[TRNSYS-users] genopt

[Jochen Doell]

Hello Johannes,

Well, Genopt tries to optimize the cost function. This means, it tries 
to get the lowest possible value for "Ppumpe_eff" by varying "factor". 
Looking at the equation "Ppumpe_eff = Phydr/0.25*factor", you do not 
need Genopt to get a result...
I assume you have set the lower boundary for "factor" to 0. So this is 
the correct value to get the "optimal", i.e. minimal value for 
"Ppumpe_eff" (0, too).

Christian was not completely wrong with his assumption about the 
one-dimensional optimization: Though Genopt algorithms can be applied, 
it usually is not necessary to use an optimizer. You need an optimizer 
if you have more than one variable which have conflictive influence on 
your result; or if you have one variable of unknown or difficult to 
mathematically express influence (e.g. the influence of the heat storage 
volume on the annuity of a solar dhw system: too small --> high aux 
energy consumption, too big --> high investment).

Best regards,
Jochen


Am 20:59, schrieb Gantner:
> Dear all,
>
>
>
> I am trying to use Genopt in order to optimize the massflow of a
> heatpump.
>
> Everything works out well and Genopt is calculating. But the result
> after every
>
> Simulation step is “0.00”. At the end the optimum is the input value.
> First I thought
>
> the problem is that the cost function is not depending on the variables
> so I tried
>
> a equation like this:
>
> Ppumpe_eff = Phydr/0.25*factor
>
>
>
> Where “factor” is the only variable. But still the result after the
> different simulations
>
> Is “0.00” and the optimum is the input value.
>
>
>
> Does anyone know what is my mistake?
>
> Would really appreciate your help.
>
>
>
> Thanks in advance,
>
> Best,
>
> Johannes Gantner
>
>
>

-- 
_________________________________________________

Dipl.-Ing. (FH) Jochen Döll
Division Thermal Systems and Buildings

Fraunhofer-Institut für Solare Energiesysteme ISE
Heidenhofstrasse 2, 79110 Freiburg, Germany

Phone: +49 (761) 4588-5468 Fax: +49 (761) 4588-9000
jochen.doell at ise.fraunhofer.de
www.ise.fraunhofer.de
_________________________________________________