Research On Intelligent Charging Management Of New Energy Vehicles Based On Big Data Analysis

New energy vehicles(NEVs)are replacing traditional fossil energy vehicles with their environmental friendliness and clean energy advantages.Promoting new energy vehicles has become a workable solution for countries all over the world to cope with the serious problems of environmental degradation and energy depletion.The widespread use of new energy vehicles has also led to a series of reforms in the energy system,economic system,and service system.More and more companies hope to obtain considerable economic profits by operating charging stations.New energy taxis(NETs),as the main service objects of charging station operators,face more serious "mileage anxiety" in fulfilling the demand of traditional petrol taxis to pick up and drop passengers.The taxi operating model that considers picking up and dropping off passengers can no longer meet the comprehensive needs of various aspects generated during the NETs operation.Advanced big data analysis technology has become a viable technology for operators to make profits in market competition.The collection and analysis of an enormous amount of effective data make the company’s operational decision-making more accurate and creates intelligent management of the new energy vehicle charging market.This thesis studies the concept and technology of intelligent charging management for NEVs(including NETs)based on big data analysis technology.This thesis summarizes the specific research subjects and results in the following three points.1)In response to the dynamic pricing problem of charging station operators in the complex and changeable market competition,this thesis proposes a novel three-layer framework for the new energy vehicle market,and proposes a dynamic pricing algorithm for charging station operators based on charging demand analysis to maximize long-term profitability of operators.First,the three-layer framework proposed in the first chapter describes the economic model of the new energy vehicle market,which is composed of smart grids,operators,and charging stations for new energy vehicles from top to bottom.Next,this thesis uses the Markov game to model the second layer of the framework as a competitive market.Besides,the thesis designs a dynamic pricing policy based on multi-agent reinforcement learning to achieve higher long-term profits for operators.Experimental results show that the effect of the algorithm on the operator’s long-term profit increases over time.Compared with the random pricing scheme and the greedy pricing scheme,the dynamic pricing algorithm increases the operator’s long-term profit by approximately 31%and 21%,respectively.In addition,the algorithm can reduce the profit loss of the original operators when there are more new operators in the market.2)In response to the contradiction between the charging process and the service process that NETs face in the promotion process,the thesis proposes a novel double queue model and proposes a new energy vehicle charging scheduling algorithm based on taxi passenger demand analysis,aiming to maximize the profit of the charging station and guarantees the demand of NETs and passengers.This thesis first proposes a dual-queue system to describe the coupling relationship between NETs,charging stations,and passengers and then model the charging scheduling as a stochastic optimization problem based on this system.To solve this problem,this thesis proposes a dynamic charging scheduling algorithm(DCSA)based on Lyapunov optimization.In the end,this thesis proves the effectiveness of DCSA.Based on real data analysis of Beijing NETs and passenger demand,the experimental results show that on the basis of guaranteeing passenger demand,the long-term average profit of DCSA,nearby charging mechanism and random charging mechanism at charging stations has increased by 33.485%and 29.897%,respectively.3)In response to the potential cost problem of the non-service period due to charging caused by NETs during the promotion process,this thesis proposes a novel NETs downwind operation mode and proposes new energy based on the fastest and shortest path algorithm(SPFA)Car routing algorithm.The thesis first divides the study area into square areas of the same size and proposes to use the pickup probability to describe the potential downwind probability of sending passengers down the wind when the NETs pass by in each area.Also,through data analysis,the thesis gets the time distribution characteristics of the travel demand of taxi passengers in Beijing.This thesis proposes a routing algorithm based on SPFA,which surveys the energy,time cost,and benefits of sending passengers down the wind during the charging routing process of NETs.Based on the experimental analysis of the real data set in Beijing,it is verified that the algorithm can reduce the routing cost of NETs.In the working day,compared with the optimal energy solution and the optimal distance solution,it can reduce up to 17.06%and 30.52%respectively.On weekends,compared with the baselines,the reduction is up to 10.18%and 20.67%,respectively.