I should solve two different PDEs in one COMSOL model (first "Stationary" and second is "Eigenvalue") in such a way that the output of first Stationary Physics(the electrostatic potential) used in the second Physics(Schrodinger equation). I prefer to solve the problem in one "Study" not in two separate Studies. Is it possible? How should I change the solver setting?
Add the two study steps 'Stationary' (step1) and 'Eigenvalue' (step2) in to one study(say study 1). Then go the 'Study settings' of study1, step 2, and select 'Values of Dependent Variables'. In this section you have the option to specify where to get your initial values and values for variables not solved for. There you can specify to get these values from for example; study1,step1. See the attached screen clipping for the location where this can be specified.
Hi,
You can add two steps in one study: the first step is the 'Stationary' and the second one is the 'Eigenvalue'. For different steps, you can activate different physics. Input the values of variables from the first step to the second step.
I hope it is helpful.
Kind Regards,
Yan
Hi, Yan Peng
Thank you for your answer.
I guess, It is nesseccary to do some changes in" Values of Dependent Variables" or other parts of Study Setting.
The problem is that I don't know who to input the values of variables from the first step to the second step?
Add the two study steps 'Stationary' (step1) and 'Eigenvalue' (step2) in to one study(say study 1). Then go the 'Study settings' of study1, step 2, and select 'Values of Dependent Variables'. In this section you have the option to specify where to get your initial values and values for variables not solved for. There you can specify to get these values from for example; study1,step1. See the attached screen clipping for the location where this can be specified.
Hi,
Anybody can help me about the precision of calculation of eigenvalue in COMSOL?
I have noticed, COMSOL will not give me same eigenvalues from a nonlinear eigenvalue PDE when I just press “compute" on study twice, three times or more. It is necessary to have same precision in calculation, since I should run the model several times through MATLAB. I have check 100 times repeatedly. The answers are not so different but it leads to bad consequences in my calculation. I have changed the precision to 1e-9 but this problem did not removed. I cannot trust to such results!
- Badges
- Science topic