Optimizing Power Electronics in Electric Vehicle Systems Through Neural Network-Based Control

EasyChair Preprint no. 12717

Abstract

This paper presents a novel approach to optimizing power electronics in electric vehicle (EV) systems through neural network-based control techniques. As EV technology continues to evolve, the demand for efficient and reliable power electronics becomes increasingly critical. Traditional control methods often struggle to address the dynamic and nonlinear characteristics inherent in EV systems. In contrast, neural network-based control offers a promising solution by providing adaptive and robust control strategies. The introduction provides an overview of the significance of power electronics in EVs, highlighting its role in energy conversion, motor control, and battery management. Challenges associated with traditional control methods, including difficulties in handling system nonlinearities and uncertainties, are discussed. The paper then delves into the fundamentals of neural network-based control, explaining how neural networks can effectively learn complex mappings between inputs and outputs. Various optimization techniques employed in neural network-based control, such as backpropagation and reinforcement learning, are explored. These techniques enable the neural network controller to adapt and optimize its performance over time.

Keyphrases: control, electric vehicles, neural networks, Optimization, Power Electronics

@Booklet{EasyChair:12717,
  author = {Basit Abbas},
  title = {Optimizing Power Electronics in Electric Vehicle Systems Through Neural Network-Based Control},
  howpublished = {EasyChair Preprint no. 12717},

  year = {EasyChair, 2024}}