Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
21 apr 2011, 11:28 GMT-4
In addition to my previous post. The model works fine, however when I solve using a parametric sweep for a given parameter. Only the solution for the last parameter value is available for post-processing. Although I can select the other solutions as options to plot, they all appear to be empty, it seems like comsol is not saving these early solutions. I am not sure if this is a bug in comsol, or if I am doing something wrong.
More confusing is that I created a simple reactive-transport model involving only 1 species and I can view and plot the model solutions for all the values solved for by the parametric sweep. However when I switch to the more complicated model I only get the last one. All the solver settings appear to be the same in both models.
In addition to my previous post. The model works fine, however when I solve using a parametric sweep for a given parameter. Only the solution for the last parameter value is available for post-processing. Although I can select the other solutions as options to plot, they all appear to be empty, it seems like comsol is not saving these early solutions. I am not sure if this is a bug in comsol, or if I am doing something wrong.
More confusing is that I created a simple reactive-transport model involving only 1 species and I can view and plot the model solutions for all the values solved for by the parametric sweep. However when I switch to the more complicated model I only get the last one. All the solver settings appear to be the same in both models.
Ivar KJELBERG
COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
21 apr 2011, 13:13 GMT-4
Hi
When you add a parameter sweep you get normality 2 solution cases, one for the last (active) one and one with all parameters, check you main plot node reference (1st entry) (sometimes you need to regenerate your solver sequence, if you have fiddled too much with it, then COMSOL might get lost ;)
Then when you have a nested parametric sweep or one parametric + 1 transient, if you postprocess via the Defined variable you get double nested tables. There you have a selection for the "inner solution" or the "outer solution" depending if you want to see the rows or the columns
--
Good luck
Ivar
Hi
When you add a parameter sweep you get normality 2 solution cases, one for the last (active) one and one with all parameters, check you main plot node reference (1st entry) (sometimes you need to regenerate your solver sequence, if you have fiddled too much with it, then COMSOL might get lost ;)
Then when you have a nested parametric sweep or one parametric + 1 transient, if you postprocess via the Defined variable you get double nested tables. There you have a selection for the "inner solution" or the "outer solution" depending if you want to see the rows or the columns
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
17 mag 2011, 10:11 GMT-4
Chris,
Have you had any success with this?
I am having a similar problem when I do a parametric sweep with a time transient problem. If I try to plot the time dependence of a variable (say Temperature) I can identify the solution set that supposedly has values for each of the parameter values. As Ivar points out, I can get the drop down menu that allows me to select parameter values, and the figure is fine if I select the last value of the parameter. However, if I select one of the earlier values, I get a plot with both axes ranging very close to zero and no data at all. If I select "all" the values, I see the single line on the plot, corresponding to the last parameter value.
Note that I have had many models that used parametric sweeps and frequency domain solutions and did not see the problem there.
Noa
Chris,
Have you had any success with this?
I am having a similar problem when I do a parametric sweep with a time transient problem. If I try to plot the time dependence of a variable (say Temperature) I can identify the solution set that supposedly has values for each of the parameter values. As Ivar points out, I can get the drop down menu that allows me to select parameter values, and the figure is fine if I select the last value of the parameter. However, if I select one of the earlier values, I get a plot with both axes ranging very close to zero and no data at all. If I select "all" the values, I see the single line on the plot, corresponding to the last parameter value.
Note that I have had many models that used parametric sweeps and frequency domain solutions and did not see the problem there.
Noa