Implementation of neural network controller for grid connected wind-solar PV charging station

Résumé fourni par la source

The main goal of this project is to use an advanced neural network controller to create an electric vehicle (EV) charging station that is connected to the grid and uses renewable energy. Renewable energy sources that don&s;t produce hazardous waste or pollutants, such as solar, wind, hydrogen, biogas, and tidal energy, are used to generate electricity. In a similar vein, EVs don&s;t produce greenhouse gasses, which drastically lowers carbon emissions and promotes environmental sustainability. Power parameters are often controlled using a proportional-integral (PI) controller in traditional systems. Power quality is impacted by PI controllers&s; poor speed response and excessive harmonic distortions (THD). In order to overcome these drawbacks, this study suggests substituting an Artificial Neural Network (ANN)-based controller for the PI controller. This improves system performance by reducing mistakes and increasing reaction time. Better power quality and a considerable reduction in harmonic distortion are guaranteed by the suggested ANN controller. Simulations in MATLAB/Simulink 2018a are used to examine the system&s;s performance and compare the ANN-based and conventional PI-based approaches. The outcomes show a significant decrease in Total Harmonic Distortion (THD) from 1.20% (PI) to 0.51% (ANN), under scoring the suggested control strategy&s;s superior efficiency.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.