Computer vision-based sun tracking control for optimizing photovoltaic power generation

Abstract

As global energy consumption rises and fossil fuel reserves dwindle, the transition to renewable energy sources becomes imperative. Solar photovoltaic (PV) technology, crucial in this shift, faces challenges in efficiency and cost. This study explores a motorized sun-tracking system employing image processing techniques to optimize solar panel orientation and maximize energy capture. Using an Arduino Mega 2560 microcontroller, L298N motor driver, Raspberry Pi 3 Model B, and webcam integration, the system dynamically adjusts solar panels based on real-time sun position detection. Experiments compare the performance of fixed and sun-tracking solar panels, revealing that sun-tracking panels consistently outperform fixed ones, particularly during low sun angles, resulting in up to 84.9% higher power output. These findings underscore the potential of sun-tracking technology to significantly enhance solar energy efficiency and support sustainable energy goals. Future research should focus on refining tracking algorithms and optimizing system design to further boost energy capture and reliability.