DESIGN AND IMPLEMENTATION OF ARTIFICIAL INTELLIGENCE BASED MAXIMUM POWER POINT TRACKING SYSTEM FOR PHOTOVOLTAIC SYSTEM USING METEOROLOGICAL DATA

The performance of photovoltaic (PV) systems is highly affected by the current-voltage (I-V) characteristics of a solar cell, which changes with sun radiation, temperature, and load conditions. Maximum power point tracking (MPPT) algorithms are vital in extracting power from PV systems under such conditions. Therefore, this study proposed an improved artificial neural network (ANN)--based MPPT controller trained using the chicken swarm optimization (CSO) algorithm to overcome these problems. This CSO uses meteorological data for the improved MPPT in varying environmental conditions. A 100W PV system was simulated in MATLAB/Simulink, and the CSO method was compared with Perturb and Observe (P&O). The result shows that the CSO-trained ANN achieves better tracking even under partial shadowing. More so, the use of hardware such as Arduino Uno interfaced with a DC-DC boost converter affirms the practical feasibility of the project. This research contributes to AI-assisted MPPT, offering a better alternative to classical MPPT methods. Future work could explore hybrid optimization techniques.