Verification of 2D Heat Transfer Models Developed in LS-DYNA For Structural Fire Engineering Applications
Verification of 2D Heat Transfer Models Developed in LS-DYNA For Structural Fire Engineering Applications
Verification of 2D Heat Transfer Models Developed in LS-DYNA For Structural Fire Engineering Applications
net/publication/319494872
CITATIONS READS
0 206
5 authors, including:
Panagiotis Kotsovinos
Arup
24 PUBLICATIONS 86 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Panagiotis Kotsovinos on 02 March 2018.
ABSTRACT: Understanding the spatial heat distribution within a structural member is crucial for
appropriately assessing the structural performance of the member in fire conditions. Obtaining an
appropriate heat transfer model that correctly captures the effects of geometry boundary condi-
tions and material non-linearities requires the use of verified finite element numerical scheme.
Arup’s preferred tool for the purpose is the commercially available multipurpose finite element
package LS-DYNA. This paper presents the methodology and selected results of a verification
exercise for the use of LS-DYNA for heat transfer for structural fire engineering applications that
was conducted by Arup.
1 INTRODUCTION
2 METHODOLOGY
The objective of this study is to verify that LS-Dyna can predict the thermal response
within different construction materials and assemblies. The focus falls on members
composed of steel, concrete and fire protection materials incorporating protective mate-
rials and radiating enclosures and gaps. LS-Dyna simulation predictions are compared
to the following models and solutions with progressively increasing degree of complex-
ity:
Analytical solutions for simple transient 0D and 1D thermal response cases; and
Benchmark cases presented in Wickström & Pålsson (1999).
3 EXAMPLE METHODS
Results from two verification models will be presented. These studies assess the heat
transfer in composite steel and concrete construction and the heat transfer in a protected
steel section with two radiating voids. These cases are both taken from Wickström &
Pålsson (1999).
LS Dyna can be set to use different numerical schemes for heat transfer analysis, for
both of the studies discussed below implicit analysis (with respect to time) was used.
This is set in the LS Dyna’s THERMAL_TIMESTEP control card by setting TIP equal
to 1.
Convection and radiation boundary conditions following the ISO 834 cellulosic fire
curve are applied to the external faces of the model. Internal radiation boundary condi-
tion is applied within the voids. Convection within the cavities was not included.
The temperature is measured at the base of the web and in the centre of the whole dura-
tion of the simulation and along line “S” at 60 minutes.
The predictions from the LS Dyna model were compared against the findings of an SP
Report 1999:36 for up to 120min of heating. It was determined that the results closely
followed the benchmark throughout the simulation. The LS-Dyna model predictions are
consistently below those of the SP Report predictions with a peak difference of ~10°C
at the base of the web, and ~30°C along the line “S” (see Figure 3). This is attributed to
the fact that the LS-DYNA model lacks a means for convective heat transfer through the
void, while SP have included some convective transfer within their model. When com-
paring this difference along “S” it is noted that it reduces along the flange as a propor-
tion of the heat transferred via the void reduces and conduction becomes more domi-
nant.
Figure 3. Flange temperatures comparison Figure 4. Temperature comparison along
"S" line
This scenario demonstrates that LS-Dyna is able to represent heat transfer via radiation
enclosures and perfect transfer of heat between two solids in contact. Provided a dense
enough mesh is used to give a number of elements through the thickness of the materi-
als, the results are not sensitive to element size.
Radiative and convective boundary conditions were applied to the exterior faces. Full
contact between the concrete and the steel was established by shared nodes.
The temperature was reported in the centre of the concrete slab and on the corner and on
the midside of the boundary between steel and concrete (Figure 6).
The results illustrated in Figure 7 show that LS-Dyna predictions correspond to the tem-
perature values reported Wickström & Pålsson (1999) at the selected locations and
times with negligible divergence. This scenario demonstrates that LS-Dyna is able rep-
resent heat transfer though multiple materials and perfect transfer of heat between two
solids in contact.
Figure 7 Temperature distribution in composite construction
4 CONCLUCIONS
This paper outlined the verification process of LS-DYNA, a multipurpose FEA software
package currently used by Arup Fire for structural fire engineering applications, through
a selected number of benchmarks.
Arup has tested and verified the package and we are confident that it provides accurate
predictions in the following areas when used correctly:
Use of 3D solid elements to represent 2D heat transfer problems
Thermal material properties that vary with temperature
Conduction through a solid material
Conduction between two solids in contact
Non-linear convective and radiative boundary conditions
Radiative heat transfer across an enclosure within a model boundary
As a result, the software can be used for advanced heat transfer modelling falling within
the scope of capabilities discussed in this paper.
Arup’s further verification studies include 3D heat transfer problems with incorporation
of convective heat transfer within an enclosure.
5 REFERENCES
British Standard Institution. 1991. BS EN 1991-1-2. Actions on Structures – Part 1-2: General
actions – Actions on structures exposed to fire. London, British Standards Institution.
International Organization for Standardization. 2014. ISO 834-11:2014. Fire resistance tests –
Elements of building construction – Part 11: Specific requirements for the assessment of fire
protection to structural steel elements. International Organization for Standardization.
Wickström, U. & Pålsson, J. 1999. Scheme for Verification of Computer Codes for Calculating
Temperature in Fire Exposed Structures. SP Report 1999:36. SP Swedish Testing and Re-
search Institute.