An intellectual hybrid algorithm for optimal monitoring of power point in solar photovoltaic energy system-based grid network

Abstract

ABSTRACT Extraction of electrical energy through a photovoltaic (PV) system plays a vital role in this booming digital era. Maximum power tracking efficiency of PV becomes crucial on grid-connected power distribution system due to non-linearity attributes and environmental changes. To address this, an intellectual hybMaximum Power Point Tracking (MPPT) technique is outlined in this paper. The combination of Extreme Learning Machine (ELM) with BAT optimization algorithm frames the hybrid efficient reliable BATs (HERBS) technique in order to trade-off maximum power and time accuracy. HERBS algorithm achieves global peak within minimal time on PV arrays by optimal detection of tracking efficiency. The photovoltaic framework design includes the DC-DC converter under resistive load is reproduced in MATLAB/Simulink. The proposed work is experimentally validated using ARM processor and also simulated to observe the peak power point under half shading condition. Performance of HERBS improved the tracking efficiency in an average of 98.9%, 3% more than PSO-ANN, and 7% than GA-ANN conventional methods. Moreover, the speed convergence achieved by HERBS is less than 0.5 seconds.