target building through a three-dimensional laser scanning technology, and generating a geometric model and a thermodynamic property model of a digital twin body of the building by combining temperature gradient distribution data acquired by a thermal imaging sensor; According to fire source diffusion mode data in a historical fire case library, extracting pyrolysis kinetic parameters of different materials, and establishing a fire spreading probability matrix; Inputting real-time meteorological monitoring data into a thermodynamic attribute model, and calculating a dynamic coupling relation between a temperature field and a smoke concentration field in a building; based on the fire spreading probability matrix and the dynamic coupling relation, a fire scene evolution path set is generated, and a traffic capacity attenuation curve of the key escape nodes is output.
2. The scene digital twin based fire emergency simulation method according to claim 1, wherein the generating a geometric model and a thermodynamic property model of a building digital twin body comprises the steps of: Performing non-uniform rational B-spline surface reconstruction on the space topology data, and eliminating topology faults caused by scanning blind areas; mapping the temperature gradient distribution data to the surface grid vertexes of the geometric model by adopting a Kriging interpolation algorithm; And matching the thermal resistance value of the building component through a material thermal conductivity coefficient database, and completing parameterization calibration of the thermodynamic property model.
3. The scene digital twinning-based fire emergency simulation method according to claim 2, wherein the establishing a fire spread probability matrix comprises the steps of: extracting a fire source ignition time sequence from a historical fire case library, and fitting heat release rate critical values of different materials by adopting Weber distribution; constructing a directed propagation graph taking heat radiation flux as a weight according to the spatial adjacent relation between building components; And (3) iteratively calculating ignition probability thresholds of all nodes in the directed propagation graph through Monte Carlo simulation to generate a fire spreading probability matrix.
4. A scene digital twinning-based fire emergency simulation method according to claim 3, wherein the calculating of the dynamic coupling relation between the internal temperature field and the smoke concentration field of the building comprises the steps of: Decomposing a wind speed vector in the real-time meteorological monitoring data into a normal component and a tangential component of a building elevation; Solving the enhancement coefficient of the normal component to the chimney effect of the vent based on a large vortex simulation method; and tracking a smoke diffusion track driven by tangential components by adopting a particle system, and establishing a space-time coupling equation of a temperature field and a smoke concentration field.
5. The scene digital twinning-based fire emergency simulation method according to claim 4, wherein the generating a set of fire scene evolution paths comprises the steps of: extracting a component number sequence exceeding a ignition probability threshold value from a fire spreading probability matrix; Predicting a temperature field mutation time point corresponding to the component number sequence according to the space-time coupling equation; and aligning the abrupt time point with the inflection point of the traffic capacity attenuation curve to generate a fire scene evolution path set with a time stamp.
6. The scene digital twinning-based fire emergency simulation method according to claim 5, wherein the outputting of the traffic capacity attenuation curve of the critical escape node comprises the steps of: marking topological center coordinates of the stairwell and the safety exit in the geometric model; Calculating the shortest path distance from the topological center coordinates to the position of the fire source through a Di Jie Style algorithm; and dynamically correcting the passing resistance coefficient of the shortest path distance according to a space-time coupling equation of the smoke concentration field to generate a traffic capacity attenuation curve.
7. The scene digital twinning-based fire emergency simulation method according to claim 6, further comprising the steps of: when a component number sequence is newly added in the fire spreading probability matrix, triggering an incremental updating mechanism of the evolution path set; And adopting a Kalman filter to fuse the measured data and the predicted data of the abrupt change time point of the temperature field, and recalibrating the parameter error of the thermodynamic property model.
8. The scene digital twinning-based fire emergency simulation method according to claim 7, wherein the incremental update mechanism is executed by: Comparing the spatial distribution similarity between the newly added component number sequence and the historical case library; if the similarity exceeds a preset threshold, calling pre-stored pyrolysis kinetic parameters to directly generate an evolution path branch; and if the similarity is lower than a preset threshold, starting a local recalculation flow of the thermodynamic property model.
