Enhancing Stability and Performance of Grid-Connected Residential PV Systems With Battery-Super Capacitor Storage Using Advanced Control Techniques

Abstract

The increasing integration of renewable energy technologies poses significant challenges to the power grid due to generation unpredictability. Variations in output, driven by weather uncertainties, highlight the need for effective storage solutions to maintain grid stability and reliability. This research proposes a novel approach for a grid-connected residential photovoltaic (PV) system incorporated with a hybrid energy storage system (HESS) comprising a battery bank and a super capacitor (SC) pack. The novelty of this paper lies in the innovation of the Red Panda Optimization (RPO) and Efficient Predefined Time Adaptive Neural Network (EPTANN). Hence, the method is named RPO-EPTANN. The objective of the proposed method is to enhance stability, reduce voltage overshoot, improve efficiency, and reduce the system's entire cost. The converter's control signal is optimized using the proposed RPO, and the EPTANN predicts the converter's ideal control signal. By then, the proposed approach is put into practice from the MATLAB working platform, and the findings are calculated using the existing process. The proposed strategy outperforms all current approaches in terms of Particle Swarm Optimization (PSO), Artificial Neural Networks (ANN), and Artificial Rabbits Optimization (ARO). The existing methods exhibit total system costs of 27 660$, 29 665$, and 30 025$, whereas the proposed method achieves a significantly lower cost of 24 540$. Efficiency of 85%, 75%, and 62% in the existing approaches are improved to 98% with the proposed method. These findings indicate that the proposed RPO-EPTANN method significantly reduces operational costs while enhancing overall system efficiency. This reflects a substantial advancement in performance, ensuring improved stability, reliability, and energy optimization in grid-connected residential PV systems.