Predictive management of batteries in microgrids equipped with electric vehicles

Abstract

In recent years, the penetration of renewable energy sources into the power distribution system has increased significantly. Power generation has evolved from centralized power plants towards a decentralized structure. In this context, microgrids, i.e., small-scale power distribution grids, are being deployed to support distributed generation. Microgrids encounter different challenges as they have to be resilient, reliable, robust, and capable of ensuring power quality and handling intermittent energy sources and new energy usages. This paper focuses on the predictive management of microgrids equipped with a bank of batteries and a fleet of electric vehicles. The reference strategy is rule-based. Two model predictive control (MPC) based strategies - in the first one, all the batteries are managed as a unique (fictitious) battery whereas in the second one, those batteries are managed independently – are discussed and evaluated in simulation. First, the results highlight the benefits of using a predictive strategy when it comes to efficiently manage electricity storage and release in microgrids equipped with electric vehicles. As an interesting result, the strategy proposed to predictively manage the batteries as a unique (fictitious) battery is a litte bit less efficient regarding economical cost and carbon dioxide emissions reduction than the predictive strategy intending to manage the batteries independently but computation time is significantly lower with a large fleet of electric vehicles.