Scholarlink

<p>This study developed a novel approach based on separated artificial neural networks (ANNs) to efficiently and accurately model the drain current (<i>I</i><sub><i>D</i></sub>)–gate voltage (<i>V</i><sub><i>G</i></sub>) characteristics of silicon carbide (SiC) power MOSFETs efficiently and accurately. We found that a single ANN cannot model the entire <i>I</i><sub><i>D</i></sub>–<i>V</i><sub><i>G</i></sub> range under a large ON/OFF current ratio (10<sup>− 12</sup> to 10<sup>− 1</sup> mA/mm), which is often observed in wide-bandgap semiconductor technologies, such SiC MOSFETs. To address this problem, we developed a method that involves using two ANNs, one each for the ON- and OFF-states. A transition layer is also used to model the transition between the ON- and OFF-states. We evaluated our method on training datasets of various sizes. This method achieved a coefficient of determination (<i>R</i><sup>2</sup>) exceeding 99.96% on 3000 <i>I</i><sub><i>D</i></sub>–<i>V</i><sub><i>G</i></sub> curves when training was conducted using only 150 randomly selected curves, with a modeling time of less than 10&#xa0;s. Our approach can thus be used to accurately and efficiently model the <i>I</i><sub><i>D</i></sub>–<i>V</i><sub><i>G</i></sub> characteristics of semiconductor devices with large ON/OFF current ratios, such as SiC MOSFETs.</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

Demonstration of accurate ID-VG characteristics modeling in SiC mosfets using separated artificial neural networks with small training dataset

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年5月29日

查看全文
  • Manopat Chankla,
  • Bang-Ren Chen,
  • Shivendra Kumar Singh,
  • Yogesh Singh Chauhan,
  • Wen-Jay Lee,
  • Nan-Yow Chen,
  • Songphol Kanjanachuchai,
  • Tian-Li Wu

摘要

This study developed a novel approach based on separated artificial neural networks (ANNs) to efficiently and accurately model the drain current (ID)–gate voltage (VG) characteristics of silicon carbide (SiC) power MOSFETs efficiently and accurately. We found that a single ANN cannot model the entire ID–VG range under a large ON/OFF current ratio (10− 12 to 10− 1 mA/mm), which is often observed in wide-bandgap semiconductor technologies, such SiC MOSFETs. To address this problem, we developed a method that involves using two ANNs, one each for the ON- and OFF-states. A transition layer is also used to model the transition between the ON- and OFF-states. We evaluated our method on training datasets of various sizes. This method achieved a coefficient of determination (R2) exceeding 99.96% on 3000 ID–VG curves when training was conducted using only 150 randomly selected curves, with a modeling time of less than 10 s. Our approach can thus be used to accurately and efficiently model the ID–VG characteristics of semiconductor devices with large ON/OFF current ratios, such as SiC MOSFETs.