Forecasting Vertical Acceleration Railway Wagons - A comparative study

Advances in modern machine learning techniques has encouraged interest in the development of vehicle health monitoring (VHM) systems. These techniques are useful for the reduction of maintenance and inspection requirements of railway systems. The performance of rail vehicles running on a track is limited by the lateral instability and track irregularities of a railway wagon. In this study, a forecasting model has developed to investigate vertical acceleration behavior of railway wagons attached to a moving locomotive using different regression algorithms. Front and rear vertical acceleration conditions have predicted using ten popular learning algorithms. Different types of models can be built using a uniform platform to evaluate their performances. This study was conducted using ten different regression algorithms with five different datasets. Finally best suitable algorithm to predict vertical acceleration of railway wagons have suggested based on performance metrics of the algorithms that includes: correlation coefficient, root mean square (RMS) error and computational complexity.

[1]  Vladimir Vapnik,et al.  An overview of statistical learning theory , 1999, IEEE Trans. Neural Networks.

[2]  V K Garg,et al.  Dynamics of railway vehicle systems , 1984 .

[3]  David J. Olive,et al.  Introduction to Regression Analysis , 2007 .

[4]  Cen Li,et al.  Classifying imbalanced data using a bagging ensemble variation (BEV) , 2007, ACM-SE 45.

[5]  A. B. M. Shawkat Ali,et al.  Data Mining. Methods And Techniques , 2007 .

[6]  Samia Nefti-Meziani,et al.  A neural network approach for railway safety prediction , 2004, 2004 IEEE International Conference on Systems, Man and Cybernetics (IEEE Cat. No.04CH37583).

[7]  Peter Wolfs,et al.  A distributed low cost device for the remote observation of track and vehicle interactions , 2006 .

[8]  Ron Kohavi,et al.  Wrappers for Feature Subset Selection , 1997, Artif. Intell..

[9]  Sakdirat Kaewunruen,et al.  ICSV 14 Cairns • Australia 9-12 July , 2007 1 RESPONSE AND PREDICTION OF DYNAMIC CHARACTERISTICS OF WORN RAIL PADS UNDER STATIC PRELOADS , 2007 .

[10]  David W. Aha,et al.  Tolerating Noisy, Irrelevant and Novel Attributes in Instance-Based Learning Algorithms , 1992, Int. J. Man Mach. Stud..

[11]  Ian H. Witten,et al.  Induction of model trees for predicting continuous classes , 1996 .

[12]  C. Esveld.  Modern railway track , 1989 .

[13]  Kate Smith-Miles,et al.  On learning algorithm selection for classification , 2006, Appl. Soft Comput..

[14]  G. Holmes,et al.  Developing innovative applications in agriculture using data mining , 1999 .

[15]  C. S. Bonaventura,et al.  Intelligent system for real-time prediction of railway vehicle response to the interaction with track geometry , 2000, Proceedings of the 2000 ASME/IEEE Joint Railroad Conference (Cat. No.00CH37110).

[16]  Ian H. Witten,et al.  Data mining: practical machine learning tools and techniques, 3rd Edition , 1999 .

[17]  P.J. Wolfs,et al.  An autonomous, low cost, distributed method for observing vehicle track interactions , 2006, Proceedings of the 2006 IEEE/ASME Joint Rail Conference.

[18]  A. Gyasi-Agyei,et al.  Survey of Wireless Communications Applications in the Railway Industry , 2007, The 2nd International Conference on Wireless Broadband and Ultra Wideband Communications (AusWireless 2007).

[19]  George D. Magoulas,et al.  Neural network-based colonoscopic diagnosis using on-line learning and differential evolution , 2004, Appl. Soft Comput..

[20]  Jason Smith,et al.  Security as a Safety Issue in Rail Communications , 2003, SCS.

[21]  Manicka Dhanasekar,et al.  Monitoring the Dynamics of Freight Wagons , 2002 .

[22]  Sankaran Mahadevan,et al.  Improving railroad wheel inspection planning using classification methods , 2007, Artificial Intelligence and Applications.

[23]  Masahiro Miwa,et al.  Measurement and analysis of track irregularity on super-high speed train - TRIPS , 2000 .