کنفرانس بین المللی ماشین و محرکه های الکتریکی

صفحه اصلی / The 5th International Conference on Electrical Machines and Drives

Design and Implementation of an NARMA-L2 Controller for Dynamic Wireless Power Transfer in Electric Vehicles

نویسندگان :

Javad Shariatzadeh¹ Seyed Ehsan Abdollahi² Sayyed Asghar Gholamian³ Jafar Adabi⁴ Seyed Reza Abdollahi⁵

1- Electrical and Computer Engineering Department Babol Noshirvani University of Technology, Babol, Iran 2- Electrical and Computer Engineering Department Babol Noshirvani University of Technology, Babol, Iran 3- Electrical and Computer Engineering Department Babol Noshirvani University of Technology, Babol, Iran 4- Electrical and Computer Engineering Department Babol Noshirvani University of Technology, Babol, Iran 5- Electrical and Computer Engineering Department University of Science and Technology of Mazandaran, Behshahr, Iran

کلمات کلیدی :

Wireless Power Transfer (WPT)،Nonlinear Autoregressive Moving Average (NARMA-L2) controller،Voltage Regulation،load variations

چکیده :

Wireless power transfer (WPT) systems provide a promising solution for efficient charging of electric vehicles (EVs); however, their performance is often challenged by nonlinear dynamics arising from variations in mutual coupling and load fluctuations. This paper introduces a novel control framework based on the Nonlinear Auto Regressive Moving Average model with exogenous inputs (NARMA-L2) to achieve robust output voltage regulation in Series-Series(SS) compensated WPT systems tailored for EV applications. The proposed methodology involves identifying the NARMA-L2 model using a multilayer perceptron neural network trained via the Levenberg–Marquardt optimization algorithm, followed by the design of an inverse feedforward controller derived from the identified model. To validate the proposed control approach, extensive simulations in MATLAB were conducted and benchmarked against the conventional proportional–integral (PI) controller. The results demonstrate that the NARMA-L2 controller achieves superior performance, with negligible overshoot (<3%) and shorter settling times in both reference tracking and load disturbance scenarios. These improvements are particularly advantageous for EV charging, where fast dynamic response and minimal overshoot are essential to ensure stable power flow and prevent potential battery degradation.