A Review of Deep Learning and AI Techniques for Enhanced Sun Detection in Solar Power Applications | ||
| ERU Research Journal | ||
| Volume 4, Issue 4, October 2025, Pages 3314-3335 PDF (628.17 K) | ||
| Document Type: Review article | ||
| DOI: 10.21608/erurj.2025.336928.1205 | ||
| Authors | ||
| Alaa Eesa* ; Wafaa Mostafa Ibrahim | ||
| Mechatronics Department, Faculty of Engineering, Egyptian Russian University, Cairo, Egypt | ||
| Abstract | ||
| This review article explores the evolution and recent innovations in solar tracking systems, with a specific focus on the integration of deep learning and artificial intelligence (AI) techniques for enhanced sun detection. The paper highlights the strategic importance of renewable energy, particularly in Egypt, and discusses Concentrated Solar Power (CSP) systems especially Parabolic Trough Systems (PTS) as a sustainable energy solution. Traditional and modern solar tracking techniques are critically analysed, showing how AI-driven methods significantly improve tracking accuracy, system adaptability, and energy efficiency across various environmental conditions. Key findings demonstrate that deep learning approaches, such as Convolutional Neural Networks (CNNs) and ANFIS, can boost tracking precision by over 90%, reduce error margins to under 0.1°, and improve energy yield by up to 76% in dual-axis systems. The review concludes that the integration of AI and machine vision not only enhances the reliability and scalability of solar tracking systems but also opens new pathways for autonomous, low-cost, and high-performance renewable energy applications. | ||
| Keywords | ||
| Solar tracking systems; computer vision; machine learning algorithms; concentrated solar power; renewable energy | ||
|
Statistics Article View: 5 PDF Download: 3 |
||
