Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Civil-Comp Proceedings
ISSN 1759-3433
CCP: 109
Edited by: Y. Tsompanakis, J. Kruis and B.H.V. Topping
Paper 36

Inverse Problems using an Evolutionary Algorithm applied to the Identification of Railway Track Properties based on Measured and Computed Train-Track Behavior

P.A. Ferreira and R. Maciel

Department of Civil Engineering, Instituto Superior Tecnico, University of Lisbon, Portugal

Full Bibliographic Reference for this paper
P.A. Ferreira, R. Maciel, "Inverse Problems using an Evolutionary Algorithm applied to the Identification of Railway Track Properties based on Measured and Computed Train-Track Behavior", in Y. Tsompanakis, J. Kruis, B.H.V. Topping, (Editors), "Proceedings of the Fourth International Conference on Soft Computing Technology in Civil, Structural and Environmental Engineering", Civil-Comp Press, Stirlingshire, UK, Paper 36, 2015. doi:10.4203/ccp.109.36
Keywords: high-speed railway, dynamic train-track model, evolutionary algorithms, inverse problems, multi-objective optimization.

Modelling railway train-track dynamic systems with particular interest on track performance under the passage of the train relies on information which often carries some uncertainties. These uncertainties are usually related to difficulties within track property characterization (laboratory or in situ tests) and simplification hypothesis assumed in modelling some materials mechanical behavior (sub-models).

As explained in this paper, one way to bypass these issues is to formulate an optimization problem with the purpose of identifying the different combinations of real parameter ranges which lead to simulation results that agree with real infield measurements. Thus, an evolutionary algorithm is applied to the identification of railway track substructure properties based on the observed behavior of a railway track subjected to the passage of high-speed trains. The paper reports an inverse problem that is formulated as a constrained multi-objective optimization problem, whose solution consists of the set of parameters that minimize deviations between the output of the computational model and the target ranges derived from field measurements. These field measurements collected during the railway line's commercial exploration are used to provide a quantitative and real description of the railway track's dynamic behavior.

In this way, in presence of certain measured track responses and the support of a validated numerical train/track model, specific track parameters can be more accurately accessed thus enabling a better acquaintance of railway high speed track performance.

purchase the full-text of this paper (price £20)

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