EV Battery Wear Cost Optimization for Frequency Regulation Support in a V2G Environment
Olalekan Kolawole · 2018
In addition to improving the ground transport and the environment through reduced greenhouse gas emissions, Electric vehicles (EVs) can support a number of power grid services through the Vehicle to Grid (V2G) system. If EVs charging and discharging are properly integrated and managed, they can be grouped to provide ancillary services like Frequency Regulation (FR), peak load leveling etc. Proper integration of EVs can help in integrating renewable energy sources and provide various demand response. There are a number of challenges that should be addressed before EVs can effectively provide these services. The main challenge is the availability of power from EVs. This challenge is related to the EV battery capacity, the available power at the time of need and the battery cycle life. Battery cycle life is inversely proportional to the number of charge-discharge cycles the battery goes through. Therefore, the battery cycle life and the cost of degradation should carefully be included when optimizing the V2G operation. In this thesis, a critical study and investigation of the V2G concept is presented. An Electric Vehicle Charge-Discharge (EVCD) optimization model is presented, developed and evaluated to incorporate actual and forecast FR, electricity prices from two electricity vendors (NYISO and PJM), battery wear and other parameters into the objective function to generate optimum solutions for individual time slots. The optimization problem is solved using Mixed Integer Linear Programming (MILP) to minimize the cost of providing the service. A case study was developed for the charge scheduling problem which encompasses the actual, forecast regulation and electricity prices. A forecasting model was investigated and incorporated with the optimization model to provide an insight to EV owners on the time to charge/discharge their batteries for the regulation service. Furthermore, the results were validated by analyzing the optimal state of charge (SOC) profile and charge/discharge schedule of the EVs participating in the regulation service. An iterative algorithm to predict the effect of the actual and forecast electricity and regulation prices on the EV battery cycle life was presented and evaluated. In our analysis, it is seen that EVs are potential players for providing the regulation service by minimizing the wear cost of the EV batteries while generating revenue for the EV owners and grid operators.