An LS-Based Improvement of Conventional MPPT Techniques for Photovoltaic Systems Under Varying Conditions
Year: 2026
Type: conference-paper
Abstract: Although conventional MPPT algorithms, such as Incremental Conductance (INC), Fractional Open Circuit Voltage (FOCV), and Perturb and Observe (P&O), are extensively employed in both scientific research and industrial applications because of their ease of implementation and reduced cost, they exhibit several inherent drawbacks. Indeed, these methods generally exhibit persistent oscillations around the maximum power point, a relatively slow convergence speed and an increased sensitivity to rapid variations in environmental conditions. In this work, an analytical enhancement of these MPPT techniques is proposed through the integration of the least squares (LS) method. For voltage-based method such as "FOCV", The LS approach is partially applied to approximate the power-voltage (P-V) characteristic with a parabolic model in order to estimate the MPP voltage, thereby replacing the fixed empirical relationships commonly employed in conventional schemes. For perturbative and derivative-based methods such as "P&O" and "INC", the full LS formulation is adopted to accurately identify the MPP by fitting the PV characteristic and analytically computing the optimal operating voltage. The estimated MPP voltage is subsequently exploited to directly regulate the duty cycle of the DC–DC boost converter. MATLAB/Simulink-based simulations demonstrate that the proposed LS-based MPPT methods substantially decrease oscillations around the maximum power point, improve convergence speed, enhance tracking accuracy and improve overall efficiency, particularly under varying irradiance and partial shading conditions. Finally, the proposed algorithms effectively overcome the drawbacks of the FOCV, INC, and P&O methods, reaching an efficiency higher than 98%.
Language: English
Field: Energy
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7201879285