Scholarlink

In order to improve the efficiency of multi-physics field coupling simulation of UHV transformer, this paper constructs a magnetic field based on Proper Orthogonal Decomposition (POD) and Radial Basis Function including Linear Polynomial (RBFLP). reduced order model. The two-dimensional winding region of UHV transformer is simulated by COMSOL, and the magnetic field data under several sets of working conditions are obtained, and the two-dimensional magnetic field reduced order modeling of UHV transformer is obtained by using POD-RBFLP; finally, the effectiveness of reduced order modeling is verified. The results show that the absolute error of the magnetic field is only 3.7 × 10−7T, and the relative error is lower than 0.00056%, which has a very high computational accuracy and the computation time is shortened from 323 s to 0.301 s, and the efficiency is improved by 1,073 times. It pro-vides an efficient solution for the magnetic field and loss of UHV transformers.

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

Reduced Order-Response Surface Modeling of Magnetic Field in UHV Transformer

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Pengfei Tang,
  • Yingqiang Zhang,
  • Xiao Liang,
  • Zhonghao Zhang,
  • Tianhang Long,
  • Yufei Dong,
  • Gang Liu

摘要

In order to improve the efficiency of multi-physics field coupling simulation of UHV transformer, this paper constructs a magnetic field based on Proper Orthogonal Decomposition (POD) and Radial Basis Function including Linear Polynomial (RBFLP). reduced order model. The two-dimensional winding region of UHV transformer is simulated by COMSOL, and the magnetic field data under several sets of working conditions are obtained, and the two-dimensional magnetic field reduced order modeling of UHV transformer is obtained by using POD-RBFLP; finally, the effectiveness of reduced order modeling is verified. The results show that the absolute error of the magnetic field is only 3.7 × 10−7T, and the relative error is lower than 0.00056%, which has a very high computational accuracy and the computation time is shortened from 323 s to 0.301 s, and the efficiency is improved by 1,073 times. It pro-vides an efficient solution for the magnetic field and loss of UHV transformers.