功率模块多物理场快速求解技术综述

ZHU Lingyu; TANG Yizheng; ZHAN Cao; KANG Zizhang; WANG Weicheng; JI Shengchang

doi:10.13336/j.1003-6520.hve.20250369

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

功率模块多物理场快速求解技术综述

更新时间：2026-06-19

- 功率模块多物理场快速求解技术综述
- Vol. 51, Issue 10, Pages: 4881-4899(2025)
- 作者机构：
  
  1. 电工材料电气绝缘全国重点实验室(西安交通大学),西安,710049
  2. 弗吉尼亚理工大学电力电子系统研究中心, 弗吉尼亚州布莱克斯堡,24061
- 作者简介：
- 基金信息：
- DOI：10.13336/j.1003-6520.hve.20250369
  CLC：
- Published：2025
- 稿件说明：
移动端阅览
ZHU Lingyu, TANG Yizheng, ZHAN Cao, et al. 功率模块多物理场快速求解技术综述[J]. 2025, 51(10): 4881-4899.
DOI：

ZHU Lingyu, TANG Yizheng, ZHAN Cao, et al. 功率模块多物理场快速求解技术综述[J]. 2025, 51(10): 4881-4899. DOI： 10.13336/j.1003-6520.hve.20250369.

摘要

功率模块作为电能变换系统的核心组件，其可靠性直接决定着新能源发电、高压直流输电等系统的安全稳定性。多物理场建模与分析是揭示功率模块运行状态与劣化机理的核心手段，然而传统数值计算方法存在计算效率瓶颈，难以满足新型电力系统对设备状态实时感知与智能诊断的需求。为此，该文总结了焊接型和压接型功率模块封装形式与多物理场建模方法；从物理模型降阶、投影降阶和深度学习降阶3个方面总结功率模块快速求解技术的研究现状，讨论了多物理场耦合模型的局部降阶方法，对比分析了各种快速求解技术的优缺点。在此基础上，进一步探讨目前迫切需要研究的重点，对进一步深入多物理场快速求解、数字孪生、智慧运维等研究具有重要的参考价值。

Abstract

As a core component of power conversion systems

the reliability of power modules directly determines the safety and stability of systems such as renewable energy generation and high-voltage direct current (HVDC) transmission. Multi-physics modeling and analysis are essential for revealing the operating states and degradation mechanisms of power modules. However

traditional numerical methods face computational efficiency bottlenecks

making it challenging to meet the real-time condition monitoring and intelligent diagnostic requirements of modern power systems. Therefore

this paper summarizes the packaging forms and multi-physics modeling methods of bond-wire and press-pack power modules. Tthe research progress in fast solution techniques for power modules is reviewed

focusing on model order reduction

projection-based reduction

and deep learning-based reduction methods. Moreover

this paper also discusses the local reduction methods for multi-physics coupled models and provides a comparative analysis of the advantages and disadvantages of various fast solution methods. Based on these discussions

this paper further explores the urgent research areas

offering significant reference value for advancing multi-physics fast solution methods

digital twins

and smart operation and maintenance.

关键词

Keywords

references

Views

880

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

基于格拉姆角场图像融合与深度学习算法的大型变压器绕组故障诊断方法

人工智能应用于电力电缆局部放电诊断的研究进展

Electric Equipment Image Recognition Based on Deep Learning and Random Forest

Multi-target Detection and Fault Recognition Image Processing Method

Pattern Recognition of Acoustic Signals of Transformer Core Based on Mel-spectrum and CNN

Related Author

GUO Lei

WANG Hongbo

FU Anzhi

LU Zhuowen

QIAN Guochao

WANG Dongyang

YANG Yong

SHI Huanyu

Related Institution

Electric Power Research Institute, Yunnan Power Grid Co., Ltd.

School of Electrical Engineering, Southwest Jiaotong University

Marketing Service Center, State Grid Anhui Electric Power Co., Ltd.

State Grid Hubei Electric Power Research Institute

State Key Laboratory of Advanced Electromagnetic Engineering and Technology(School of Electrical and Electronic Engineering, Huazhong University of Science and Technology)

AI问答

⁰