Design of Modified Droop Controller and Aggregator for Frequency Support in Microgrid using Electric Vehicles

Abstract

Residential, commercial, and industrial sector’s dependency on electricity is increasing day by day. However, the generated electricity provided to them is limited which have imposed a lot of burden on existing grids. This increasing load needs to be managed in an efficient manner so as to fulfil the growing demands of electricity of the end users in an uninterrupted, and efficient manner. Hence, rapid demand growth has led to the emergence of distributed energy sources (DES) such as-solar, wind etc. These DES when connected to a low voltage distribution network forms a microgrid (MG). However, the combination of inertial and non-inertial DES leads to various frequency fluctuations in MG. To cater frequency fluctuations at MG, droop control has been extensively used by academia and industry. This controller generates series of reference signals for various DES based on their respective maximum capacities. The integration of varieties of DES in MG is still debatable amongst researchers. On the same line, Electric Vehicles (EVs) are considered the future for reliable DES, since they have rechargeable batteries. However, due to their random arrival and departure patterns, their capacity varies rapidly. Thus, conventional droop controller cannot dispatch reference signals to EVs. Hence, the need arises to feed back their instantaneous information to droop controller so that they can be utilized in frequency support. For this purpose, conventional droop control has been integrated with a feedback mechanism and is referred as modified droop controller (MDC) in the proposed scheme. With this feedback mechanism, MDC would be now able to dispatch reference signals to both diesel generator and EVs. However, for meeting reference signal given to fleet of EVs, an aggregator (AG) is proposed. AG can alter the charging and discharging rates of EVs’ thereby satisfying EVs’ needs in terms of battery’s state of charge (SoC). Extensive simulations have been done using the proposed scheme on data available from Santa Rita Jails’ MG