Grid-connected solar photovoltaic-supported dynamic voltage restoration for enhanced voltage stability and power quality

Grid-connected PV systems are important in modern power systems as they enhance voltage stability, improve power quality (PQ), and support renewable energy integration under varying solar generation conditions. However, optimizing voltage restorers remains challenging due to reactive power compensation requirements, system stability issues during PV variations, and diverse grid conditions and load profiles. To overcome this problem, this work proposes a grid-connected solar photovoltaic fed quasi-impedance source driven dynamic voltage restorer employing optimized switching strategy for enhanced voltage gain and power quality. The proposed technique is the joint execution of both House Swallow Optimizer (HSO) and Skeleton guided convolutional neural network (SGCNN) and is commonly referred as HSO- SGCNN method. The main objective of the proposed technique is to reduce the Total Harmonic Distortion (THD) and enhance power output quality. The proposed HSO is utilized to optimize THD performance and SGCNN is used for maximizing the power of PV. The proposed technique is simulated in MATLAB and its performance is compared with various existing techniques includes Multi-Objective Bees Algorithm (MOBA), Genetic Algorithm (GA) and Butterfly Optimization Algorithm (BOA). The proposed control and HSO-SGCNN achieve the better performance, with voltage gain exceeding MOBA 16, GA 12, and BOA 9 at modulation index 0.5, reaching over 15 at M = 1, voltage stress reducing to 1.1 pu, and THD minimized to 4 % compared to 8–10 %, ensuring maximum stability analysed to other existing techniques.