search

orcmkit / ORCmKit

ORC Modeling Kit: Tools for the modeling of ORC power systems in Python, Matlab and EES
MIT License
23 stars 18 forks source link

Add calibration example for expander semi-emp model in Matlab #7

Open davidT9 opened 7 years ago

davidT9 commented 7 years ago

Hi @RDickes @davideziviani

Thanks for sharing the great ORC model.

I was trying to run the expander model (ExpanderModel.m) using SemiEmp approach in Matlab. I also went through the mentioned paper (Lemort et al.).

I have a quesiton about the parameter identification process of the model.

If I am planning to use the provided expander model to simulate a diffrent scroll expander performance, how can I know the input parameters of the scroll expander is valid for simulation? Any suggestions for changing the parameters as there are more than ten parameters to be defined for different expander.

Thanks a lot.

RDickes commented 7 years ago

Hi @davidT9 ! You are welcome, hope this models will be usefull for you. Do you have any experimental or reference data for characterizing this new expander? If so, you can tune the parameters accordingly so as to fit you data. To do so, I personnaly use a derivative-free optimization algorithm (quite long to run, but it is stable) and I minimize the residuals committed onto the different model outputs. If you want to, I can add this calibration code in the library.

If you do not have data, you can keep the same parameters as proposed with just updating the built-in volumetric ratio and the machine swept volume. It would not be perfect, but it is still better than nothing.

Also, if you are using the ORCmKit for research purpose (and you are happy with it), could you make reference to it in your reports so we could get greater visibility? Thanks in advance!

Best regards,

Rémi

davideziviani commented 7 years ago

@RDickes : shall we add two examples of calibrating semi-emp model with Matlab and Python? I have added only examples for Pacejka equation

davidT9 commented 7 years ago

Hi @RDickes and @davideziviani

Thanks for your reply.

Sure, I will absolutely reference your model.

I only have only serval set experiment data (but only the speed, output and pressure value), not the expander details. So I think I would like to try the calibration code first to see whether I could get the identified parameter values for scroll expander.

It would be great if you could upload the calibration code in the library.

Thanks in advance.

T. D.

RDickes commented 7 years ago

@davideziviani yep, we can add this. I must finish some work on my side in the next days, then I'll add a reference calibration code for the matlab version.

davideziviani commented 7 years ago

Sure thing. Same here, I am a bit overwhelmed but I have the code ready

davidT9 commented 7 years ago

Dear @RDickes and @davideziviani,

Thanks in advance for uploading the calibration code.

RDickes commented 7 years ago

Hi David,

I'll upload the Matlab version by tomorrow! sorry for the delay..

Le 31/01/2017 à 15:32, davidT9 a écrit :

Dear @RDickes https://github.com/RDickes and @davideziviani https://github.com/davideziviani,

Thanks in advance for uploading the calibration code.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/orcmkit/ORCmKit/issues/7#issuecomment-276378303, or mute the thread https://github.com/notifications/unsubscribe-auth/AHOunuPGWkwuFcPuhLGREX3cYKeBKxHiks5rX0X3gaJpZM4LjLUW.

-- Rémi DICKES

PhD candidate/FRS-FNRS research fellow University of Liège - Faculty of Applied Sciences Energy System Research Unit Campus du Sart Tilman, Quartier Polytech 1 Allée de la Découverte 17 - Bât. B49 (P33) B–4000 LIEGE (BELGIUM) rdickes@ulg.ac.be +32 494.59.2253 +32 43.66.48.16

L'absence de virus dans ce courrier électronique a été vérifiée par le logiciel antivirus Avast. https://www.avast.com/antivirus

davidT9 commented 7 years ago

Sure, I will check the Matlab calibration code once you uploaded. Again, thanks @RDickes