An advanced MPPT scheme for standalone solar powered PV system using neurofuzzy estimator based on measured data

Abstract

An innovative technique for implementing the maximum power point tracking system for the Photo-Voltaic panel is proposed in this paper. The system comprises of a Neural Network Estimator (NNE), followed by a conversion coefficient and a calculation stage of the optimal duty cycle. The proposed NNE technique, implemented in MATLAB / Simulink, calculates the ratio of open circuit voltage corresponding to each solar radiation for various value of temperature to the corresponding standard open circuit voltage. A regularization coefficient is determined, which estimates the voltage corresponding to the maximum power directly from the open circuit voltage for each solar radiation. Finally, from the input/output equation of boost converter, the optimal duty cycle is evaluated. Simulation results have been used to assess the system performance for various situations obtained by using an existing photovoltaic model and real weather data, and relevant comparisons are also done to substantiate the performance advantages of the method. The Simulation results demonstrate that the optimization of the P&O MPPT control with a NN algorithm provides better results and performance in terms of accuracy and complexity. The results support the effectiveness and performance of using NNE-based MPPT controller approach. It is demonstrated that this controller can achieve almost 99% of the real PVP maximum power.