Solving PDE with COMSOL Script: Parametric Study & Initial Solution

[py_engineer]

Hi,

I am using COMSOL Script and I am trying to figure out something.

Let's say I am only solving one PDE which solution is 'phi'. And I am doing a parametric study, so I am solving it for different biases 'Va' from 0V to -1V. My geometry is simply a rectangle.

As part of the PDE, I have a parameter 'equ.f' which I set to be equal to 'Rec'. 'Rec' is defined in 'fem.globalexpr'.

For Va=0 (initial solution), I set 'Rec' to be 0. But then, for all the following steps, I want 'Rec' to be a function of the solution 'phi' found by COMSOL at Va=0. Is there a way to do that? Basically, I want to be able to use the initial solution in order to define an expression which is to be used in all the following steps..

I tried to solve for Va=0, then save the FEM structure, and use it to define 'Rec' by doing something like Rec=posteval(fem,'phi','solnum'1), and then solve again for the following steps. But it seems the format is not compatible. I get an error like 'cannot evaluate 'Rec', etc. So it didn't work.. (or maybe I didn't do it right..)

Any ideas would be GREATLY appreciated! Thanks!

 

[PerennialII]

May be I'm not getting this all (please educate ), but have you tried looking at solver manager and what's in there with respect to initial solution, storing a solution, using the current solution as an initial value etc.?

 

[py_engineer]

Thanks for your answer! I actually looked a little bit into the Solver Manager, and the Help docs on this. And from what I understand, I can use a previous solution as an initial solution for the next step.

But what I'm trying to do, is to use a previous solution in order to define an equation parameter which will be used in the next step.. So it's a little bit different and, although I need to look more into COMSOL Help, I haven't found how to do that yet..

 

[PerennialII]

In principle it sounds something which you might be able to do by introducing a nonlinear dependency (or history dependency might be a term for this case), or an additional "coupling" perhaps (not perhaps in the most traditional sense of a coupling, but occasionally use them [a PDE form for example] to introduce dependencies over time history). There are a bunch of ways of defining different sorts of expressions you can use in your equations (all the expression types, global equations, function definitions, couplings, script m - files etc.) so would think something might be appropriate (if nothing a tuned comsol script can do a whole lot of things to a model like read initial values etc.).

 

[SKH]

Hi,

I was wondering if you ever figured out how you could do this. I have the same problem and I have no clue on how to solve it. I am able to save final solution of the first solver as the initial value for the second one but in my second solver I have expressions that are dependent on this initial value and I am not able to make them work.

Any Ideas??

Thanks!