EANFIS-based Maximum Power Point Tracking for Standalone PV System

Abstract

The design and development of eco-friendly renewable energy sources is a critical process in the power generation system. Power generation of photovoltaic system depend on temperature and irradiation. Variation of atmospheric conditions need to find points for every instant on V-I characteristics of PV in which maximum power transfer from source to load is achieved. This work deals with Maximum Power Point Tracking (MPPT) method based on Adaptive Neuro Fuzzy Interference System (EANFIS) in standalone operation. The novelty is introduced in the design of inverter, motor selection, and maximum power point tracking. Quasi-Z-source inverter (qZSI) is designed with Z-shaped impedance network to continuously draw constant current from solar panel. MPPT enhance the efficiency of PV panel via load matching; however, it may be affected by environmental changes. Hence, an EANFIS-based MPPT technique is used in the proposed work to confirm maximum power delivery to current motor. The proposed method is the combination of Particle Swarm Optimization (PSO) and Adaptive Neuro Fuzzy Inference System (ANFIS). Training stage of ANFIS is optimized by PSO to handle switching angle of Multi-Level Inverter (MLI) and generate harmonic-less control voltage, hence named Enhanced ANFIS (EANFIS). Voltage and current control of solar panel decide maximum power generation which is verified using Simulink and practical environment. Thus, EANFIS-based MPPT technique achieved the maximum tracking efficiency of 94% which is better than other comparison methods, namely P&O, RBFNN, ANN, and IDISMC.