Enhanced energy management of DC microgrid: Artificial neural networks-driven hybrid energy storage system with integration of bidirectional DC-DC converter

Abstract

Standalone microgrids using Photovoltaic (PV) systems might be a feasible alternative for powering off-grid populations. However, this form of application necessitates the use of energy storage systems (ESS) to control the intermittent nature of PV production. This paper proposes a novel energy management strategy (EMS) based on Artificial Neural Network (ANN) for controlling a DC microgrid using a hybrid energy storage system (HESS). The HESS connects to the DC Microgrid using a bidirectional converter (BC), that enables energy exchange between the battery and supercapacitor (SC). The effectiveness of the proposed approach in relation to the traditional control method is based on performance metrics such as overshoot and settling time. The proposed ANN control attains 18 % overshoot while step increases in PV generation and attains 21 % overshoot while decreasing PV generation. The results show that the proposed control method ensures continuous power supply from the PV system by optimising load balancing under various operating conditions, maintaining state-of-charge (SoC), and improving the voltage profile.