how we will feel that solution is correct if we dont have experimental results : on the basis of convergence or any any thing else

Convergence is only one of the criteria, but not the only one. You should also consider the validity of the model, the quality of the numerical solution (order of the discretization scheme, grid quality, ...), the precision in the boundary conditions settings...

So there is no sure answer to your question, in my opinion. I think it's always worth to compare the results of the calculations with experimental data. If they're not available, try to look for similar cases in the literature for example.

Hi

ap