how to judge convergence

  • 19 Views
  • Last Post 17 February 2019
Weiqiang Liu posted this 16 February 2019

hello, all,

I am doing a methane catalytic combustion in a micro-channel. When I run the model, all the residuals can be low than 1e-5 and keep almost unchanged all the time like the picture shows below:

However, the temperature contour keeps changing with the iteration going on. The reaction zone would move very slowly to the outlet of the channel and finally the whole channel temperature would be 300K. What happens is I guess failure of ignition. The picture below shows the temperature contours during calculation:

In this picture, the reaction zone is already too close to the outlet of the channel. 

Let me just put the ignition problem aside. I am just wondering if all the residuals keep unchanged, then all the results should keep unchanged, right?

or am I misunderstanding something?

weiqiang

abenhadj posted this 17 February 2019

Check the Judging Convergence Chapter in Fluent User's Guide. Recommend checking:

 

1/Imbalances

2/Evolution of key variables

3/Using local residuals scaling

 

to judge convergence

Best regards,

Amine

Close