Development of a new Mathematical Optimization Technique for Enhancing MPPT Performance in Photovoltaic Systems

The Maximum Power Point Tracking (MPPT) algorithms monitor and operate photovoltaic (PV) systems under various environmental conditions, ensuring maximum power output and energy generation. Bio-inspired and hybrid techniques have improved engineering applications, but their use in MPPT for PV systems is being investigated due to poor energy harvesting in dynamic conditions. A novel mathematical optimization, Intelligent Elephant Herding Optimization (IEHO) approach is developed to enhance MPPT performance in solar systems by optimizing the classification of dynamic long short-term memory (DLSTM) parameters. The dynamic LSTM to predict the MPP level of the solar PV system under varying environmental conditions. To collect solar irradiance, temperature, voltage, panel tilt angle, azimuth angle, current, and weather conditions, from the solar PV system, those parameters are cleaned by using the Box-Cox transformation and then characterize the MPP level. The proposed mathematical model IEHO method has been evaluated in MATLAB/Simulink with various operating circumstances, including temperature, grid connection/disconnection, voltage fluctuations, solar irradiation, different PV module types, and partial shade. The suggested mathematical model was compared with existing optimization methods. The proposed mathematical model greatly enhanced MPPT performance, providing more energy efficiency and stability than previous optimization strategies across a variety of operating situations. The results show that IEHO, when paired with dynamic LSTM predictions, significantly improves PV system power production and efficiency.