Over past decades, cold-formed steel (CFS) sections are widely used in wall panel assembly of light gauge steel framed structures. The cold-formed steel sections are primarily used as load bearing structural elements due to their advantages such as higher strength to weight ratio over the other structural materials such as hot-rolled steel, timber and concrete. Fire safety design of building structures is an essential requirement as the fire events cause loss of lives and property. Therefore it is essential to understand the fire performance of cold-formed steel sections under fire condition.

The buckling behavior of cold-formed steel compression member under fire condition is not well modeled using finite element software. As the first phase of this thesis, a detailed review was undertaken on the mechanical and thermal properties of cold-formed steel section at elevated temperature and most reliable properties and stress-strain models was adopted for modeling. As the second phase of this project, the local buckling behavior was investigated based on finite element modeling. The ultimate load carrying capacity for lipped channel section used was identified.

As the third phase of this project, the different modeling used by the other researchers was investigated for flexural-torsional buckling of CFS. The first model was based on the application of fire temperature variation along the length of plasterboard is assumed to be uniform. Later on next model was based on non-uniform temperature distribution along the length of the board which was simulated by dividing the board into sections for different temperature.

This thesis presents the detail comparison of different modeling methods and also presents a proposed equation method of modeling to obtain the more accurate outputs similar to that of the experiment.

The ultimate failure time, failure temperature and the deflected shape were compared with the different modeling methods and the experimental results obtained by (Feng, 2004). Based on the comparison study of finite element analysis methods with experimental results following conclusive remarks were drawn.

• The uniform temperature method of modeling along the length of the wall panel gives the deformation shape of the wall assembly and the cold-formed steel studs. However, failure time could not be achieved when compared to real condition.
• The failure time of uniform temperature method was below the standards and experimental failure time. This difference in temperature could be attributed to the fact that, the same amount of temperature was applied along the length of the stud which resulted in the maximum total temperature affecting the steel stud.
• The board division method of modeling was used to incorporate the nonlinearity of temperature affecting the steel stud along its length. This method resulted in a reduction of failure time and temperature. This may be due the large temperature difference (around 250-300 °C) at certain portions of the section caused by the delay of 2 minutes in temperature growth which resulted in material instability.
• To stimulate such difference in temperature along the length of the wall assembly, the new proposed equation for emissivity was adopted in such a way that the temperature affecting the central portion is higher than the edges.
• The equation method of modeling found to have good agreement with the experimental results in terms of failure temperature, failure time and the deflected shapes.
• The study was also extended to observe the effect in variation of load ratio. The load ratio of 0.4 and 0.2 was applied on the uniform and equation method of modeling. It was clearly seen that increasing load ratio decreases the performance of the wall assembly.

It is noteworthy that the equation method of modeling gives the results similar to that of the experimental and this method helps in reduction of material, time and experimental cost for testing the wall assembly in the laboratory.