Research On Evacuation And Rescue Planning Mechanism For Victims In Disasters

In recent years,various kinds of natural hazards or man-made disasters happened with heavy casualties.These disasters were diversified and complicated,and difficult to predict and control.In disaster scenario,how to use informational technology to carry out fast and effective disaster response is the key,which attracts more and more attentions of governments,academia and industry around the world.In this thesis,we study the evacuation planning and rescue route planning problems in disasters.The main research work of this thesis include:(1)An Evacuation Planning Algorithm based on Artificial Potential Field(EPAAPF)is proposed.The basic idea of EPAAPF is to establish a model of artificial potential field to describe the complex large-scale evacuation problem with the potential field function,which simplifies the difficulty of modelling the complex disaster scenario.The experimental results show that EPAAPF can enhance the efficiency and accuracy effectively,and shorten the evacuation route length and time.The distribution of shelters reach a balanced state.(2)In order to realize a more humanitarian evacuation,we take the relationships into consideration,and propose an Evacuation Planning Algorithm based on Artificial Potential Field with Relationship Attractions(EPAAPF-RA).We introduce the relationship attraction potential field into the ultimate resultant potential field function.Under the premise of the safety,evacuees with relationships can be evacuated to the same shelter to achieve humanitarian evacuation.Experiments show that the algorithm can achieve high efficiency and humanitarian evacuation.(3)We abstract Emergency Rescue Planning Problem(ERRP)to Multiple Traveling Salesman Problem(MTSP).At first,we propose an Obstacle Constrains and Task of Equal Division based K-means++ Clustering Algorithm(OT-K-means++)to decompose MTSP into several Traveling Salesman Problems(TSP),and cluster multiple victims to be rescued in disasters,so that the difference of nodes in the clusters is smaller and the number of nodes is more uniform.Then,a Glowworm Swarm Optimization Algorithm based on Chaotic Initialization(GSOCI)is proposed to solve each TSP.By introducing Chebyshev mapping,the fireflies are initialized to improve the quality of initial distribution and population diversity,and introduce C2 Opt for local adjustment and optimization to enhance the convergence speed of the algorithm.The experimental results show that this algorithm can improve the accuracy and convergence speed of the solutions effectively,shorten the rescue route length and time,and improve the rescue efficiency.(4)Based on the Mobile Cloud Computing(MCC)platform,we construct an emergency evacuation and rescue route planning system for disasters.The system realizes the cooperation between mobile terminals and cloud servers: mobile terminals submit the location information and other collected information to cloud servers,the cloud server push the information of disaster to mobile terminals and then provide the planning schemes for mobile terminals.Finally,EPAAPF and EPAAPF-RA proposed in this thesis are applied to the evacuation module of the system,which can realize rapid and effective evacuation,shorten the evacuation route length and time,make the remaining capacity of the surrounding shelters balanced,and realize humanitarian evacuation aims.Then OT-K-means++ and GSOCI proposed in this thesis are applied to the rescue module of the system,which can improve the accuracy and convergence speed of the solutions effectively,shorten the rescue route length and time,and achieve high efficiency rescue.