Can deep learning be used in pattern recognition for time series? For example, is there any research on mechanical fault pattern recognition based on noise signal?

Posted on

Can deep learning be used in pattern recognition for time series? For example, is there any research on mechanical fault pattern recognition based on noise signal? -The answer of Yuncheng Wanlihttps://www.zhihu.com/question/40992219/answer/1325531901

The deep residual shrinkage network is a kind of mechanical fault pattern recognition method specially for noise and vibration signals.

As shown in the figure below, the deep residual shrinkage network adopts soft thresholding in its structure, so it is suitable for processing vibration signals with noise.

Can deep learning be used in pattern recognition for time series? For example, is there any research on mechanical fault pattern recognition based on noise signal?

This paper is published in IEEE Transactions on industrial information

M. Zhao, S. Zhong, X. Fu, B. Tang, M. Pecht, Deep residual shrinkage networks for fault diagnosis, IEEE Transactions on Industrial Informatics, 2020, 16(7): 4681-4690.

https://github.com/zhao62/Deep-Residual-Shrinkage-Networks

Leave a Reply

Your email address will not be published.