Enhancing Solar Panel Efficiency with Adaptive Solar Reflectors Using ANFIS and Real- Time Environmental Control

Abstract

ABSTRACT The increasing demand for efficient and sustainable solar energy solutions has led to the development of advanced tracking systems that enhance photovoltaic (PV) panel performance. This study presents the Adaptive Solar Reflector (ASR) system, an intelligent, machine-learning-based solar optimization system designed to improve solar energy capture by dynamically adjusting the tilt and curvature of reflective mirrors. The ASR system integrates an Adaptive Neuro-Fuzzy Inference System (ANFIS) for real-time decision-making and an ESP32-based Internet of Things (IoT) module for remote monitoring and control. The system was specifically designed and tested under Sri Lankan environmental conditions, where high solar irradiance levels and variable wind speeds influence energy generation. The primary objectives of this research included enhancing ASR performance, analysing cost-effectiveness, implementing a real-time adaptive tracking mechanism, and ensuring structural durability while maximizing solar efficiency. The system was evaluated through experimental testing and comparative analysis against fixed solar panels. Results indicated that the ASR system improved energy capture by 20-30% while effectively adapting to environmental conditions, such as shading and wind disturbances. The ANFIS model demonstrated high prediction accuracy, with Servo 2 and Servo 3 achieving 100% classification accuracy, while Servo 1 exhibited 80% accuracy, suggesting areas for further refinement. The IoT-based monitoring system achieved a 97% data transmission success rate, enabling real-time remote control and performance analysis. Despite minor challenges, including a 20% power reduction under strong wind conditions, the ASR system proved to be a highly effective and scalable solution for solar power optimization. Future enhancements will focus on advanced wind adaptation algorithms, deep-learning-based servo control improvements, and large-scale field testing in diverse climatic regions. The findings from this research contribute to the ongoing advancements in intelligent renewable energy systems, paving the way for more efficient and adaptive solar energy solutions.