@inproceedings{d255c6c9f62d4b77a0b8d48e8e7997e1,
title = "A Method for Intermittent Fault Diagnosis of Electronic Equipment Based on Labeled SOM",
abstract = "Intermittent fault of electronic equipment is an essential problem in the industry, and its diagnosis is paid more attention. The data-driven approach is commonly utilized, but it focuses on the mapping relation more than the inner data structure characteristics. In this paper, a method for intermittent fault diagnosis based on SOM. SOM topology is obtained based on unsupervised learning, and it is labeled using the majority voting strategy in the associated samples of every neuro. Inputs are judged as states excluding intermittent fault are processed again using SVM, and two new features are proposed for the SVM training. The 'one-versus-one' is utilized to establish multi-classification SVM. Voltage conversion board is taken as an example, in which the presented method is applied and compared with ANN and SVM. The result shows that the proposed method is effective.",
keywords = "BIT, SOM, SVM, fault diagnosis, intermittent fault",
author = "Weiwei Cui and Weikang Xue and Lianfeng Li and Junyou Shi",
note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 4th International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2020 ; Conference date: 05-08-2020 Through 07-08-2020",
year = "2020",
month = aug,
day = "5",
doi = "10.1109/SDPC49476.2020.9353111",
language = "英语",
series = "Proceedings of 2020 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2020",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "149--154",
editor = "Yong Qin and Zuo, \{Ming J.\} and Xiaojian Yi and Limin Jia and Dejan Gjorgjevikj",
booktitle = "Proceedings of 2020 International Conference on Sensing, Diagnostics, Prognostics, and Control, SDPC 2020",
address = "美国",
}