Optimizing PID control for maximum power point tracking in photovoltaic systems under variable and partial shading conditions

The development of MPPT algorithms is crucial for optimizing energy utilization in photovoltaic systems. This manuscript introduces a hybrid method to enhance MPPT performance under partial shading conditions by optimizing PID control. The proposed AOA-DWNN method integrates the Aquila Optimization Algorithm (AOA) for tuning PID gain settings and the Dynamic Wavelet Neural Network (DWNN) for predicting optimal converter parameters. This combination improves power tracking accuracy and optimizes energy utilization in varying environmental conditions. The method is put into practice in MATLAB and compared with existing methods, including the Salp Swarm Algorithm and Sine Cosine Algorithm (SSA-SCA), Firefly Algorithm and Particle Swarm Optimization (FA-PSO), and Artificial Bees Colony and Cuckoo Search Algorithm (ABC-CSA). The results demonstrate a significantly lower error rate of 0.89 %, compared to 1.5 %, 1.2 %, and 1 % in existing approaches. Additionally, the proposed technique achieves an efficiency of 99.96 %, surpassing the 98.94 %, 97.95 %, and 96.94 % of other methods. The findings highlight the effectiveness of the AOA-DWNN technique in improving photovoltaic system performance, ensuring more accurate and reliable MPPT operation under partial shading conditions.