Optimization Of Electric Vehicles’ Charging Station Location

With the rapid depletion of oil resources and aggravation of environmental pollution, nations across the world are faced with severe energy and environmental crisis. In this case, the adoption of electric vehicles (EV) can replace traditional fuel into clean fuel and reduce vehicle exhaust emissions, which is vital for securing energy safety and controlling air pollution. Nevertheless, there are several obstacles to EV adoption, the most important being the chicken-and-egg problem related to EV charging infrastructure. Considering the high capital cost of EV charging infrastructure, the optimization of charging infrastructure locations under budget constraint becomes the hottest issue. This dissertation address such issue by researching optimization method for EV charging infrastructure locations.First of all, this dissertation introduced concepts and characteristics of electric vehicles (EV) and EV charging infrastructures, discovering that EV’s limited driving range and lengthy charging time will probably cause queueing in charging stations and therefore make an impact on EV charging station layout. This dissertation then addressed the "range anxiety" issue, and studied the travel pattern and charging behavior of EV users on the foundation of household travel survey results. On top of that, the dissertation developed the idea that EV charging demand can be predicted, using arrival rate as a proxy.After that, the dissertation took a zoomed-in three step approach to predict EV’s charging demand: ①use elastic coefficient method to predict the amount of EV and chargers needed; ②use gravity model to quantify traffic analysis zones’(TAZs) trip production volume, trip attraction volume, traffic impedance, then calibrate coefficients and predict trip distribution volume among TAZs;③se peak hour volume ratio and high charging tendency population ratio to predict charging demand rate of all TAZs.Finally, the dissertation analyzed possible factors that may influence charging station layout, made several assumptions in uses’charging behavior, took chargers’ amount constraint into consideration, and built EV charging station location optimization model oriented at minimizing users’total time cost. A genetic algorithm is proposed to solve the model. The dissertation then took Chengdu Hi-tech Zone as an example, wrote a GA based program on MATLAB software, through which the optimal charging station location and charger allocation plan is found. The result shows that when available charger amount increases, charging stations’location tend to be more dispersed; in certain extent, the marginal benefit of adding one charger is substantial, and therefore charging facilities are advised to be constructed moderately excess of demand. The example shows that the EV charging station location optimization model is robust and worth spreading to larger urban area.