A Machine Learning Approach to Boost the Vehicle-2-Grid Scheduling

TL;DR

This paper presents a machine learning approach to optimize the charging and discharging schedules of electric vehicles acting as grid energy storage, enabling efficient, adaptable, and scalable capacity services to the power grid.

Contribution

It introduces a data-driven method for EV scheduling that achieves near-optimal performance with faster computation compared to traditional optimization techniques.

Findings

Comparable objective values to CPLEX and dynamic programming for large EV fleets.

Significantly shorter run times enabling frequent re-optimization.

Effective in providing capacity-related grid services.

Abstract

Electric Vehicles (EVs) are emerging as battery energy storage systems (BESSs) of increasing importance for different power grid services. However, the unique characteristics of EVs makes them more difficult to operate than dedicated BESSs. In this work, we apply a data-driven learning approach to leverage EVs as a BESS to provide capacity-related services to the grid. The approach uses machine learning to predict how to charge and discharge EVs while satisfying their operational constraints. As a paradigm application, we use flexible energy commercialization in the wholesale markets, but the approach can be applied to a broader range of capacity-related grid services. We evaluate the proposed approach numerically and show that when the number of EVs is large, we can obtain comparable objective values to CPLEX and approximate dynamic programming, but with shorter run times. These reduced run times are important because they allow us to (re)optimize frequently to adapt to the time-varying system conditions.