An Improved MPPT Approach Using Artificial Neural Network for PV Grid System

Nowadays, in transportation systems, the clean energy aspect of solar photovoltaic (PV) energy is becoming more and more popular. However, the nonlinear environmental dependence of solar PV is its main drawback. Voltage management and effective Maximum Power Point Tracking (MPPT) techniques are essential to maximize the power produced from PV systems. This proposed work aims to integrate an Artificial Neural Network (ANN) based MPPT for PV-tied grid systems with a Boost converter. The fluctuating DC voltage from the PV panels is converted by the proposed Boost converter into a stable and appropriate voltage level for grid integration with high efficiency and low Total Harmonic Distortion (THD). Additionally, this work uses the ANN-based MPPT technique to track the PV system's optimal power, leading to better tracking accuracy and faster convergence. The single phase VSI converts the DC input to AC output for power supply to the grid system with the aid of a PI controller. The MATLAB/Simulink is used to implement the entire proposed system, and a comparison is made with the existing topologies (MPPT, P&O Based MPPT) to demonstrate the significance of the implemented work.