I am trying to predict the temperature of the hot rolled steel section after cooling down from 1300 °C. My predicted temperature reached room temperature at 30 minutes, but journal prediction only came to room temperature at 150th minutes. I considered all input data as per the journal (DOI: 10.1061/(ASCE)ST.1943-541X.0001739). I am sceptical about modelling input parameters like convection and radiation interaction properties. I want to know where I made a mistake. Kindly help me with this. Please provide your valuable suggestions and, if possible, any input data support my study.
Note: I have attached my Abaqus file and my predicted result
for convection you need to have shell elements on surface (generally referred as shell coating), also you need to have complete understanding of direction and orientation of elements (I will suggest to use ANSA as pre-processor).
Regards
Hi Mahalingam,
You are using linear elements and only one element through the thickness of the steel section. This means that you will not be able to correctly predict the sub-surface temperature distribution. At the start of cooling, only a small surface layer starts to cool. Later, the depth to which cooling has occurred grows. This means that the internal temperature distribution will be non-linear and will require many elements through the thickness to correctly reproduce this distribution. I suggest you perform a mesh sensitivity study. Repeat the calculation with models of different mesh density. This would be time consuming on your 3D models, so start by using 2D models of the cross section on the part.
Regards,
Simon
- Badges
- Science topic
- Similar topics
- Social Science
- Media
- Journalism