welche Parameter für PI Regler

[tbi]

or you look here: https://knx-user-forum.de/knx-eib-fo...tml#post215157

Also to warm there and I expect the sun is much better at your house, Raymond.

regards

Tbi

[Gaston]

Is it that what they mean by "continuous PI" ?
But in the 3K/60, what is this 60 exactly ?

Hi Raymond,

The term "Continous regulation" is used for a type of regulation that constantly adaptes the regulated value and thus adapting the energy flow, as opposed to discontinous regulation that only able to distinguish between "heating needed" and "no heating needed".

KNX regulators are not really continous but discrete regulators as they are digital implementations. But that distinction is marginal if short computation intervals are used.

In 3K/60 the 60 is the time required for the I-Regulator to adapt the valve position about the same amount of the P-Regulator value.

Example:

Current temperature: 20°C
Target temperature: 21°C
Parameters: 4K/180

Now assume that the regulator is just starting now with these values (I=0). In this case P=100/4*(21-20)=25%

In this case the 180 would yell for another 25% in 180 minutes, meaning that after 3 hours we will have the valve at 50%, hours 75%,...

However these are theoretical values since it assumes that the room temperature does not change during that time. This is a standard method used to show how regulators behave (called step response).

[Warichet]

Oh yes, you have. It's 0 as it is by every RTR on the market. But it should be somewere around 70% or 80%. The value which would be nessasery to keep the wanted temperature. RTRs are starting to go down to 0, as soon as they reached the wanted temperature.

I've attached a few screenshots of my RTR.
Can you please tell me which one it is ?

I'm not sure to understand the "Arbeitspunkt" concept, can you elaborate a bit on this please ? i.e. the value is based on what ? just experiment i.e. trial & error ? What's the problem of leaving it at the default value ?

[Gaston]

I've attached a few screenshots of my RTR.
Can you please tell me which one it is ?

I'm not sure to understand the "Arbeitspunkt" concept, can you elaborate a bit on this please ? i.e. the value is based on what ? just experiment i.e. trial & error ? What's the problem of leaving it at the default value ?

The regulator bias point (=Arbeitspunkt) cannot be changed. The bias point of a regulator is the point at which the system is in balance. This means the valve position at which the wanted temperature is constantly kept if no change is brought into the system.

The regulator will at some point in time reach its bias point. However if for instance you start the standard heating regulator with th esystem currently in balance (meaning current temperatur = target temperature) it will start at 0% which is for sure not the correct bias point.

It would be much better to have a regulator that starts at the correct bias point. And even then there is room for improvement in regard of the starting point.

Computing the correct bias and thus starting point is not trivial but it would be much better to have it at an arbitrary 50% than at 0%.

[Warichet]

It would be much better to have a regulator that starts at the correct bias point. And even then there is room for improvement in regard of the starting point.

Computing the correct bias and thus starting point is not trivial but it would be much better to have it at an arbitrary 50% than at 0%.

Thank you for the explanation.
I understand what you are saying, but don't see the parameter I should change.
Do you mean on picture T4 "adjustement value for room temperature measurement" ?

[Uwe!]

, but don't see the parameter I should change.

This bias point (thanks for the word, Gaston) can NOT be changed! This is what i tryed to say. It should be changeable, but it isn't!

[Warichet]

This bias point can NOT be changed!

OK, thanks.
It looks like I'm unable to find something that doesn't exist

Would you be so kind to have a look at the screenshots and check if I didn't make too many dumb things ?

[Micha]

siehe Dokumentation Touch_IT C3. Hier kann man den Arbeitspunkt einstellen. Interessant die Begründung für die Verwendung 50% als AP

"Dieser Regler kann vorteilhaft in korrekt entworfenen Systemen verwendet werden"

Gruß Micha

[Gaston]

siehe Dokumentation Touch_IT C3. Hier kann man den Arbeitspunkt einstellen. Interessant die Begründung für die Verwendung 50% als AP

Ja, und dann sowas:

"Verwenden Sie diesen Parameter, wenn Sie lediglich einen P-Regler verwenden (I=0)."

[Micha]

in der Gleichung für den symetrischen AP ist der I-Anteil trotzdem enthalten. Ich vermute, dass er dann auch berücksichtigt wird. Das ist doch erstmal positiv. Ich werde das gelegentlich mal austesten.

Gruß Micha

[Uwe!]

Would you be so kind to have a look at the screenshots

Done!
Picture 12, Scaling Setpoint:
not sure, what this is. 3K?

Picture 14, Change of Setpoint...:
0.5K is quite much, but absolutely ok, if it fits your need. I have 0.2K

Adjustment:
0K. Normally RTRs gauge quite wrong, so you should check with a good thermometer and adjust

So everything regarding the PI seems fine.

[Gaston]

in der Gleichung für den symetrischen AP ist der I-Anteil trotzdem enthalten. Ich vermute, dass er dann auch berücksichtigt wird.

Es ging ja nicht um die Formel sondern um die Rechtfertigung, bzw. die Empfehlung

[enertegus]

Zurück zum PI Makro:
So wird das ausschauen - wird als Standardseite im Visuassistenten konfigurierbar sein und benötigt einen Speicherplatz im Flash für die Daten. Der "Raumkontroller" im EibPC benötigt dann nur die Isttemperaturmessung.

[Uwe!]

dann macht doch bitte gleich noch den Arbeitspunkt einstellbar mit rein!

[enertegus]

dann macht doch bitte gleich noch den Arbeitspunkt einstellbar mit rein!

Was meinst du damit?