Keywords: load models, high-speed train, dynamics of railway bridges.

This paper investigates the influence of track-bridge interaction (TBI) in the comparison of dynamic effects of actual (or envisaged) high-speed trains vs. load models prescribed by national or international standards. The focus is set on concentrated load models of high-speed vehicles travelling over simple bridges, following a general methodology previously presented by the authors. For such comparison, two variables named exceedance (in amplitude) and required speed increase were used to analyse whether a real train is or is not duly covered by a model prescribed in a standard. Two particular cases are analysed here of one regular and one conventional train, both compliant with Annex E in EN1991-2, which are compared vs model HSLM-A in two slab-type bridges of spans 10.8 m and 25.2 m. Two different mathematical models are considered for each bridge, characterised by very different levels of refinement. The frequencies of the bridges are selected intentionally low, close to the limit of the frequency band defined in EN1991-2, section 6.4.4; therefore, it is of interest to note how the resonant speeds are also low, particularly for the second example. The main conclusions of this work are two. First, the difference between the response predicted at mid-span by a simple beam model (1-DOF) and a much more elaborated 2D FE model with TBI is very small (in the second example, totally negligible). The reason behind it is that resonances of the fundamental mode prevail in the envelope response at mid-span of these two examples, as in the majority of simply-supported bridges. The remarkable similitude is obtained particularly because damping is derived from free vibration in the FE model, and subsequently assigned to the 1-DOF model. Secondly, since the responses obtained with both models are so similar, their influence in the comparison of real trains vs. HSLM-A in the cases analysed here can be considered negligible in engineering terms, i.e. the areas of non-coverage predicted by the two models are essentially the same. Therefore, TBI has no influence in the two bridges studied in this paper. Further research is required to establish whether this relevant finding could be retained as a general conclusion.

download the full-text of this paper (PDF, 7 pages, 597 Kb)

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