Joint Temporal-Spatial Coordination Of Charging Decision Of Plug-In Electric Taxi Fleet

The increasing traditional fuel-engined vehicles have been causing many worldwide concern-s,such as greenhouse gas emissions,fuel oil energy crisis.Due to economical and environmental friendly advantages,electric vehicles(EVs)are vigorously promoted by policy makers in recent years,and expected to penetrate deeply in the near future.Meanwhile,EVs as large storage ap-pliances offer huge potential for smart grid.Plug-in electric taxis(PETs)as an important part of EVs,have pressing demands on scheduling.It is of great significance to develop efficient charging scheduling algorithms.In this study,we consider a city with a large fleet of PETs and studies the charging coordi-nation problem of the fleet.The goal is to reduce charging cost,defined as the potential loss of service income caused by charging,for each PET by wisely choosing when and where to charge in an online fashion upon receiving real time information.Through analysis it is seen that the exact position of PETs are critical factors in decision when choose charging stations(CSs).Since the exact future positions remain highly randomly for PETs,it’s hard for PETs to make decision in advance before make sure when to charge.This problem is approached in two stages.In the first stage,a thresholding method is proposed to assist PET driver choosing a proper time slot for charg-ing,with joint consideration of state of charge(SOC)of PET,time varying income and queuing status at CSs.In the second stage,a game-theoretical approach is devised to dispatch PETs to CSs,so that the traveling and queuing time of each PET can be reduced with fairness.The proposed algorithms can effectively reduce the charging cost of PETs,as well as enhance the utilization ratio of CSs,as illustrated by extensive numerical simulations.