Re: [TRNSYS-users] Questions about PID 23

[Marcus <jones.0bj3@gmail.com>]

I have used this type in the control of coils and fans, but it's been a year since I looked into it... 

The key thing to remember is that TRNSYS is based on discrete timesteps, and this does have an effect on the controller cycling, as you have noticed. If your timestep is large, i.e. 1 hour, the temperature in the zone has had 1 hour to rise and therefore if the controller was ON previously, it would turn fully OFF for the next hour, and then fully ON for the next, etc.... 

Therefore, as I recall it, we first reduced the Proportional gain to close to zero and the integral gain at least equal to the timestep (essentially a pure integral control), and simulated with a very short (maybe 5 seconds) timestep. Once that works, you can try to increase the proportional gain and decrease integral, if the response is too slow. So in your case, keep the Kp = 10^-4, but increase your Ti. 

Cheers,

Marcus
-- 
Marcus Jones, LEED AP, M.Sc. 
Freelance energy consultant
Vienna, Austria

On Wed, May 11, 2011 at 4:54 PM, Matt Duffy <duffy@tess-inc.com> wrote:

Hello Erwan,

I apologize, but I do not have much experience with the component. However, there are two examples that demonstrate the use of the component in the following directory, ".\Examples\Feedback Control".

Best regards,

Matt Duffy

----- Original Message -----
From: "LIBIOT, Erwan" <e.libiot@bouygues-construction.com>
Sent: Mon, May 9, 2011 3:49
Subject: [TRNSYS-users] Questions about PID 23

Hello everybody,

I'm a new trnsys user…

 

Currently, I simulates an heating coil (type 670) with fan (type 662) regulated by a PID type 23. I fixed an input temperature outside, incoming air flow, HR, pressure and inlet fluid temperature of heating coil.

 

My goal is to analyze the functioning of the PID 23. It regulates the heating coil fluid flow rate by comparing the output temperature of the fan to a set temperature

(gain Kp = 10 ^-4 and Ti = 10 ^-4; Td = 0).

 

By several simulations, I noticed that:

 

_The time step simulation is significant in the convergence of the regulator (Doesn't it perform iterations between each time step?). A large time step does not allow the system converge.

 

 

_Acting on the gain of the PID, convergence can be possible! Do you know why there is a dependency of these two parameters?

 

 

_PID values, gain and Ti (integration time) are not to my knowledge in the orders of magnitude of value of real PID of this system... why this difference in value?

 

If you have some information and/or model working fine about the PID 23...

 

Thanks!!!

 

LIBIOT Erwan

Bouygues Batiment International

service mechanical
01.30.60.56.66
06.78.02.81.97

e.libiot@bouygues-construction.com

 

--
Les donnees et renseignements contenus dans ce message sont personnels, confidentiels et secrets. Toute publication, utilisation ou diffusion, meme partielle, doit etre autorisee. Si vous n'etes pas le bon destinataire, nous vous demandons de ne pas lire, copier, utiliser ou divulguer cette communication. Nous vous prions de notifier cette erreur a l'expediteur et d'effacer immediatement cette communication de votre systeme.

Any data and information contained in this electronic mail is personal, confidential and secret. Any total or partial publication, use or distribution must be authorized. If you are not the right addressee, we ask you not to read, copy, use or disclose this communication. Please notify this error to the sender and erase at once this communication from your system.

---

_______________________________________________
TRNSYS-users mailing list
TRNSYS-users@cae.wisc.edu
https://mailman.cae.wisc.edu/listinfo/trnsys-users

_______________________________________________
TRNSYS-users mailing list
TRNSYS-users@cae.wisc.edu
https://mailman.cae.wisc.edu/listinfo/trnsys-users