[kyo022]

Hello,

I'm an undergraduate student majoring in mechanical engineering,

and now I have been working on writing a paper about some CFD simulations regarding high-speed trains.

Few days ago, I got a result on my first domestic paper, and the examiners questioned about the accuracy of my results.

So, I'm planning to do some grid sensitivity studies to re-submit my paper.

And here is the thing that I'd like to ask if I may. I have total 4 different results, and each case similar geometry(meaning, for example, each case's train has different type of window shape.).
I want to know if I have to conduct the grid sensitivity study for each case OR just it's enough to conduct one case and defend myself with this one GSS.

Sorry for the lengthy story, but I would appreciate if you share your knowledge on this.

Have a nice day.

[piu58]

If you have to run a set of cases which are very similar, it should be sufficient to make a grid sensitivity check only once.

> I have total 4 different results, and each case similar geometr

That doesn't sound good. You have perhaps problems well beyond the simple mesh density.

[lcarasik]

Originally Posted by piu58

If you have to run a set of cases which are very similar, it should be sufficient to make a grid sensitivity check only once.

> I have total 4 different results, and each case similar geometr

That doesn't sound good. You have perhaps problems well beyond the simple mesh density.

I disagree, if this work is intended to be of journal quality, the author really should rerun the same grid sensitivity for each geometries. This is assuming they have sufficient computational resources to do so.