Research On Real-time Information Electric Vehicle Path Planning

With the rapid growth of the demand for electric vehicles in various scenarios,the transportation network is becoming more and more complex and changeable.The traditional vehicle routing problem(VRP)solution method is difficult to meet the diversified actual needs of today’s electric vehicle(EV)users.Therefore,studying the electric vehicle routing problem(EVRP)is crucial to the development of route navigation and logistics services.Based on the previous researches on traditional VRP at home and abroad,this paper focuses on the new dynamic EVRP model and solution algorithm.The main research contents of this thesis are as follows:(1)Study the effect of waiting time for charging in the EV path planning on the overall travel time,and develop a dynamic real-time information processing system.Based on the research on the queuing theory of multiple service desks(M/M/S),based on the real-time traffic node inflow rate,the probability that EV users need to be queued for charging is derived,and the estimated queue length is normalized.Strategies to generate the queued charging congestion factor of each charging node.(2)Model the optimal time vehicle routing problem for electric vehicle users,and design an improved ant colony algorithm(IACO)to solve the model,and provide path planning and solution strategies.The IACO algorithm proposed in this chapter optimizes the global pheromone update rule,and uses the 2-opt heuristic algorithm to optimize the local path exploration ability,in order to prevent the impact on the search for the global optimal value,and adopts the ant regression strategy to deal with ants Deadlock situation.Based on the previous charging node congestion factor,the algorithm initialization pheromone matrix is improved to make it differentiated distribution and enhance the global search performance of the path planning algorithm in the early stage.Finally,in the experimental part,in order to verify the effectiveness of the IACO algorithm in this scenario,data simulations are carried out based on three different traffic networks with different degrees of congestion,and the experimental results are compared with the Dijkstra algorithm and the ant colony algorithm,and finally performed Convergence conclusion analysis.(3)Model and solve the problem of real-time dynamic logistics distribution of electric vehicles.A mathematical model was established based on the constraint characteristics of EVRP,and an ant colony algorithm(GA-ACO)fused with genetic algorithm was designed to solve it.The algorithm uses the advantages of ant colony algorithm and genetic algorithm,adopts island parallel algorithm structure,uses single-point crossover mutation operation to perform local optimization,and enhances the algorithm’s ability to find better solutions based on the concept of population exchange individuals,and The specific update rules of genetic information are given.Finally,in the experimental part,the results of GA-ACO and ACO algorithms are compared on data sets of different scales to verify the effectiveness of the solution algorithm,and the convergence analysis is carried out.