9. The scene digital twinning-based fire emergency simulation method according to claim 8, wherein the local recalculation process comprises: isolating a geometric model subarea corresponding to the newly added component number sequence; improving the solving precision of the heat conduction equation of the subarea by adopting the self-adaptive grid encryption technology; and carrying out Gaussian smoothing splicing on the calculation result and the global temperature field to finish updating of the evolution path set.
10. A scene digital twinning based fire emergency simulation system comprising a memory, a processor and a computer program stored in the memory and running on the processor, characterized in that the processor, when executing the computer program, realizes the steps of the scene digital twinning based fire emergency simulation method according to any of claims 1 to 9.

Description

Scene digital twinning-based fire emergency simulation method and system Technical Field The invention relates to the technical field of fire simulation, in particular to a scene digital twin-based fire emergency simulation method and system. Background Fire is one of the most damaging incidents in the field of building security, and the sudden nature of the outbreak, the complexity of fire spread and the uncertainty of smoke spread form serious threats to personnel life safety and property protection. The effective fire emergency simulation can predict the fire evolution trend in advance and clear the critical escape path, and is an important means for improving the fire emergency treatment efficiency and reducing the disaster loss. However, the current mainstream fire emergency simulation technology still has a plurality of problems to be solved. In the aspect of building model construction, the traditional method depends on two-dimensional drawing digitization or simple three-dimensional modeling, and is difficult to accurately restore the actual space topological structure of a building, especially the detail information such as beam column distribution, pipeline trend and the like in a complex building. Meanwhile, the existing model only pays attention to the re-engraving of geometric forms, lacks of accurate representation of thermodynamic properties of building materials, cannot truly reflect temperature response characteristics of different areas in a fire, and causes larger deviation between subsequent fire simulation and actual scenes. In addition, although some technologies attempt to introduce sensor data, a single type of sensor is often adopted, so that effective fusion of space data and thermodynamic data cannot be realized, and the reliability of the model is further reduced. In the fire spreading prediction link, the existing model is generally calculated based on a fixed diffusion formula, and the influence of the pyrolysis dynamics difference of different building materials on fire spreading is ignored. The data of the diffusion mode of rich fire sources contained in the historical fire cases are not fully mined, so that the probability prediction of the fire spread is insufficient, and the random change in the fire development process is difficult to deal with. Meanwhile, weather conditions are taken as important external factors for influencing fire spread, the current simulation method is often regarded as static parameters, the calculation process of a temperature field and a smoke concentration field cannot be integrated in real time, the dynamic influence of weather factors such as wind, humidity and the like on fire evolution cannot be accurately reflected, and the coupling calculation result of the temperature field and the smoke concentration field is disjointed from the actual situation. In the aspect of emergency escape guidance, the prior art can only provide fixed escape route suggestions, and lacks assessment of dynamic change of the traffic capacity of key escape nodes in the fire development process. After a fire disaster occurs, factors such as temperature rise, smoke accumulation and the like can continuously weaken the traffic capacity of a channel, a fixed escape route can be invalid due to channel blockage in an actual fire scene, timely and effective guidance can not be provided for people evacuation, and the safety and efficiency of emergency evacuation are further affected. The existence of the problems causes that the current fire emergency simulation technology is difficult to meet the actual requirements of accurate emergency decisions in complex building scenes, and a new method capable of integrating multi-source data and realizing dynamic accurate simulation of fire scenes is needed. Disclosure of Invention The invention aims to provide a scene digital twinning-based fire emergency simulation method and a scene digital twinning-based fire emergency simulation system, which are used for solving the problems in the background technology. To achieve the above object, the present invention provides a scene digital twin-based fire emergency simulation method, which includes: acquiring space topology data of a target building through a three-dimensional laser scanning technology, and generating a geometric model and a thermodynamic property model of a digital twin body of the building by combining temperature gradient distribution data acquired by a thermal imaging sensor; According to fire source diffusion mode data in a historical fire case library, extracting pyrolysis kinetic parameters of different materials, and establishing a fire spreading probability matrix; Inputting real-time meteorological monitoring data into a thermodynamic attribute model, and calculating a dynamic coupling relation between a temperature field and a smoke concentration field in a building; based on the fire spreading probability matrix and the dynamic coupling relation, a fire scene