Scholarlink

<p>Electric vehicles (EVs) are growing an integral part of today's means of transportation as we shift to more environmentally friendly and greener ways to get around the city. Wireless Power Transfer (WPT) has become known as a vital technology to facilitate this transition facilitating simple, conventional cable unlimited charging and lowering overall emissions. In this article, a new smart control architecture for the primary side of a WPT charger is proposed by combining Hybrid Adaptive Swarm–Herd Optimisation (HAS–HO) and a feed-forward Artificial Neural Network (ANN). The HAS–HO technique is used to improve the controller parameters to prevent the switching losses whereas the ANN offers real-time performance prediction and adaptive control during steady-state current charging. By combining these methods, stable operation, high fault tolerance against communication loss, and enhanced energy transfer efficacy are guaranteed. Simulation results show an efficiency of 93.45%, and the experimental validation on a prototype of 1&#xa0;kW at 85&#xa0;kHz shows a good agreement with computational predictions. Therefore, the proposed HAS–HO + ANN structure is a robust and efficient approach for future generation EV wireless charging systems, which directly supports sustainable transportation goals.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Design and experimental validation of wireless electric vehicle charger control using hybrid adaptive swarm–herd optimisation (HAS–HO) and feed-forward artificial neural network

  • 期刊论文

  • 发表日期：

    2026年6月29日

查看全文
  • P. Geetha,
  • S. Usha

摘要

Electric vehicles (EVs) are growing an integral part of today's means of transportation as we shift to more environmentally friendly and greener ways to get around the city. Wireless Power Transfer (WPT) has become known as a vital technology to facilitate this transition facilitating simple, conventional cable unlimited charging and lowering overall emissions. In this article, a new smart control architecture for the primary side of a WPT charger is proposed by combining Hybrid Adaptive Swarm–Herd Optimisation (HAS–HO) and a feed-forward Artificial Neural Network (ANN). The HAS–HO technique is used to improve the controller parameters to prevent the switching losses whereas the ANN offers real-time performance prediction and adaptive control during steady-state current charging. By combining these methods, stable operation, high fault tolerance against communication loss, and enhanced energy transfer efficacy are guaranteed. Simulation results show an efficiency of 93.45%, and the experimental validation on a prototype of 1 kW at 85 kHz shows a good agreement with computational predictions. Therefore, the proposed HAS–HO + ANN structure is a robust and efficient approach for future generation EV wireless charging systems, which directly supports sustainable transportation goals.