Variable Step-Size ANN-Based MPPT Controller for Standalone PV Systems under Varying Climatic Conditions

The large-scale consumption of oil worldwide, combined with concerns about depletion, prompts investment in renewable energy sources to meet rising demand and mitigate the environmental impacts of fossil fuels. There are many renewable energy technologies, among which solar photovoltaic systems are particularly prominent. However, solar photovoltaic systems have some drawbacks due to intermittent irradiance, which generates volatile power output requiring cutting-edge technologies for effective utilization. In addition, conventional maximum power point tracking techniques often suffer from slow response and power oscillation that reduce energy yield. In this regard, an attempt is made to propose a technique that mitigates the aforementioned drawbacks in PV systems. For Maximum Power Point Tracking (MPPT), a Variable Step Size Artificial Neural Network Control (VSS-ANN) is used on solar PV systems in different scenarios and two meteorological experimental cases. The system is standalone, and the operating conditions are online. A comparative study has been made between the presented approach and other methods, such as Sliding Mode Control (SMC) and Perturb and Observe (P&O). The VSS-ANN approach ensures an advanced control framework to obtain the desired operation of the photovoltaic system in terms of limited fluctuations and stable operation by efficiently handling the maximum power point in a short time.