Research On Dynamic Bi-objective Evacuation Path Planning For Building Fires

When a building fire occurs,risk factors such as smoke and high temperature can seriously threaten the mobility and safety of escapees,which in turn can lead to larger casualties.In the evacuation process of building fires,the lives of evacuees should be safeguarded to the greatest extent possible.Therefore,how to plan a safe and efficient evacuation path for the evacuees in the event of a fire is of great importance in the field of public safety.This study firstly takes the risk factors in building fires as the entry point,and takes five major risk factors in building fires,namely carbon monoxide,hydrogen cyanide,ambient temperature,visibility distance and crowding,as the target of the study.And quantifies the threat of the risk factors to the evacuees through a semi-quantitative calculation method.Secondly,a single objective path planning model with fire risk as the optimization objective is established based on the quantified risk,and a single objective path planning model with path length as the optimization objective is established to take into account evacuation efficiency.The weighted ideal point method is used to transform the two single objective models into a bi-objective path planning model,through which the Pareto-optimal solutions on the fire risk objective function and the path length objective function can be obtained.The bi-objective control of the path optimization results is finally realized.Meanwhile,the classical Dijkstra algorithm is improved.The improved algorithm can dynamically adjust the weights of the objective function and ensure the safety of the planned path and the efficiency of the algorithm operation through the mechanisms of forbidden threshold and virtual endpoint,etc.The algorithm can cooperate with the bi-objective path planning model to realize the multi-objective planning of building fire evacuation paths for optimization.Finally,the numerical simulation of fire in a city building is carried out by Pyro Sim software to obtain the values of smoke and ambient temperature at different moments.The set parameters and the numerical values of the fire obtained from Pyro Sim simulation are loaded into the program for simulation using MATLAB.The results show that the evacuation method using the bi-objective path planning model and the improved Dijkstra algorithm is significantly better than the disorderly evacuation method.The total evacuation time for all people to evacuate to the safety exit is reduced by 40 s and the total walking distance for all people is reduced by 3650 m.Subject to the constraints of the fire objective function,the optimal evacuation path avoids the highrisk area and satisfies the safe evacuation demand while ensuring the efficient evacuation due to the constraints of fire objective function and path length objective function.The above analysis and verification prove the superiority of this evacuation path planning method,which provides new ideas for future research on building fire path planning and has certain practical significance.