Intelligent control strategy for grid-integrated PV systems with enhanced dynamic performance

With the increasing integration of solar photovoltaic (PV) systems into modern power grids, grid stability and power quality have become a critical challenge due to environmental variability and non-linear load dynamics. To deal with these problems, this research proposes a novel control strategy by incorporating Deep Attention Dilated Residual Convolutional Neural Network (DADRCNN) with Hippopotamus Optimization Algorithm (HOA) to optimally manage energy in a grid-connected photovoltaic system. The key goal of this study is to mitigate power quality issues like harmonic distortions, voltage fluctuations, and load imbalances, thereby improving the power system's overall performance and stability. To achieve this, DADRCNN precisely tracks the maximum power point, while HOA optimizes the converter's duty cycle to ensure effective control. The proposed control technique is simulated in MATLAB and evaluated against existing strategies. Findings demonstrate that the proposed strategy achieves a high accuracy of 99.84 %, the lowest total harmonic distortion of 1.08 %, and a low computation time of 1.05 s, outperforming existing models. In addition, statistical analysis confirms the robustness and reliability of the proposed technique, indicating its practical applicability for PV energy conversion in smart grid environments.