最大功率点跟踪控制器的优化以提升太阳能光伏系统性能

Abstract Solar energy is a key renewable source, but the low efficiency of photovoltaic (PV) systems limits energy extraction. To maximize power, a maximum power point tracking (MPPT) controller is essential. This manuscript proposes an optimized MPPT method to enhance solar PV system performance. The proposed technique integrates the capabilities of Hippopotamus Optimization (HO) and Convolutional kolmogorov–arnold networks (CKAN), forming a model termed HO-CKAN. The primary intent of this work is to lower the tracking errors, improve the power generation efficiency, and reduce power loss on solar PV systems. The HO algorithm optimizes MPPT performance under varying conditions, while CKAN predicts PV power accurately, enabling adaptive control for improved energy extraction and system stability. The proposed HO–CKAN is implemented in MATLAB and compared with Genetic Algorithms (GA), particle swarm optimization (PSO), and Adaptive salp swarm algorithm (ASSA). The proposed technique establishes the highest average power output of 3.4 × 10 4 W, total energy harvested of 3.61 × 10 4 J, and average load power of 108 W, outperforming existing approaches in terms of power delivery and energy efficiency. The proposed HO-CKAN approach effectively optimizes MPPT performance, enhancing power generation efficiency and stability in solar PV systems compared to existing methods.