An Adaptive Duty-Cycle Methodology for PV Power Maximization Using a Single Variable

Abstract

This paper presents a new methodology to maximize the power output of Photovoltaic panels (PV), based on an adaptive duty-cycle methodology. The approach embeds the DC/DC converter characteristic in the cost function, allowing an optimization based on a single measured variable. Two cost functions, and respective learning rules, are derived. The first, more complex and comprehensive, traces the ground for the second which is less computational intensive and solves stability issues and implementation difficulties. It is also demonstrated that the system is asymptotically stable around the optimum duty-cycle, in the Lyapunov sense. Both methods are compared through simulations and deviations from the optimal solution are assessed.