Comparison of the Adaptive Neural-Fuzzy Interface System (ANFIS) based Solar Maximum Power Point Tracking (MPPT) with other Solar MPPT Methods

Abstract

Maximum power point tracking (MPPT) is a desirable factor in the Photovoltaic (PV) systems and is used to increase the extracted power of the PV system. There are many existing conventional techniques such as Perturb & Observe (P&O), Hill Climb (HC) and Incremental Conductance (IC) for the MPPT in PV systems. However, recently with inclusion of the artificial intelligence (AI) based techniques, the MPPT has become more efficient. This work presents the MPPT in PV system by an artificial intelligence technique called as Adaptive Neural-Fuzzy interface system (ANFIS). ANFIS is considered more accurate because of fast response as this technique is integration of Artificial Neural Network (ANN) and Fuzzy Logic Controller (FLC). ANFIS uses the selectivity of FLC and Training of ANN in order to achieve the MPPT in PV systems which results in good computation and robust response. In this work, an ANFIS based MPPT system has been designed and compared in MATLAB/SIMULINK with other MPPT techniques. The comparison results verify that ANFIS based MPPT outperforms than ANN and FLC in terms of convergence time and output power under fixed as well as varying solar irradiance. Moreover, it is worth mentioning that the convergence time is ten times less in ANFIS method than FLC, IC and P&O techniques which results in less chance of error and accurate tracking of maximum power point (MPP).