Termisk analys av brandutsatta tunnelkonstruktioner med olika ytegenskaper

Abstract: Tunnels are constructed to withstand a fire during a designed time period.
In Sweden tunnels are dimensioned according to the Hydrocarbon (HC)
temperature curve during 180 minutes. Fires in tunnels are different from
fires in compartments; the temperatures in a tunnel are more severe. The
gas temperatures may be higher than fires in compartments, which may
affect the tunnel construction. A worst case scenario is that it may collapse.
This will have a devastating outcome both for the tunnel owners and the
future traffic situation. There are different ways of protecting tunnel
structures. One is to use insulation called Promatect-T boards. The product
is used worldwide and used for analysis in the present study.
The aim of the study is to gain knowledge about thermal impact on tunnel
structures. A finite element program called TASEF (Temperature Analysis
in Structures Exposed to Fire) has been used throughout this study. In
TASEF, tunnels have been dimensioned and different thermal loads has
been used as an input to see what temperatures are obtained in the
reinforcement bars inside the concrete in the tunnel construction. A
verification of a part in TASEF called TASEF-tube has been made, and an
analysis of the most optimal element size in TASEF-tube was also carried
out.
During the verification of TASEF-tube, a comparison between the
temperatures at the end of the tunnel obtained from TASEF-tube and the
temperatures at the end of the tunnel obtained from an equation made from
DeWitt and Incropera was made. The length of the tunnel, the radius of the
tunnel, the ventilation velocity and the heat transfer coefficient were
changed to see when the temperatures would correlate with each other.
That occurred when a long tunnel and a high value of the heat transfer
coefficient was used. Overall, TASEF and TASEF-tube is a quite good
program for dimensioning simple tubes to see the temperature flow inside
them.
The optimal element size was analysed and calculated to three m, which
gave a relative error of 0.2%. Since the calculation times in TASEF are very
short, and due to a limitation in TASEF regarding the number of nodes and
the number of elements, a longer element size could be used. For example,
an element size of six m gave an absolute error at 5 degrees Celcius.
A comparison between gas temperatures calculated by TASEF-tube and a
model developed by Jonatan Gehandler was made. Focus was only on the
gas temperatures downstream a fire in a tunnel. Two different types of
tunnel surfaces was used; concrete and calcium silicate (Promatect boards).
The temperatures obtained from TASEF-tube and Gehandlers model did
correlate well after approximately 200 seconds when using concrete as a
surface in the tunnel structure. When calcium silicate was used, the
temperatures from Gehandler's model were higher than for TASEF-tube
during the first 200 seconds.
To study the thermal impact on concrete and the reinforcement bars in
concrete, different standardized temperature curves and experimental
temperature curves obtained from full scale tests were compared. The
temperature curves from the full scale test simulate a heavy good vehicle
fire. The temperature curve from the full scale test also showed the same
behaviour as the temperature curve HC during a 60 minutes fire. A
suggestion is then to compare these results obtained in this study and
compare them to Swedish regulations, and see if a change could be made.
Instead of dimensioning tunnels to withstand a HC fire during 180 minutes,
a HC fire during 60 minutes could be enough. This is because if a fire would
last during 180 minutes, a spreading of the fire is needed.
A last case during this study was to analyse if a spreading fire is more
dangerous than a fire in one position. Input data from a model scale test
was used in TASEF, where one, two and three burning wood piles
corresponded to a burning heavy good vehicle. The obtained results showed
that a similar temperature development inside the reinforcement bar inside
the concrete could be seen when the temperatures were measured directly
above the first burning wood pile. Further away from the first burning wood
pile, the temperature evolution was different. The highest temperatures
were obtained approximately 50 m downstream the tunnel, and not directly
above the fire source. The temperature evolution inside the concrete showed
that the differences between the burning wood piles are larger closer to the
surface of the concrete, but near the first reinforcement bars a similar
behaviour could be seen.