Design of DC-DC Buck Converter for Battery Charging Application Using Neural Network Based CC and CV Method

The DC-DC Buck converter is a widely used power electronic converter topology for voltage reduction applications. Traditional control methods for achieving Constant Current (CC) /Constant Voltage (CV) regulation often rely on complex feedback loops and parameter tuning, making them susceptible to varies in load conditions and component variations. This project presents a unique approach to control the buck converter in CC/ CV mode using Neural Network (NN) - based controller. In this study, the NN architecture is designed and trained to predict the duty cycle required to maintain a constant output current, thus eliminating the need for conventional feedback control loops. Practically proved the outcome valued the efficient of the NN-based controller in maintaining a constant output current/constant output voltage, even under dynamic load changes and uncertainties. The comparative analysis with traditional control methods highlights the superior performance and adaptability of the proposed approach. Overall, this research contributes to the advancement of DC-DC buck converter control techniques, offering a promising avenue for efficient and reliable constant current/constant voltage regulation in various power electronics applications.