Discussion Forum

October 11, 2012 5:28am UTC in response to Marjan

Re: inverse initializing model

Hi

I can propose, once you have drawn your geometry and defined some physics + loaded at least the initial conditions to populate the FEM matrices (and allow you to do postprocessing) then you can select the domain (2D-axi surface) you consider, and integrate it with a surface integration, while you turn on the "2D-axi volumetric function which simply, implicitely multiplies the operand by 2*pi*r the loop length. You can then add a bolean function to your operand defining the height, such as *(z<H) where H is your height, defined in the correct coordinate frame (with the correct z0 initial height).

If you need to use these variables in the solution (and not only the postprocessing) you need to define the integration in the top Definition node, and define the olume as a variable in the same Definitions but under Variables node

In V4.3 we have now a series of new functions/operators, check the help, so you can perhaps also define the integration directly

--
Good luck
Ivar

Reply  |  Reply with Quote  |  Send private message  |  Report Abuse

October 11, 2012 2:22pm UTC in response to Marjan

Re: inverse initializing model

Hi

no problem, that was possible even back in 3.3 ;)

--
Good luck
Ivar

Reply  |  Reply with Quote  |  Send private message  |  Report Abuse