Control and Power Management of a Battery-Supercapacitor-PV-Wind Based Grid-Independent DC-Microgrid

Renewable energy sources (RESs), particularly wind and solar, are among the very promising and widely utilized options. However, these RESs are inherently intermittent and unpredictable, which makes it challenging to rely on them as a long-term power source, lacking hybrid energy storage systems (HESSs). To address these challenges, this paper presents a novel power management strategy (PMS) for sharing power between battery and supercapacitor energy storage systems. This approach seeks to mitigate the gap concerning power demand and generation while maintaining stable regulation of DC bus voltage (DBV). To regulate the DBV and balance the disparity between demand and generation, we present a conventional control concept enhanced by intelligent strategies to compute control gains. A hybrid artificial neural network (ANN) constructed proportional-integral (PI) gain tuner has been developed to optimally estimate the gain parameters by minimizing error deviations using acquired datasets. This ANN-optimized strategy generates control pulses to efficiently manage the DBV in the counted DC Microgrid (DCMG). Moreover, we have explored the efficacy of this proposed control strategy against conventionally tuned strategies, focusing on performance metrics such as voltage overshoot and settling time. The hardware-in-the-loop (HIL) HESS-PV-Wind-based grid-independent DCMG's functionality has been validated using an OPAL-RT® real-time simulator within the MATLAB/Simulink® environment across various scenarios.