Research On Distribution Scheduling And Algorithms Of Electric Unmanned Vehicles In Urban Logistics

With the rapid growth of online shopping in urban,the significance of urban logistics has attracted extensive attention.In order to deal with the sustainable growth of logistics distribution volume efficiently in the city and satisfy the distribution needs of customers to better improve the consumer experience,the urban unmanned logistics distribution represented by the electric unmanned vehicle has become a reality and received widespread attention.Currently,the theoretical research mainly focuses on the feasibility of unmanned logistics distribution and the operation planning of electric unmanned vehicles.Further the research considering the distribution characteristics of electric unmanned vehicles is insufficient.The principle of electric unmanned vehicles routing,from distribution center to the completion of delivery task,can refer to the famous travelling salesman problem.Nevertheless,it is necessary to further describe the charging decision of electric unmanned vehiclesThis thesis studied the application of electric unmanned vehicles in urban logistics distribution.To guarantee that the power of electric unmanned vehicles is adequate while delivering,the thesis proposes a model for the location of charging stations,and the improved k-means algorithm is designed to solve the problem,the number and location of charging stations are reasonably arranged.Considering the uniqueness of electric unmanned vehicles,this thesis put forward the corresponding charging principle of the electric unmanned vehicle and analyzes the logistics cost,and establishes the corresponding nonlinear integer programming model.This model is an extension of the classical vehicle routing problem(VRP)and can be seen as a NP-hard problem,which can be solved more effectively by adopting intelligent algorithm.Consequently,we designed the genetic simulated annealing algorithm and introduced the solution idea and process of the algorithm.Finally,lingo solver is used to solve the path planning model,proving the correctness of the path planning model and the feasibility of genetic simulated annealing algorithm.In the stage of case analysis,the operation of M company is briefly described,and 800 pieces of customer data of M company are selected for optimization analysis.Starting from the construction cost of the charging station,considering the large scale of data,the improved k-means algorithm is used to locate the charging station.By comparing the cost of building different charging stations,the reasonable number and location of charging stations are determined.In the routing of electric unmanned vehicle,using the established mathematical model,taking the vehicle fixed use cost,driving cost,waiting cost and charging cost as the objective function,vehicle load,electricity and customer time window as constraints,and algorithm is used to plan the large-scale data in this case.Finally,the performance of genetic simulated annealing algorithm is further analyzed to verify its superiority in solving large-scale problems.The sensitivity analysis of parameters such as driving speed,coefficient of power consumption and charging is carried out.The results show that these factors have a certain impact on the distribution cost.The maximum running range of electric unmanned vehicles in urban logistics distribution is limited by the electric quantity.Therefore,how to make the most reasonable layout of charging stations and how to find and select charging stations for electric unmanned vehicles to ensure the safety of distribution have become key problems in the current city.This study can provide reference and guidance for the development of urban logistics distribution business of logistics companies.