Scholarlink

The main objective of this paper is to investigate the feasibility of employing Physics-Informed Neural Networks (PINNs), particularly PINNs based on Kolmogorov-Arnold Networks (KANs), for facilitating Electromagnetic Interference (EMI) simulations. This work first introduces a common electromagnetic problem and the corresponding formulation and then shows how it can be solved using AI-driven solutions instead of lengthy and complex full-wave numerical simulations. This research may open new horizons for green EMI simulation workflows with less energy consumption and feasible computational capacity.

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

Investigating KAN-Based Physics-Informed Neural Networks for EMI/EMC Simulations

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Kun Qian,
  • Mohamed Kheir

摘要

The main objective of this paper is to investigate the feasibility of employing Physics-Informed Neural Networks (PINNs), particularly PINNs based on Kolmogorov-Arnold Networks (KANs), for facilitating Electromagnetic Interference (EMI) simulations. This work first introduces a common electromagnetic problem and the corresponding formulation and then shows how it can be solved using AI-driven solutions instead of lengthy and complex full-wave numerical simulations. This research may open new horizons for green EMI simulation workflows with less energy consumption and feasible computational capacity.