Information Technology Reference
In-Depth Information
Fig. 3. Distribution of rail removals for a fixed grinding cycle depending on the
grinding depth
The curve plotted on figures 4 shows that the two risks are at the same
level. Thee internal rail defect can only been “neglected” in the first life time
(around 10 years of use)!
The previous results confirm the data collected on the 3870km French
high speed lines:
λ track 5 · 10 6 /hour/trackkm
(4)
with an average of 16 year old rails.
Petri nets modelisation
The problem of comparing the economic and safety parameters of the
two systems is not so simple. The Petri nets can be used to formalize the
stochastical and logical links between the various situations.
The parameter x is the % rate of the undetected internal defect by the
measuring cars, z is the % rate of non eliminated contact defect by the cyclic
grinding operations and y is the % rate of detected broken rails by the track
circuit. For HSL: x
4%
, z
60%
(one grinding per year) and y
95%
(with out the switches).
For HSL: x
4%
(the detection by the driver comes to late), z
100%
(three grinding operations per year).
In the both cases, the simulation give us the distribution of the time to
reach, either position 6, traducing the fact that the broken rail is crossed
Search WWH ::




Custom Search