Analysis of a Traditional and a Fuzzy Logic Enhanced Perturb and Observe Algorithm for the MPPT of a Photovoltaic System

Abstract

This paper presents the results obtained for the maximum power point tracking (MPPT) technique applied to a photovoltaic (PV) system, composed of five solar panels in series using two different methodologies. First, we considered a traditional Perturb and Observe (P&O) algorithm and in a second stage we applied a Fuzzy Logic Controller (FLC) that uses fuzzy logic concepts to improve the traditional P&O; both were implemented in a boost converter. The main aim of this paper is to study if an artificial intelligence (AI) based MPPT method, can be more efficient, stable and adaptable than a traditional MPPT method, in varying environment conditions, namely solar irradiation and/or environment temperature and also to analyze their behaviour in steady state conditions. The proposed FLC with a rule base collection of 25 rules outperformed the controller using the traditional P&O algorithm due to its adaptative step size, enabling the FLC to adapt the PV system faster to changing environment conditions, guessing the correct maximum power point (MPP) faster and achieving lower oscillations in steady state conditions, leading to higher generated energy due to lower losses both in steady state and dynamic environment conditions. The simulations in this study were performed using MATLAB (Version 2018)/Simulink.