Railcar Diagnostics Using Minimal-Redundancy Maximum- Relevance Feature Selection and Support Vector Machine Classification
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
Abstract
Railcar condition is an important factor in the complex web of relationships between railroads, railcar leasing companies, shippers and railcar builders. The most important reasons for this are operational safety and economic considerations pertaining to equipment maintenance. In this study, an approach is presented for the diagnostics of railcar component health from vibration data, utilizing mutual information (MI) based minimal-redundancy-maximalrelevance (mRMR) feature selection and multi-class support vector machine classification. The proposed monitoring solution is a data-driven method which was developed with measurements taken at a railroad test laboratory under controlled conditions. Vibration data was collected from multiple locations on a railcar over several test runs, each utilizing wheelsets with different levels of wear. The input of controlled wheel wear levels was aimed at varying the system outputs to resemble those of cars with different levels of mileage in revenue service. The measured data sets were processed in the time domain, frequency domain and through
wavelet transforms, resulting in the extraction of a set of 687 features from the acceleration signals. A maximum-relevance minimum-redundancy feature selection algorithm was used
to find the optimal combination of features for classification. The algorithm performance was tested for the effect of feature set size, different kernels and scaling techniques on classification accuracy. The results and methods of this assessment are presented in the paper. The paper concludes with a proposal for a monitoring strategy aimed at specifically detecting faulty components and practicing predictive maintenance.
##plugins.themes.bootstrap3.article.details##
diagnostic, Support Vector Machine, Mutual Information, Railcar, Asset management, Minimal-Redundancy Maximum-Relevance, Bogie
Arulampalam, M. S., Maskell, S., Gordon, N., & Clapp, T. (2002). A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking. Signal Processing, IEEE Transactions on, 50(2), 174-188.
Barke, D. a. C. K. W. (2005). Structural health monitoring in the railway industry: A review. Structural Health Monitoring, 4(1), 81 - 94.
Bishop, C. M. (2006). Pattern recognition and machine learning (Vol. 4): springer New York.
Cortes, C., & Vapnik, V. (1995). Support-vector networks. Machine learning, 20(3), 273-297.
Haykin, S.,. (1999). Neural Networks - A Comprehensive Foundation.
Hopkins, B., Seidel, A., Maraini, D., & Shahidi, P. (2015). End-of-car Device Condition Monitoring with Onboard Sensors. Paper presented at the ASME Joint Rail Conference, San Jose, CA.
Hsu, C.-W., Chang, C.-C., & Lin, C.-J. (2003). A practical guide to support vector classification.
Hubbard, P., Ward, C., Goodall, R., & Dixon, R. (2013). Real time detection of low adhesion in the wheel/rail contact. Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit, 0954409713503634.
Kalman, R. E. (1960). A new approach to linear filtering and prediction problems. Journal of Fluids Engineering, 82(1), 35-45.
Kappaganthu, K., & Nataraj, C. (2011). Feature Selection for Fault Detection in Rolling Element Bearings Using Mutual Information. Journal of Vibration and Acoustics, 133(6), 061001-061001.
doi:10.1115/1.4003400
Klingel, W. (1883). Über den Lauf der Eisenbahnwagen auf gerader Bahn. Organ für die Fortschritte des Eisenbahnwesens, 20, 113-123.
Lagnebäck, R. (2007). Evaluation of wayside condition monitoring technologies for condition-based maintenance of railway vehicles: Luleå University of Technology Luleå.
Li, P., & Goodall, R. (2004). Model-based condition monitoring for railway vehicle systems. Paper presented at the Proceedings of the UKACC international conference on control, Bath, UK.
Maraini, D., & Nataraj, C. (2015). Freight Car Roller Bearing Fault Detection Using Artificial Neural Networks and Support Vector Machines. In K. J. Sinha (Ed.), Vibration Engineering and Technology of Machinery: Proceedings of VETOMAC X 2014, held at the University of Manchester, UK, September 9-11, 2014 (pp. 663-672). Cham: Springer International Publishing.
Maraini, D., Shahidi, P., Hopkins, B. M., & Seidel, A. (2014). Development of a Bogie-Mounted Vehicle On-Board Weighing System. Paper presented at the 2014 Joint Rail Conference.
Mei, T., & Li, H. (2008). Measurement of vehicle ground speed using bogie-based inertial sensors. Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid
Transit, 222(2), 107-116.
Peng, H., Fulmi, L., & Ding, C. (2005). Feature selection based on mutual information criteria of max-dependency, max-relevance, and minredundancy. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 27(8), 1226-1238. doi:10.1109/TPAMI.2005.159
Schölkopf, B., Sung, K.-K., Burges, C. J., Girosi, F., Niyogi, P., Poggio, T., & Vapnik, V. (1997). Comparing support vector machines with Gaussian kernels to radial basis function classifiers. Signal Processing, IEEE Transactions on, 45(11), 2758-2765.
Shahidi, P., Maraini, D., Hopkins, B., & Seidel, A. (2014). Estimation of Bogie Performance Criteria Through On-Board Condition Monitoring. Paper presented at the Annual Conference of the Prognostics and Health Management Society 2014, Fort Worth, TX.
Shahidi, P., Maraini, D., Hopkins, B., & Seidel, A. (2015). Railcar Bogie Performance Monitoring using Mutual Information and Support Vector Machines. Paper presented at the Prognostics and Health Management Society 2015, San Diego, CA.
Tournay, H. M., & Lang, R. (2007). History and Teardown Results of Five Loaded Coal Cars Identified as Poor Performers while Passing accorss a Truck Performance Detector R-985. Washington, DC: Association of American Railroads/ Transportation Technologies Center, Inc.
Ward, C. P., Goodall, R. M., Dixon, R., & Charles, G. (2010). Condition monitoring of rail vehicle bogies.
Ward, C. P., Weston, P., Stewart, E., Li, H., Goodall, R. M., Roberts, C., . . . Dixon, R. (2011). Condition monitoring opportunities using vehicle-based sensors. Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit, 225(2), 202-218.
Xia, F., & True, H. (2003, 22-24 April 2003). On the dynamics of the three-piece-freight truck. Paper presented at the Rail Conference, 2003. Proceedings of the 2003 IEEE/ASME Joint.
Zakharov, S. M., & Zharov, I. A. (2005). Criteria of bogie performance and wheel/rail wear prediction based on wayside measurements. Wear, 258(7–8), 1135-1141. doi:http://dx.doi.org/10.1016/j.wear.2004.03.025