跳到主要导航 跳到搜索 跳到主要内容

Study on life prediction method of mosfet thermal environment experiments based on extended kalman filter

  • Ke Li*
  • , Yuxiang Zhang
  • , Shimin Song
  • , Zhijian Zhao
  • , Lijing Wang
  • *此作品的通讯作者
  • Beihang University
  • China Aerospace Science and Technology Corporation
  • Naval Aeronautical and Astronatical University

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

The prediction method of the residual service life of power MOSFET is studied in this paper. By analyzing the data collected under the existing thermal overload accelerated aging experiment, after processing the data, the failure threshold was set by using the prediction algorithm based on data drive and model to predict the residual life of power MOSFET. The traditional SVR algorithm requires a lot of parameter selection and only a few parameter convergence. The prediction algorithm model is based on Extended Kalman Filter, the extended Kalman filter is relative to the advantage of support vector machine (SVM) regression is used to predict the variance is small, and can be found in a wide range of required to predict the sample interval and the predicted results are more accurate, the test results verify the feasibility of this method.

源语言英语
主期刊名Man-Machine-Environment System Engineering - Proceedings of the 20th International Conference on MMESE, 2020
编辑Shengzhao Long, Balbir S. Dhillon
出版商Springer Science and Business Media Deutschland GmbH
495-503
页数9
ISBN(印刷版)9789811569777
DOI
出版状态已出版 - 2020
活动20th International Conference on Man-Machine-Environment System Engineering, MMESE 2020 - Zhengzhou, 中国
期限: 24 10月 202026 10月 2020

出版系列

姓名Lecture Notes in Electrical Engineering
645
ISSN(印刷版)1876-1100
ISSN(电子版)1876-1119

会议

会议20th International Conference on Man-Machine-Environment System Engineering, MMESE 2020
国家/地区中国
Zhengzhou
时期24/10/2026/10/20

指纹

探究 'Study on life prediction method of mosfet thermal environment experiments based on extended kalman filter' 的科研主题。它们共同构成独一无二的指纹。

引用此