Optimized Solar EV Charging System Using Monarch Butterfly MPPT and Neural Network Control
TL;DR
The goal of this work is to maximize amount of electrical energy produced by solar panels in the context of fluctuating temperature and irradiance by developing an MPPT regulating model called Monarch Butterfly Optimized—Recurrent Neural Network (RNN).
Abstract
This study presents an Electric Vehicle Charging System (EVCS) based on a DC microgrid consisting of a grid, battery, and solar photovoltaic system (SPVS). This paper's innovative contribution is the development of a novel converter and Maximum Power Point Tracking (MPPT) approach to augment the PV‐EV system's overall power tracking performance, voltage gain, power quality, and efficiency. The goal of this work is to maximize amount of electrical energy produced by solar panels in the context of fluctuating temperature and irradiance by developing an MPPT regulating model called Monarch Butterfly Optimized (MBO)—Recurrent Neural Network (RNN). This work successfully increases PV output by using a cutting‐edge Trans Z‐Source Coupled Inductor Boost Converter (TZSCIBC), which boosts voltage‐gain efficiency with less switching stress and power loss. The effective conversion and transfer of electrical energy between the EV battery and the grid system is the responsibility of a bidirectional DC‐DC converter, which gets regulated with the assistance of an Artificial Neural Network (ANN) controller. Additionally, a variety of performance metrics, such as overall conversion efficiency (96.66%) and tracking accuracy (99.95%), are obtained from MATLAB/Simulink and experimental analysis of the proposed regulating model and converter.