I am modeling a composite slab subjected to fire (ISO 834-1). I have two models, a heat transfer model in ABAQUS / standard and a mechanical model in ABAQUS / Explicit due to convergences issues in concrete damage plasticity.
In my heat transfer model, I apply heat for 6000 seconds according to my experimental data.
In my mechanical model I firstly apply a pressure load in the first step, and in the second step I import the nodal temperatures as a function of time as a predefined field. The problem is the computational costs in this second step that greatly increase my analysis time.
I would like to know how can I accelerate the development of temperatures in my mechanical model?
Thanks in advance.
Hello
subscribe to and seek assistance from the
ABAQUS mailing list
https://groups.yahoo.com/neo/groups/Abaqus/info
Search the archive of the list before posting in it.
Good luck
Frank
With ABAQUS/Standard the calculations should be faster. ABAQUS/Standard is more suitable for longer times. The large deformations are no problem. A suitable choice of the time steps is also helpful.
In ABAQUS / Explicit, you can use the Predefined Field in the Load Module to define your thermal field. The value you input will be assigned to every nodes automatically so that you don't have to use the Standard & Explicit at the same time. If you still prefer faster calculation, bigger mass scaling coefficient could be helpful.
Dear Henry,
My model presents a lot of convergence issues in the concrete due to the plasticity algorithm in abaqus standard. Thus the time increments become very small. That's the reason to use Abaqus / Explicit.
Thank you.
Dear Bozhou,
The only way to increase the stable time incremet is through the mass scaling technique?
Thank you.
It is not the ONLY way, of course, but it's the most efficient way. However, bigger mass scaling coefficient means that you have to pay a price for your calculation accuracy. You can also decrease the time of your load step to accelerate your analysis. 6000s may be quite long.
Yes I understand,
I'm using the mass scaling, but to achieve a stable time incremet reasonable I have to increase the mass too much and thus I lose quality in the results.
In fact, the 6,000 seconds are due to the previous heat transfer model, in which I submit my model to the fire situation and record the nodal temperatures at pre-established times.
Therefore, to apply the development of nodal temperatures as a function of time as a predefined field in my mechanical model (in ABAQUS/Explicit), I have to define an equivalent time in this step.
I would like to know if there is another way to accelerate this nodal temperature development in my mechanical model.
Thank you.
Elisson Ferreira Personally, it's not necessary to use Standard to define temperature field. You can employ the Explicit for all of your analysis. Loading and hear transferring could be divided into two sub-steps. The thermal field could be realized by defining the predefined field. In this way, the time and mass scaling could be defined separately, which is convenient.
- Badges
- Science topic
- Similar topics
- Engineering
- Thermal Engineering
- Thermofluid
- Heat Transfer