Adaptive V2G Peak Shaving and Smart Charging Control for Grid Integration of PEVs

Abstract

The stochastic nature of plug-in electric vehicle (PEV) driving behavior and distribution grid load profile make it challenging to control vehicle-grid integration in a mutually beneficial way. This article proposes a new adaptive control strategy that manages PEV charging/discharging for peak shaving and load leveling in a distribution grid. For accurate and high fidelity transportation mobility modeling, real vehicle driving test data are collected from the field. Considering the estimated total required PEV battery charging energy, the vehicle-to-grid capabilities of PEVs, and the forecasted non-PEV base load, a reference operating point for the grid is estimated. This reference operating point is updated once at the end of peak hours to guarantee a full final state-of-charge to each PEV. Proposed method provides cost-efficient operation for the utility grid, utmost user convenience free from range anxiety, and ease of implementation at the charging station nodes. It is tested on a real residential transformer, which serves approximately one thousand customers, under various PEV penetration levels and charging scenarios. Performance is assessed in terms of mean-square-error and peak shaving index. Results are compared with those of various reference operating point choices and shown to be superior.