Soft Computing-Based Optimal Solar Tracking and MPPT

This study focuses on improving the efficiency of solar energy systems by addressing the challenges posed by fluctuating solar radiation and environmental conditions. The research employs machine learning algorithms, specifically Feedforward Neural Networks (FNN) and Random Forest (RF), to optimize the tilt angles of solar panels for maximum energy capture. The study utilizes data from the National Renewable Energy Laboratory (NREL) to train and test the models. The FNN model demonstrated higher accuracy in predicting tilt angles but required more computational time, while the RF model offered a balance between accuracy and computational efficiency. The integration of these algorithms with Maximum Power Point Tracking (MPPT) using the Perturb and Observe (P&O) algorithm further enhances the system's performance. The study concludes that combining machine learning for tilt optimization with MPPT for power maximization significantly improves solar energy system efficiency. Future research should focus on location-specific datasets and advanced ML techniques to adapt to rapidly changing environments.