Abstract:
The rail wheel fatigue failures are a serious problem that has been facing the international railway
industries for more than two centuries. The railway industries have been spending millions of
dollars annually for this problem. Rail wheel fatigue is rolling contact fatigue, which is caused by
cyclic contact stress under rolling motion. Crack is formed due to this cyclic contact stress or
small defect, and propagated to certain values at which fracture of rail wheel occurs. This fatigue
crack propagation in railway wheels occurs under mixed mode (I, II and III) conditions. The aim
of this study is to model and analysis fatigue behaviour of rail wheel rolling contact problem. In
this study, both analytical and FE approaches have been used to predict the expected fatigue life
of rail wheel. Extended finite element (X-FEM) in ABAQUS has been used for FE analysis. The
results obtained from both approaches were compared and fatigue life of the rail wheel was
predicted.
Three multiaxial fatigue models (S-J fatigue model, SWT fatigue model and F-S fatigue model)
which can predict both the fatigue crack initiation life and fatigue cracking plane orientation are
used to predict number of cycles to initiate the crack. All of these models are categorized under
strain life approaches. The number of cycles for the crack initiation that obtained from the S-J,
SWT and F-S fatigue models are 1 x 105
, 5.78 x 105
and 1.32 x 105 respectively.
In order to determine the number of cycles to propagate the crack from the initial flaw to the critical
crack length, the Liu mixed mode fatigue crack model was used. This model has the capability to
consider non-proportionality required for a rail wheel fatigue crack modelling. The number of
cycles for crack propagation which obtained from both analytical and FE results are 2052 and 2020
respectively. This number of cycles are the life of the wheel from the initiated crack grow up to
the critical length of the crack. After the critical length, the crack will propagate rapidly.
