Improving Grid-Tied Electric Vehicle Charging Efficiency with a Neuro-Network-Inspired Intelligent Maximum Power Point Tracking (MPPT) Controller

Abstract

A new solution is being sought for as the demand for power continues to climb throughout the world. As a possible solution, grid-synchronized electric vehicles are being explored. The extensive usage of electric cars and their subsequent integration into the grid causes a multitude of power system challenges, such as imbalances in supply and demand, voltage, and frequency. One solution to these issues is the incorporation of solar power into this interface. A smart grid advancement that enables energy exchange between the grid and the EV is vehicle-to-grid (V2G) technology. The purpose of this project is to use a smart maximum power point tracking (MPPT) controller driven by a neural network to enhance the efficiency of charging grid-connected electric vehicles.In order to get electricity from solar panels to an inverter, a Relift-Luo converter is used. In order to track the solar panels' output, the proposed system employs an innovative method called artificial neural network maximum power point tracking (ANN-MPPT). The proposed system might provide energy management with the aid of a battery system and a bidirectional battery converter. In order to bring STATCOM devices into grid synchronization, this study employs a Recurrent Neural Network (RNN) controller in conjunction with the D-Q theory of transformation. The goal of this research is to find ways to improve power quality using STATCOM and to apply these improvements in a variety of situations. The whole system is validated using a MATLAB simulation 2021a.