Predicting Potential Railway Operational Disruptions with Echo State Networks

Abstract

European passenger rail systems are massively interconnected and operate with very high frequency. The effects of single-component failures on these types of systems can significantly affect technical and operational reliability. Today advanced diagnostic tools with broad functionalities are being added to systems and system components. These tools control the operation of, support the maintenance of, and monitor the highly sophisticated and interconnected components. A set of diagnostic event data from a passenger train exterior door system was used to predict the occurrence of events that might evolve into operational disruptions that affect train operation and therefore railway reliability. This approach used a neural network algorithm with dynamic temporal behavior (the echo state network) in combination with principal component analysis. The proposed approach exhibited a prediction accuracy of up to 99%.

Get full access to this article

View all access options for this article.

References

Vachtsevanos

Intelligent Fault Diagnosis and Prognosis for Engineering Systems. John Wiley & Sons, Hoboken, N.J., 2006.

Fararooy

, and Allan

On-line Condition Monitoring of Railway Equipment Using Neural Networks. In IEE Colloquium on Advanced Condition Monitoring Systems for Railways, Institute of Electrical Engineers, London, 1995, pp. 21–29.

Marquez

F. P. G.

, Schmid

, and Conde Collado

A Reliability Centered Approach to Remote Condition Monitoring: A Railway Points Case Study. Reliability Engineering & System Safety, Vol. 80, No. 1, 2003, pp. 33–40.

Lehrasab

, Dassanayake

H. P. B.

, Roberts

, Fararooy

, and Goodman

C. J.

Industrial Fault Diagnosis: Pneumatic Train Door Case Study. Journal of Rail and Rapid Transit, Vol. 216, No. 3, 2002, pp. 175–183.

Roberts

, and Silmon

Advanced Techniques for Monitoring the Condition of Mission-Critical Railway Equipment. In Railway Safety, Reliability, and Security: Technologies and Systems Engineering ( Flamini

, ed.), Information Science Reference, Hershey, Pa., 2012.

Smith

A. E.

, Coit

D. W.

, and Yun-Chia

Neural Network Models to Anticipate Failures of Airport Ground Transportation Vehicle Doors. IEEE Transactions on Automation Science and Engineering, Vol. 7, No. 1, 2010, pp. 183–188.

Fink

, and Weidmann

Scope and Potential of Applying Artificial Neural Networks in Reliability Prediction with a Focus on Railway Rolling Stock. In European Safety and Reliability Conference (ESREL), Troyes, France, Taylor & Francis Group, London, 2011.

Jaeger

, and Haas

Harnessing Nonlinearity: Predicting Chaotic Systems and Saving Energy in Wireless Communication. Science, Vol. 304, No. 5667, 2004, pp. 78–80.

Haykin

S. S.

Neural Networks and Learning Machines, 3rd ed. Pearson Education, Upper Saddle River, N.J., 2009.

10.

Mandic

D. P.

, and Chambers

J. A.

Recurrent Neural Networks for Prediction Learning Algorithms, Architectures, and Stability. John Wiley & Sons, Chichester, United Kingdom, 2001.

11.

Verstraeten

Reservoir Computing: Computation with Dynamical Systems. PhD dissertation. Faculty of Engineering, Ghent University, Ghent, Belgium, 2009.

12.

Jaeger

Tutorial on Training Recurrent Neural Networks, Covering BPTT, RTRL, EKF and the Echo State Network Approach. GMD Report 159. German National Research Center for Information Technology, St. Augustin, Germany, 2005.

13.

Lukoševičius

A Practical Guide to Applying Echo State Networks. In Neural Networks: Tricks of the Trade ( Montavon

, Orr

, and Müller

K.-R.

, eds.), Springer, Berlin, 2012.

14.

Duda

R. O.

, Hart

P. E.

, and Stork

D. G.

Pattern Classification, 2nd ed. John Wiley & Sons, New York, 2001.

15.

Jolliffe

J. T.

Principal Component Analysis, 2nd ed. Springer, New York, 2004.

16.

Bengio

Gradient-Based Optimization of Hyperparameters. Neural Computation, Vol. 12, No. 8, 2000, pp. 1889–1900.

17.

Bishop

C. M.

Neural Networks for Pattern Recognition, reprinted 2005 ed. Oxford University Press, Oxford, United Kingdom, 2005.

18.

Rossum

G. V.

Python Setup and Usage, Release 2.7.3. Python Software Foundation, Beaverton, Ore., 2012.

19.

Peterson

F2PY: A Tool for Connecting Fortran and Python Programs. International Journal of Computational Science and Engineering, Vol. 4, No. 4, 2009, pp. 296–305.

20.

Hunter

J. D.

Matplotlib: A 2D Graphics Environment. Computing in Science and Engineering, Vol. 9, No. 3, 2007, pp. 90–95.

21.

Zito

, Wilbert

, Wiskott

, and Berkes

Modular Toolkit for Data Processing (MDP): A Python Data Processing Framework. Frontiers in Neuroinformatics, Vol. 2, 2008.

22.

Verstraeten

Oger toolbox. Reservoir-Computing, 2010. http://organic.elis.ugent.be/ORGANIC/ENGINE. Accessed Aug. 24, 2011.