Computational Technology Resources - CCC

H. Jung¹, O. Nelles², P. Kraemer³, K. Weinberg⁴, G. Kampmann² and C.-P. Fritzen¹

¹Arbeitsgruppe Technische Mechanik, University of Siegen, Germany
²Arbeitsgruppe Mess- und Regelungstechnik - Mechatronik, University of Siegen, Germany
³Lehrstuhl für Mechanik mit Schwerpunkt Schädigungsüberwachung, University of Siegen, Germany
⁴Lehrstuhl für Festkörpermechanik, University of Siegen, Germany

doi:10.4203/ccc.1.27.3

H. Jung, O. Nelles, P. Kraemer, K. Weinberg, G. Kampmann, C.-P. Fritzen, "Automatic System Identification for Robust Fault Detection of Railway Suspensions", in J. Pombo, (Editor), "Proceedings of the Fifth International Conference on Railway Technology: Research, Development and Maintenance", Civil-Comp Press, Edinburgh, UK, Online volume: CCC 1, Paper 27.3, 2022, doi:10.4203/ccc.1.27.3

Keywords: structural health monitoring, automatic fault diagnosis, subspace identification, eigenfrequency density estimation.

Abstract

Vehicle dynamics and safety against derailment are directly influenced by the primary and secondary suspension of a railway vehicle. During the operation faults of components like broken springs or dampers can occur. To prevent a complete system failure, the early detection of faults in the suspension of trains is thus of high importance. A novel approach to sensitive and robust structural health monitoring is proposed. It is based on (i) acceleration measurement, (ii) time-series modeling, (iii) eigenfrequency and possibly mode-shape extraction, (iv) probability density estimation, and finally (v) classification. Compared to traditional approaches the new kernel-based probability density estimation allows to aggregate the results from different data sets. This approach suppresses the spurious eigenfrequencies and emphasizes the physical ones. If, in addition, the mode-shapes are incorporated into the system, the probability density estimator becomes multivariate and the diagnosis accuracy improves further.

download the full-text of this paper (PDF, 7 pages, 1304 Kb)

go to the previous paper
go to the next paper
return to the table of contents
return to the volume description

	Computational & Technology Resources an online resource for computational, engineering & technology publications
	not logged in - login
Front Page Browse CCC CCP CSETS CTR IJRT Other Authors Search Purchase Guide FAQ Contact us	Civil-Comp Conferences ISSN 2753-3239 CCC: 1 PROCEEDINGS OF THE FIFTH INTERNATIONAL CONFERENCE ON RAILWAY TECHNOLOGY: RESEARCH, DEVELOPMENT AND MAINTENANCE Edited by: J. Pombo Paper 27.3 Automatic System Identification for Robust Fault Detection of Railway Suspensions H. Jung¹, O. Nelles², P. Kraemer³, K. Weinberg⁴, G. Kampmann² and C.-P. Fritzen¹ ¹Arbeitsgruppe Technische Mechanik, University of Siegen, Germany ²Arbeitsgruppe Mess- und Regelungstechnik - Mechatronik, University of Siegen, Germany ³Lehrstuhl für Mechanik mit Schwerpunkt Schädigungsüberwachung, University of Siegen, Germany ⁴Lehrstuhl für Festkörpermechanik, University of Siegen, Germany doi:10.4203/ccc.1.27.3 Full Bibliographic Reference for this paper H. Jung, O. Nelles, P. Kraemer, K. Weinberg, G. Kampmann, C.-P. Fritzen, "Automatic System Identification for Robust Fault Detection of Railway Suspensions", in J. Pombo, (Editor), "Proceedings of the Fifth International Conference on Railway Technology: Research, Development and Maintenance", Civil-Comp Press, Edinburgh, UK, Online volume: CCC 1, Paper 27.3, 2022, doi:10.4203/ccc.1.27.3 Keywords: structural health monitoring, automatic fault diagnosis, subspace identification, eigenfrequency density estimation. Abstract Vehicle dynamics and safety against derailment are directly influenced by the primary and secondary suspension of a railway vehicle. During the operation faults of components like broken springs or dampers can occur. To prevent a complete system failure, the early detection of faults in the suspension of trains is thus of high importance. A novel approach to sensitive and robust structural health monitoring is proposed. It is based on (i) acceleration measurement, (ii) time-series modeling, (iii) eigenfrequency and possibly mode-shape extraction, (iv) probability density estimation, and finally (v) classification. Compared to traditional approaches the new kernel-based probability density estimation allows to aggregate the results from different data sets. This approach suppresses the spurious eigenfrequencies and emphasizes the physical ones. If, in addition, the mode-shapes are incorporated into the system, the probability density estimator becomes multivariate and the diagnosis accuracy improves further. download the full-text of this paper (PDF, 7 pages, 1304 Kb) go to the previous paper go to the next paper return to the table of contents return to the volume description
Back to top	©Civil-Comp Limited 2023 - terms & conditions