@inproceedings{50ebc906aa454f53947eceba277590e9,
title = "Application of unsupervised linear discriminant analysis for condition monitoring of rotating machinery",
abstract = "As a popular feature extraction method, Linear Discriminant Analysis (LDA) is used to deal with the discrimination recognition directly. However, a supervised method like LDA is difficult to apply in online condition monitoring of rotating machinery due to the lack of data label. In this paper an unsupervised LDA method integrated with Principle Component Analysis (PCA) and K-Means clustering is proposed to simplify the application of LDA for the rotating machinery dataset. The proposed method was analyzed and verified for the actual rotating machinery dataset collected from rotor shaft under different fault condition. K-nearest neighbor was adopted to show the effectiveness of the proposed unsupervised LDA method in extracting features from rotor dataset. The result shows that the new method not only can identify different conditions of the rotors, but also can be easily applied to online condition monitoring for rotating machinery.",
keywords = "Feature extraction, K-means clustering, Linear Discriminant Analysis, Principal Component Analysis, Unsupervised method",
author = "Noman Khandaker and Ruxin Zhang and Shuzhi Dong and Bin Xu and Zhifeng Zhang and Guangrui Wen",
note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 9th International Conference on Modelling, Identification and Control, ICMIC 2017 ; Conference date: 10-07-2017 Through 12-07-2017",
year = "2017",
month = jul,
day = "2",
doi = "10.1109/ICMIC.2017.8321687",
language = "英语",
series = "Proceedings of 2017 9th International Conference On Modelling, Identification and Control, ICMIC 2017",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "43--48",
booktitle = "Proceedings of 2017 9th International Conference On Modelling, Identification and Control, ICMIC 2017",
}