Scholarlink

<p>Typically, eddy currents induced in a laminated core are counted as losses, and only their magnitude is considered. On the other hand, as the operating frequency increases, current sensors using laminated cores need to calculate the eddy currents directly to compensate for the phase errors caused by them. A surrogate model based on a deep learning algorithm that uses the output of finite element analysis for training was proposed to compensate for the phase error caused by eddy currents. The proposed method is expected to have higher precision than the existing first-order interpolation function. The proposed method was applied to inverter control and showed superior performance than the existing methods.</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

Compensation of Phase Errors in Current Sensors Induced by Eddy Currents Using a Deep Learning-Based Surrogate Model

  • 期刊论文

  • 发表日期：

    2025年3月26日

查看全文
  • Jong-Hoon Park,
  • Ji-Hoon Han,
  • Seung-Min Song,
  • Sun-Ki Hong

摘要

Typically, eddy currents induced in a laminated core are counted as losses, and only their magnitude is considered. On the other hand, as the operating frequency increases, current sensors using laminated cores need to calculate the eddy currents directly to compensate for the phase errors caused by them. A surrogate model based on a deep learning algorithm that uses the output of finite element analysis for training was proposed to compensate for the phase error caused by eddy currents. The proposed method is expected to have higher precision than the existing first-order interpolation function. The proposed method was applied to inverter control and showed superior performance than the existing methods.