Design of LLC Resonant Converter for EV Charging Using Neural Network-Based PI Controller

The proposed work presents the design and modeling of an LLC converter for electric vehicle battery charging applications, specifically using a 12 V lead-acid battery output. The LLC converter has opted for its soft switching characteristics, ability to produce smooth waveforms, high power density, low electromagnetic interference, and high efficiency, which are desirable for electric vehicle battery charging. The design process includes establishing the suitable transformer turns ratio and selecting resonant tank components such as the resonant inductor, resonant capacitor, and magnetizing inductor. These choices are critical for achieving optimal performance in the LLC converter. Furthermore, this work explores the implementation of closed loop control using a neural network controller. The advantages of neural networks, such as improved control performance and adaptability, are discussed. MATLAB/SIMULINK software is utilized to simulate the LLC converter, and the results demonstrate the benefits of employing a neural network-based controller.