Discriminant autoencoder for feature extraction in fault diagnosis
Nowadays, some traditional autoencoders and their extensions have been widely applied in data-driven fault diagnosis for feature extraction. However, because of the fact that traditional autoencoders could not make use of label information, the representations extracted by these traditional autoenco...
Saved in:
Published in | Chemometrics and intelligent laboratory systems Vol. 192; p. 103814 |
---|---|
Main Authors | , , , |
Format | Journal Article |
Language | English |
Published |
Elsevier B.V
15.09.2019
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | Nowadays, some traditional autoencoders and their extensions have been widely applied in data-driven fault diagnosis for feature extraction. However, because of the fact that traditional autoencoders could not make use of label information, the representations extracted by these traditional autoencoders may show disappointing results when handling ultimate discriminative task. In this paper, we propose a novel semi-supervised autoencoder, which is named as Discriminant Autoencoder. The training of proposed Discriminant Autoencoder includes a supervised process and an unsupervised process. And a distance penalty is added into the loss function, which enables the proposed Discriminant Autoencoder to extract more suitable representations from industrial data samples. In order to explain the effectiveness of this semi-supervised autoencoder, we carry out some experiments and give out a mathematical derivation. Here we use an industrial batch process dataset as the criterion dataset to test the performance of proposed Discriminant Autoencoder and other conventional autoencoders.
•A novel semi-supervised autoencoder (Discriminant Autoencoder) is proposed to extract features for fault diagnosis.•The new proposed loss function could increase the interclass separability and retain the most information of input.•A mathematical derivation based on MI is provided to explain the effectiveness of the proposed Discriminant Autoencoder.•Batch process dataset is used as the test dataset. |
---|---|
ISSN: | 0169-7439 1873-3239 |
DOI: | 10.1016/j.chemolab.2019.103814 |