Grid-Connected Photovoltaic System for EV Applications

This research discusses about artificial neural net-works, or ANNs, for tracking Maximum Power Points(MPP). Neural Networks(NN) are trained via the error back propagation approach. One benefit of neural networks is their capacity to track the MPP with accuracy and speed. A 2-10-1 neuron structure artificial neural network (ANN) is used in the lab. It was trained using test data from the photovoltaic panel. 25 percent of the data are used for validation and 75 percent for training. This method uses an ANN to determine the reference voltage for the MPP under different environmental factors, such as temperature and sunlight exposure. To track the MPP, the dc-dc boost converter must be correctly controlled. To lower total harmonic distortion (THD), an LCL filter is utilized. MATLAB/SIMULINK is used to obtain simulation results to verify theory analysis.