Abstract | There is a large quantity of literature available on longitudinal train dynamics and
risk assessment but nothing that combines these two topics. This thesis is focused at
assessing derailment risks developed due to longitudinal train dynamics. A key
focus of this thesis is to identify strategies that can be field implemented to correctly
manage these risks. This thesis quantifies derailment risk and allows a datum for
comparison. A derailment risk assessment on longitudinal train dynamics was
studied for a 107 vehicle train consist travelling along the Monto and North Coast
Lines in Queensland, Australia. The train consisted of 103 wagons and 4
locomotives with locomotives positioned in groups of two in lead and mid train
positions. The wagons were empty hopper wagons on a track gauge of 1067mm.
The scenarios studied include: the effect of longitudinal impacts on wagon dynamics
in transition curves; and the effects of longitudinal steady forces on wagon dynamics
on curves. Simulation software packages VAMPIRE and CRE-LTS were used.
The effects of longitudinal impacts from in-train forces on wagon dynamics in curves
were studied using longitudinal train simulation and detailed wagon dynamics
simulation. In-train force impacts were produced using a train control action. The
resulting worst-case in-train forces resulting from these simulations were applied to
the coupler pin of the wagon dynamics simulation model. The wagon model was
used to study the effect of these in-train forces when applied in curves and transitions
at an angle to the wagon longitudinal axis. The effects of different levels of coupler
impact forces resulting from different levels of coupling slack were also studied.
Maximum values for wheel unloading and L/V ratio for various curve radii and
coupler slack conditions were identified. The results demonstrated that the
derailment criteria for wheel unloading could be exceeded for a coupler slack of
50mm and 75mm on sharper curves, up to 400m radii.
A detailed study of the effect of steady in-train forces on wagon dynamics on curves
also was completed. Steady in-train forces were applied to a three wagon model
using VAMPIRE. Maximum and minimum values of wheel unloading and L/V ratio
were identified to demonstrate the level of vehicle stability for each scenario. The
results allowed the worse cases of wheel unloading and L/V ratio to be studied in
detail.
Probability density functions were constructed for the occurrence of longitudinal
forces and coupler angles for the Monto and North Coast Lines. Data was simulated
for a coupler slack of 25, 50 and 75mm and force characteristics were further
classified into the occurrences of impact and non-impact forces. These probability
density functions were analysed for each track section to investigate the effects of
coupler slack, track topography and gradient on wagon dynamics. The possible
wagon instability in each of these scenarios was then assessed to give a measure of
the potential consequences of the event. Risk assessment techniques were used to
categorise levels of risk based on the consequences and likelihood of each event. It
was found that for the train configuration simulated, the Monto Line has a higher
derailment risk than the North Coast Line for many of the scenarios studies in this
thesis. For a coupler slack of 25mm no derailment risks were identified, 50mm
coupler slack derailment risks were only identified on the Monto track and the
majority of derailment risks were identified for a 75mm coupler slack. |