I am running a fluid flow problem in channel, but facing problem with different solutions with different relaxation factors.
1). Initially i set default relaxation factors, then residual curves getting smoothed but solution is not converging but once i decreased relaxation factors to some value, solution is converged and getting balanced net fluxes(mass flux and heat fluxes).
2). secondly, i set low relaxtaion factors, here solutions are getting converged but net fluxes(mass flux and heat fluxes) are not balanced.
I suggest not to change all the factors at the same time ,you may check the residuals and look for the variable that diverges or starts diverging first.I think in your case ,you should leave all factors and only try to focus on momentum factor.
heloo , according to what your saying , your changing lot of factors in the same step, try to change one factor at a time to see what will reach .
Regards
Article An Approaching Fourier Analysis of the Relaxation Effect on ...
It is known that the convergence rate can be improved by the so_called relaxation parameters .The convergence rate of the itrative solution is rapidly reduced as the mesh becomes finer.Hence, the relaxation parameters are problem and mesh dependent and only change the iteration path without changing the final solution.So using the successive over_ relaxation (SOR) which depends on a similar problem under study rather than the fault parameter may help in improving the convergence rate.The using of under_relaxation factors depends on the case by case study and the large value leads to wavy solution and small value causes slow convergence.
- Badges
- Science topic