AST (devc vs. statistic vs boundary quantity) ((#13357))

[BastianStahra]

Describe the bug
Dear Kevin, please see issue #13357. I don't know how to reopen the old issue so I opened a new one with the same topic. We simulated some testcases in the meantime. I posted them below.

I think I got your point. I understand the different "behaviour" of the AST due to different T_END because of T_END/1000. I also get that you deal with devc, statistic_devc and BC in a different way.

But when I look at the different ways of measuring the AST with point devices, statistic devices and boundary quantity; I see differences in the results of up to 80°C. Is that really in an acceptable range for you? When looking at the same quantity of the same simulation?

In addition to that there is the T_End/1000 point. In sum this could lead to differences of up to more than 100°C in the testcase when looking at the point devices for very long T_END

FH0011 - T_END = 300
FH0013 - T-END = 10.000

FH0011 - T_END = 300

FH0013 - T-END = 10.000

FH0013 - T-END = 10.000
Flame_FH0013.txt

FH0011 - T_END = 300
Flame_FH0011.txt

[drjfloyd]

These are not equivalent outputs.

DT_DEVC defaults to (T_END-T_BEGIN)/1000.
DT_BNDF defaults to (T_END-T_BEGIN)/500

You are not comparing the same output intervals. Set the same DT_DEVC and DT_BNDF for both cases and you should see the same results.