Hybrid Operational Strategy for Power Quality Enhancement in Photovoltaic System With H-Bridge Multilevel Cascaded Inverter

Abstract

This research proposes a hybrid control approach to enhance power quality in a grid-connected photovoltaic (PV) system under nonlinear load conditions. Conventional power quality improvement techniques often exhibit slow convergence and may become trapped in local optima. To address these limitations, a hybrid vision transformer (HViT) combined with a bidirectional long short-term memory (BiLSTM) network is employed alongside a Clan Co-operative based Bluefin Trevally Optimizer (CCBTO). The HViT extracts deep features from nonlinear load harmonic currents, while the BiLSTM captures sequential dependencies in power regulation for improved control accuracy. The CCBTO is applied to optimize system performance, ensuring maximum power output, reduced switching complexity in the inverter, and improved operational efficiency. The proposed framework enhances maximum power point tracking (MPPT) under varying solar irradiance conditions, mitigates harmonic distortions, and improves power factor through dynamic reactive power compensation. MATLAB simulation results demonstrate that the proposed hybrid model significantly improves MPPT efficiency, reduces total harmonic distortion, and enhances overall power factor compared to conventional methods. The results confirm that the integrated HViT–BiLSTM–CCBTO approach provides a robust and adaptive solution for maintaining high-quality power transfer and stable grid integration in photovoltaic energy systems.