<![CDATA[This research describes a simple modeling technique for Maximum Power Point Tracking based on Artificial Neural Network (MPPT-based ANN) for photovoltaic (PV) systems. The proposed ANN model utilizes a feed-forward backpropagation architecture. The PV system was developed and tested in a simulation environment under uniform irradiation levels of 1000 W/m², 800 W/m², and 600 W/m², and rapidly varying irradiation changes. The simulation results demonstrate that the MPPT-based ANN accurately tracks the MPP, achieving stable power outputs of 98.36 W, 79 W, and 57.45 W, respectively. Although the system experiences initial transient oscillations during the tracking phase, it stabilizes within 80 milliseconds, showcasing rapid convergence and high steady-state accuracy. Under dynamic conditions, the MPPT-based ANN adapts effectively to fast-changing irradiation, restarting the algorithm to track and maintain the system at the updated MPP accurately. These results highlight the reliability, adaptability, and suitability of the MPPT-based ANN for real-time applications in dynamic environments. Nonetheless, further improvements to the ANN model are suggested to minimize transient oscillations and enhance overall performance.]]>
Copyrights © 2025
