Abstract
In the metal railway bridges' design there is the problem of fatigue, because with the passage of trains, multiple loading and unloading cycle are induced in the structural elements.
The first quantity to considet in fatigue verification is the load: it isn't known in every load case, so it should be measured experimentally to obtain the load spectrum, in other words the variability of the load in a reference period.
The target of this thesis is to analyse the data coming from monitoring system to estimate the load to use in the bridge verifications.
For the monitoring, strain gauges can be used to evaluate the deformation caused by the passage of trains.
It is possible to calculate the speed of train and its behaviour in time creating two sections at a known distance; with the latter, the distances between the axles that is fixed for each type of carriage are calculated and a comparison is made with the Italian rolling stock catalog to identify the type of train.
A non-linear regression is then carried out on experimental data to obtain a better representation of the peak of maximum deformation; the regression is performed with a function chosen by studying the behavior of a rail modeled throug finite element software, which simulates the passage of a load with a series of static analyzes.
Once the maximum deformation has been exstimated, the load value is calculated through a conversion factor which is validated analytically.
|