Feature extraction and analysis of switch machine curves based on one-dimensional residual convolutional auto-encoder

Cong Ding; Chao Huang; Haosen Zhao; Zepeng Chen

doi:10.1117/12.2652202

10 November 2022 Feature extraction and analysis of switch machine curves based on one-dimensional residual convolutional auto-encoder

Cong Ding, Chao Huang, Haosen Zhao, Zepeng Chen

Author Affiliations +

Proceedings Volume 12331, International Conference on Mechanisms and Robotics (ICMAR 2022); 123310L (2022) https://doi.org/10.1117/12.2652202
Event: International Conference on Mechanisms and Robotics (ICMAR 2022), 2022, Zhuhai, China

Abstract

Railway switch machine circuit action Curve feature is essential for railway turnout control system fault diagnosis research. In recent research, manual extraction of curve features or supervised learning based on data sets with fault labels to construct diagnostic models has been most adopted, which requires enormous energy and time. Therefore, a method based on one-dimensional residual convolutional Auto-encoder is proposed for curve features extraction in this paper, which can decrease the workload of manual annotation of data sets and lower the influence of the minor duration difference of the switch curves data on code over-fitting through the flexibility of unsupervised learning auto-encoder and convolutional neural network. According to the experiments, the proposed model can effectively extract key switch machine curve features and can featuring strong adaptability.

Citation Download Citation

Cong Ding, Chao Huang, Haosen Zhao, and Zepeng Chen "Feature extraction and analysis of switch machine curves based on one-dimensional residual convolutional auto-encoder", Proc. SPIE 12331, International Conference on Mechanisms and Robotics (ICMAR 2022), 123310L (10 November 2022); https://doi.org/10.1117/12.2652202

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available