Artificial neural network-optimized bridgeless Landsman converter for enhanced power factor correction in electric vehicle applications

Abstract

Electric vehicles (EVs) are gaining popularity globally due to their energy-efficient battery storage systems, low carbon emissions, and eco-friendly operation. By transforming both the transportation and electrical sectors, EVs could create a synergistic relationship that reduces fossil fuel use and improves renewable energy integration. However, this convergence emphasizes the necessity for appropriate power factor correction (PFC) methods, especially in EV battery charging systems, to alleviate supply-end PQ concerns. Use of a bridgeless Landsman converter (BLC), noted for its efficiency and link voltage monitoring, is innovative in this research. A proportional-integral (PI) controller tuned by an artificial neural network (ANN) improves prediction and classification, especially response time. The ANN-based PI controller optimises system performance in real time using adaptive control. Using a hysteresis controller attached to a pulse width modulation (PWM) generator regulates the converter's steady-state switching frequency for accurate and consistent output. The proposed approach reduces harmonic distortions and improves operating efficiency. This comprehensive architecture improves power factor and addresses significant PQ concerns in EV charging infrastructure. Integrating improved control tactics and converter design shows that this approach may support electric car technology developments. MATLAB simulations show that power factor correction (PFC) charges EV batteries quickly and effectively. Findings suggest the technique could increase power quality, system efficiency, and EV uptake.