Enhanced harmonic reduction and energy optimization in grid-connected photovoltaic-battery systems for improved power quality

Grid-connected photovoltaic (PV)-battery systems combine solar energy with battery storage to guarantee a stable and efficient power supply. However, challenges include the intermittent nature of solar generation, which impacts supply consistency, and the complexity of managing power flows between the grid, photovoltaic system and battery. To address these challenges, this manuscript introduces an innovative method for enhancing energy management (EM) in grid- linked PV battery systems. The major goal of the proposed approach is to minimize Total Harmonic Distortion (THD), reduce energy cost, and develop the system's performance. The proposed strategy utilizes Binary Waterwheel Plant Optimization (BWPO) to control the DC to AC converter and Temporal Inductive Path Neural Network (TIPNN) to predict power consumption and improve power quality; hence, it is named the BWPO-TIPNN technique. This method effectively reduces harmonic distortion, optimizes power flow, and ensures better energy efficiency in hybrid energy systems, ultimately leading to improved cost-effectiveness and system stability. The proposed methodology is assessed and contrasted using the MATLAB platform with other existing methodologies. The proposed technique demonstrates its effectiveness by maintaining a low Levelized Cost of Energy (LCOE) of $0.059 and a Net Present Cost (NPC) of $700,500.84, ensuring improved EM and enhanced power quality (PQ) in grid-connected PV-battery systems. Additionally, it achieves a THD of 1.3 % and a Root mean square error (RMSE) of 8.31643, highlighting its superiority over existing techniques. These results demonstrate that the proposed technique offers significant improvements in power quality, cost-effectiveness, and overall system performance. • A BWPO-TIPNN method is proposed for grid-connected PV-battery energy management. • BWPO optimizes converter control to reduce THD and improve power quality. • TIPNN predicts power demand for efficient scheduling and energy balancing. • Method achieves lower LCOE, NPC, and improved cost-effectiveness. • Proposed Approach has high efficiency with reduced error and improved stability.