A Novel Fault Diagnosis Method for High-Speed Railway Turnout Based On DCAE-Logistic Regression

Turnout fault diagnosis is an essential means to reduce maintenance cost and ensure safety in high-speed railway operation. Aiming to improve diagnosis accuracy, this paper proposes a novel hybrid deep learning framework combining Deep Convolutional Auto-encoder (DCAE) and Logistic Regression (LR) for turnout fault diagnosis. The raw turnout current signal data is converted to 2-D image, and DCAE is employed to automatically extract features of 2-D images. Then the feature data is fed into LR for turnout fault diagnosis. Thanks to extracting features automatically, the proposed method can overcome the weakness that manual feature extraction depends on much expertise and prior knowledge in traditional data-driven diagnosis method. The proposed method can achieve an accuracy of 99.52% on historical field data collected from a real high-speed railway turnout.

[1]  Wei Dong,et al.  Fault Diagnosis of High-Speed Railway Turnout Based on Convolutional Neural Network , 2018, 2018 24th International Conference on Automation and Computing (ICAC).

[2]  José Sá da Costa,et al.  Application of a novel fuzzy classifier to fault detection and isolation of the DAMADICS benchmark problem , 2006 .

[3]  Yang Liu,et al.  Review on Fault Diagnosis Techniques for Closed-loop Systems , 2013 .

[4]  Steven X. Ding,et al.  Model-based Fault Diagnosis Techniques: Design Schemes, Algorithms, and Tools , 2008 .

[5]  Li Xia,et al.  Fault diagnosis of high-speed railway turnout based on support vector machine , 2016, 2016 IEEE International Conference on Industrial Technology (ICIT).

[6]  Fan Zhang,et al.  Turnout Fault Diagnosis through Dynamic Time Warping and Signal Normalization , 2017 .

[7]  Hao Ye,et al.  A Fault Detection Method for Railway Point Machine Operations Based On Stacked Autoencoders , 2018, 2018 24th International Conference on Automation and Computing (ICAC).

[8]  Liang Gao,et al.  A New Convolutional Neural Network-Based Data-Driven Fault Diagnosis Method , 2018, IEEE Transactions on Industrial Electronics.

[9]  Wang Xiao-fen.  Review on modern fault diagnosis technologies , 2013 .

[10]  Muhammad Numan,et al.  Logistic regression and feature extraction based fault diagnosis of main bearing of wind turbines , 2016, 2016 IEEE 11th Conference on Industrial Electronics and Applications (ICIEA).

[11]  Mehmet Sevkli,et al.  A Simple State-Based Prognostic Model for Railway Turnout Systems , 2011, IEEE Transactions on Industrial Electronics.

[12]  Gehao Sheng,et al.  An Integrated Data Preprocessing Framework Based on Apache Spark for Fault Diagnosis of Power Grid Equipment , 2017, J. Signal Process. Syst..