[Getdp] Induction_Motor_TimeSteping_Initial_Values

[Christophe Geuzaine]

Hi Nicola,

This is indeed a good idea. There are several ways you can impose the initial values. I think that the easiest in your case is to directly specify the constraint (of Init type) in the function space of the relevant quantities (the vector magnetic potential a; as well as the current Iz if you have circuit relations), using a list of values extracted from the first time-harmonic solution.

(The small technical issue that prevents the use of the classical "TransferSolution" method is that the frequency-domain formulation leads to different degrees of freedom (they are complex) than the time-domain formulation: you can thus not simply transfer values from one system to the other: the unknown vector is different.)

In order to import list of values for the Init constraint, you could do as follows:

- perform the time-harmonic simulation, followed by a PostOperation that
   - plots the (real part of the) vector potential in the domain, and export it as a "NodeTable" file ("Format NodeTable") 
   - saves the value of the current in a run-time variable (StoreInVariable $MyCurrent)
- add an "Init" type constraint for the relevant quantities in the function spaces used for the time-domain solution
   - for the vector potential, use "Value ValueFromIndex[MyPotential[]], where MyPotential[] = ListFromFile[]{MyNodeTableFile.txt}
   - for the current, use "Value $MyCurrent"

Let us know how this works out. If it works well we might want to add this to the template.

Christophe

> On 18 Mar 2016, at 15:04, Nicola Chiodetto <nicola.chiodetto at gmail.com> wrote:
> 
>  Dear GetDP users and experts,
> 
> I am willing to simulate an induction motor with time stepping technique.
> 
> I alredy used and studied the examples provided in the Electrical Machines folder available online.
> However, I need to avoid the initial electrical transient: in other words I need to set proper initial conditions. Since the Induction machines is intrinsically operating due to eddy currents reaction I need to perform a time-harmonic. At this point I was thinking to read the real part of the vector magnetic potential, over all the domains, and take its real part. However, I do not know how to force the conditions of global rotor current variables in the domain bars. Is it possible witout touching the source code? 
> Moreover, usually in literature it is proposed a time harmonic frequency analysis exploiting an effective permeability function, instead of a simpler linear one. Is it possible to do that? It could be again useful for determining better initial conditions for the time stepping simulation first point.
> 
> Thanks in advance.
> Best regards,
> 
> Nicola Chiodetto
>  
> _______________________________________________
> getdp mailing list
> getdp at onelab.info
> http://onelab.info/mailman/listinfo/getdp

-- 
Prof. Christophe Geuzaine
University of Liege, Electrical Engineering and Computer Science 
http://www.montefiore.ulg.ac.be/~geuzaine

Tetrahedron V, July 4-5 2016: http://tetrahedron.montefiore.ulg.ac.be
Free software: http://gmsh.info | http://getdp.info | http://onelab.info