A Hybrid Model of Modified Robust Linear Regression Optimized by Ant Colony Optimization for Photovoltaic System Efficiency Improvement Under Sudden Change of Environmental Conditions

• Published in: IECON 2021 – 47th Annual Conference of the IEEE Industrial Electronics Society pp. 1 - 6
• Main Authors: Nkambule, Mpho Sam, Hasan, Ali N., Ali, Ahmed
• Format: Conference Proceeding
• Language: English
• Published: IEEE 13.10.2021
• Subjects: Ant colony optimization; Ant Colony Optimization (ACO); Heuristic algorithms; Linear regression; Machine learning algorithms; Maximum power point trackers; Maximum Power Point Tracking (MPPT); Modified Robust Linear Regression (MRLR); Partial Shading Conditions (PSC); Photovoltaic (PV); Photovoltaic systems; Training
• ISSN: 2577-1647
• DOI: 10.1109/IECON48115.2021.9589195

In recent years, the Photovoltaic (PV) system has notably progressed towards achieving the incessant growth in energy demand. The motivation being a cleaner source of energy and growth of its technology. Nonetheless, the low PV panel efficiency conversions remain as a challenge under dynamic weather conditions. Therefore, it is indispensable to appoint maximum power point tracking (MPPT) controller to optimize a high efficiency out of a PV module. The traditional conventional MPPT algorithms demonstrated the capabilities of tracking maximum power point (MPP) under uniform insolation levels. However, under partially shaded conditions (PSC) dismally failed to locate a real Global MPP. The metaheuristic and machine learning (ML) MPPT based techniques have been introduced to harvest MPP, but the response time and convergence speed remain a challenge. This paper introduces the hybrid system that incorporates the Modified Robust Linear Regression (MRLR) and Ant Colony Optimization (ACO) to solve PV module optimization problems with rapid response time and excellent convergence speed under PSC. The simulated results proved that the proposed hybrid MPPT system is more effective and accurate under PSC with respect to single approach.