Research On Dynamic Allocation Of Emergency Resources Based On Meteorological Disaster Data Driven

Due to the characteristics of various types,wide range,high frequency,long duration,prominent cluster occurrence,significant chain reaction and heavy disaster,meteorological disasters have continuously sounded the alarm to the world with their serious destruction and huge loss of life and property.Therefore,it is the common demand of all countries in the world,including China,to enhance the awareness of the inherent characteristics of these increasingly frequent meteorological disasters and enhance the emergency response capacity to deal with such disasters.Based on this,aiming at the meteorological disaster,this paper takes the emergency logistics activities and emergency response decision-making in the early stage after the disaster as the research object,takes improving the emergency response ability and decision-making optimization in response to such disasters as the research goal,and focuses on the allocation of emergency rescue resources(including materials,vehicles and personnel,etc.),so as to reveal the inherent significant characteristics of meteorological disasters.According to the research idea of "identifying problems,defining problems and solving problems",and around the inherent requirements of "dynamic,spatiotemporal and integrated" meteorological disaster emergency decision-making,this paper uses artificial neural network,heuristic algorithm,decomposition algorithm theory and epsilon constraint algorithm theory to carry out emergency aid for meteorological disasters research on modeling and algorithm design of source dynamic allocation.Specifically,the dynamic allocation optimization of emergency rescue resources includes the following five parts:(1)In view of the meteorological disaster emergency disaster emergency rescue background,proactive scheduling model,establishing emergency rescue in the model considering the robustness of emergency rescue and punishment cost,using the back propagation(BP)neural network to predict the dynamic requirements,using the improved differential evolution algorithm and genetic algorithm,the algorithm is validated by the instance when calculating the emergency rescue a proactive scheduling model,has good effectiveness.The results show that the improved differential evolution algorithm is superior to the improved genetic algorithm in terms of CPU computation time.The difference evolution algorithm is superior to the genetic algorithm in convergence,but the genetic algorithm is superior to the difference evolution algorithm in accuracy.Provide reference for decision makers,and then provide experience for future emergency decision.(2)In view of the real-time variation of meteorological disasters and the differences between disaster areas,the dynamic distribution of emergency relief materials in the early stage of meteorological disasters was studied.By analyzing the real-time update characteristics of the demand and supply information of emergency relief materials,this paper studies the dynamic scheduling method of emergency relief materials under the condition of dynamic update of space-time information from two dimensions of time and space.For disaster situation in the first time and the dynamic change of characteristics of the geographical area,using the time-space network flow method,the construction of emergency relief supplies dynamic scheduling of time and space network flow model,and then consider the height of the scheduling information characteristics of uncertainty and unpredictability,predict using extreme learning machine,study more decision-making cycle rolling mode of emergency relief supplies real-time dynamic scheduling time-space network flow model,in order to solve the information under the condition of uncertainty and dynamic update of meteorological disaster emergency scheduling decision problem provides a new train of thought.Benders decomposition algorithm was used to solve the model,and an example was set to solve the model and algorithm,so as to verify the effectiveness of the algorithm and model.(3)Aiming at the contradiction between the severity of meteorological disasters and the scarcity of emergency relief materials,the dynamic scheduling of emergency relief materials in the early stage of meteorological disasters is studied.Firstly,the dispatch model of emergency relief materials is established,and the utility of emergency demand satisfaction is quantified and defined from the perspective of utility analysis.The functional relationship between emergency demand utility and emergency relief materials is analyzed,and the emergency demand is analyzed by grey neural network.Then,from the perspective of utility,urgency and equity,this paper explores the internal principle and realization mechanism of dynamic allocation of emergency relief materials in meteorological disaster environment,and proposes a multi-objective model of dynamic allocation of emergency relief materials based on multi-demand points.Finally,based on the dynamic distribution model of multi-demand emergency relief materials,three multi-demand emergency relief materials distribution strategies are proposed,including relative rescue time priority strategy,fair scheduling strategy and delay minimum priority strategy.The feasibility and rationality of the model are proved.(4)Aiming at the problem of the interdependence between the allocation and scheduling of emergency relief materials for meteorological disasters,the integrated modeling of multiple emergency rescue decision-making links under the condition of uncertain decision information and coupling of decision variables is studied.From rescue link integration point of view,considering the distribution and scheduling link intrinsic correlation,emergency relief supplies will be allocated dynamically integrated consideration,and the dynamic scheduling problem using a distribution model of distribution network flow of the variables in the decision variables and scheduling model between conservation conditions,research to build the dynamic allocation and dynamic scheduling integrated emergency relief supplies the multi-objective model of emergency rescue for many meteorological disasters,many departments processes real-time rescue decision this complex problem provides quantitative auxiliary decision making tool.The wavelet neural network is used for demand analysis,and the genetic algorithm is used to solve the material scheduling model.A series of large-scale numerical experiments are conducted to verify the effectiveness of the algorithm and model.(5)In view of the meteorological disaster affecting the universality and emergency rescue of the rescue team assignment problem,from the perspective of complex problem solving,using particle swarm algorithm and genetic algorithm(GA)theory,the research of the meteorological disaster emergency rescue teams assignment model to solve the problem,for post-disaster emergency decision-making scheme generation of real-time and timeliness required to provide stable convergence algorithm of protection.Firstly,the model is abstracted mathematically into a kind of emergency rescue decision model.Secondly,in the context of meteorological disasters,the fuzzy neural network is used to predict the emergency rescue demand analysis.According to the basic framework of particle swarm optimization algorithm and genetic algorithm,the algorithm for solving the rescue team assignment model in the emergency rescue of meteorological disasters is designed,and the convergence of the two algorithms is proved.Finally,an actual example used to verify the effectiveness of the two algorithms.Through numerical simulation and analysis,it is found that the particle swarm optimization(PSO)algorithm has a good performance in solving the problem of emergency rescue team assignment in meteorological disasters,and the genetic algorithm has a significant improvement in the computational performance of the problem.This thesis research work for the meteorological disaster emergency rescue the complex management decision problem provides a new solution,the emergency relief materials obtained dynamic allocation and dynamic scheduling model,strategy and algorithm research,the formation of innovation system of meteorological disaster response to theoretical research,to improve the timeliness of meteorological disaster emergency response system and real time,the implementation of scientific and effective control strategy,all plays an important role in promoting,for emergency rescue decision makers timely,efficient and properly dispose of highly uncertain and dynamic evolution of meteorological disasters provide scientific decision basis,has important theoretical significance and application value.