Validation for numerical modeling

[Batzo]

Hi, Before my study , I verified experiment of concrete slab subjected to blast with an article.
I am using CDP and Concrete behavior is dependent on strain rate.

In my study case I have different blast load and different concrete strength.
How do I know that my numerical model is correct? I have nothing to compare to and I have no article to calibrate to...

 

[FEA way]

I would run a simulation with data corresponding to experiment. If you get good results (response similar to that of the test sample) then there’s a great chance that your actual model is also correct. All this assuming that the experiment was similar to your main numerical model and there were only a few differences like load magnitude and material’s strength.

 

[Batzo]

But there is a difference with blast range (if the explosion is in 5m or 20m)
So how do I know how to calibrate my model?

 

[FEA way]

It seems that there’s no other way to validate these results, apart from making sure that they look reasonable. In many cases we don’t have experimental data for comparison and we can only do some standard checks to avoid common errors.

 

[IceBreakerSours]

In this case, you may be better off leveraging statistics on top of FEA. There are various ways to do it. Here's one: Use the high fidelity model (FEA, in this case) to come up with a meta-model (transfer function) and run statistical simulations off of that meta-model assuming distributions of various inputs. This will give you a probabilistic output.

*********************************************************
Are you new to this forum? If so, please read these FAQs:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083