Research On Charging Behavior Of Electric Vehicles Based On Multiple Objectives

With the rapid development of society,technology has made great progress.While enjoying the dividends brought by technology,people should also face up to a series of problems caused by the energy crisis since resources and the environment are facing severe global challenges.Exhaust emissions caused by extensive use of traditional fuel vehicles will affect human health.Therefore,the research progress of electric vehicles(Electric Vehicle,EV)must be accelerated.Due to the gradual improvement of the system and the continuous follow-up of equipment,electric vehicles have good development prospects in both private and public fields.For the charging research of traditional electric vehicles,many scholars only consider a single influencing factor,such as the charging cost of electric vehicles,the probability of successful travel of electric vehicles,and the peak value of electric vehicle charging with its impact on the grid.Few scholars have discussed and studied the problem that electric vehicles cannot be charged in time due to the unbalanced vehicle-to-pile ratio,but this is an inevitable problem in daily life.According to the indicators that the public cares about,this thesis proposes a multi-objective-based queuing charging strategy for electric vehicles.This strategy combines the common charging modes,that is,random charging mode,time-of-use electricity price mode and stop-and-charge charging mode,and considers the unbalanced vehicle-to-pile ratio in these common modes,and then introduces relevant problem with queuing up electric cars to charge.In the process of building the model,the problem that the non-uniform battery capacity of different types of electric vehicles on the market is considered,and the battery capacity is divided into two categories: large battery capacity and small battery capacity.An emergency charging mode is also considered,in which the user can realize emergency charging by increasing the price.Aiming at the optimization problem in the proposed multi-objective-based queuing charging strategy for electric vehicles,the differential evolution algorithm(Differential Evolution,DE)is improved,and the Two-Stage Guided Constraint Differential Evolution(TSGCDE)algorithm is proposed.By using the international common CEC2013 and CEC2014 benchmark test sets as well as the MATLAB platform,the optimization experiments of 58 test functions were carried out.It is verified by experiments that the TSGCDE algorithm proposed in this thesis has strong competitiveness in terms of convergence and optimization ability.This thesis then uses the TSGCDE algorithm to set different numbers of electric vehicles for the established multi-objective electric vehicle queuing charging strategy,and conduct Monte Carlo(MC)simulation and establishe a comprehensive average index including the average peak ratio,charging cost and successful travel probability.The experimental demonstration of the proposed strategy of comparing multiple strategies for1000 electric vehicles shows that the charging cost savings of the multi-objective queuing combination charging strategy is 0.635 and the comprehensive evaluation index is0.570.Among 500 EVs,the combined multi-objective queuing charging strategy still has the highest overall evaluation index of 0.583;the charging cost saving rate is 0.635.The multi-objective EV queuing charging strategy takes into account the unavoidable queuing problem in reality and reduces charging cost while improving the convenience of life.From the perspective of the power grid,the proposed strategy makes the charging load curve of electric vehicles tend to be flat,which greatly reduces the burden on the power grid caused by charging load fluctuations,and truly achieves a win-win situation between the supply side and the demand side.It makes a certain contribution to the research of electric vehicle charging behavior.