Optimal Charging Management of Microgrid-Integrated Electric Vehicles

Introduction

The constant growth of Renewable Energy Systems (RESs) as a sustainable replacement for carbon-intensive fossil fuels for energy production is a key priority at the international level. Typically, RESs are intermittent and geographically distributed energy sources. Moreover, RESs enable the transition of the traditional electricity consumers to become prosumers actively engaged in energy markets through renewable production, storage and demand-side management. Additionally, new actors such as aggregators are present and electric vehicles (EVs) may represent an additional, intermittent and distributed load. In this framework, smart grid management and optimization technology solutions must become reality in order to guarantee the operation of the grid and the satisfaction of all connected users. To this end, Energy Management Systems (EMSs) based on optimization models are currently considered essential tools for the optimal management of an electrical grid characterized by different actors, distributed production, flexible loads, storage systems and EVs. Microgrids can be seen as a distribution grid at a small scale, which requires EMSs for the optimal management of production plants, storage systems, flexible loads, renewables, etc. When also EVs are present, the complexity of the optimization problem arises, and it is necessary to find new mathematical models and solution approaches. In fact, EVs’ integration is particularly important to reduce greenhouse gas emissions in cities. Emissions can be reduced by using EVs when renewable resources are used to produce feeding the vehicles. As confirmed by the authors in (Ferrero et al., 2016; Shareef et al., 2016), mass distribution of EVs may be a good solution, but, unfortunately, their wide usage may have important drawbacks and cause technical problems. As an example, the power grid can be harmfully affected by uncontrolled charging, long charging times and interruptions and can become unstable for distributed and intermittent loads (Qian et al., 2015; Sbordone et al., 2015). Thus, it is necessary to define planning and management strategies that are able to ensure the operation of the grid and to satisfy EVs demand. Moreover, it is necessary to consider new technologies, such as Vehicle-to-Grid (V2G), Smart Charging (SC) and Vehicle-to-Building (V2B), allow the injection of power by EVs into the electrical grid and/or the modulation of the power during the charging process.

Finally, it should be considered that EVs could be charged in a much longer time because the charging station is not able to manage multiple vehicles through optimized power management strategies, thus creating long queues and users’ dissatisfaction.

In this chapter, attention is focused on the optimal management of microgrids that include EVs and charging stations with multiple sockets.

In the literature, discrete-time decision models are used by several approaches that aim at the optimal scheduling of EVs and result in problems difficult to be solved, especially due to a large number of variables. Then, also heuristics and metaheuristics have been applied as well as decision architectures based on decentralized optimization. In the following, a discrete-event approach is used for the definition of an optimization model for the integration of EVs in the electrical grid. The approach is particularly advantageous because it allows reducing the number of variables and computational time. The considered case study includes a microgrid with a multiple sockets charging service station, photovoltaics, storage system, and a connection to the external grid. The resulting optimization model is (nonlinear) Mixed Integer Problem (MIP). This serious complication, with respect to the case considering the charge of one single EV, can still be managed when problem instances have moderate size, like those usually solved in real cases (perhaps in real-time). Anyway, the discrete-event approach counterbalances this complication, by considering a much lower number of time intervals, and consequently a significant reduction of the optimization problem size.

In summary, the contribution of the chapter are:

• •

  A description of the role of optimization for EVs’ scheduling in microgrids;

• •

  A discrete event optimization model for multiple sockets charging stations;

• •

  The application to a case study and discussion of results.

This chapter is organized as follows: