An anamoly in PI Speed Control

Hey there,

I am working on a project for speed control of an Induction Motor. The control is a non linear one. But ultimately, I am using a PI controller to generate a torque reference.

The thing is, the current error is becoming zero, but my torque reference and load torque doesn't match. What could be the problem? Will be really helpful if someone can point out any mistake I could have made ......

Thank you in advance!
 
Hey there,

That would be a good thing to post a sketch/draft
of the block diagram used....

Also can you precise OEM/Model/Type /Maker of the mentionned equipment..
 
Ok thank you for the reply

We still do not know about converter configuration as i stated before ( an hardware schematic would be also welcome )

It seems that it is slid mode control that is used here ..I will have a search on this thema...

What type of current sensors used ...Speed sensor..

You saying that you send a Torque reference ...We see only "switching signals to inverter " at the output of the block diagram..
This block diagram is very basic..

The MPCC/CSCC block need to be developed ...if you want to get better overview on what is happening on this system..

Also what you mean by current error become zero ( which current are you referring...)
 
Sorry for not mentioning in detail

Actually I'm just in the initial phase. Have not decided on the hardware.

I'm considering the two-level three phase inverter (the usual bridge configuration with three legs and six switches).

No sensors as of now just simulation in Simulink. So, I don't think the sampling errors will be a problem.

Te* is the Torque reference I had mentioned. Ultimately the controller sends switch signals.

* It's speed error not current error. My bad...

Refer to P C Krause electric machinery for CSCC and Predictive control for power converters and drives - Jose Rodriguez and Patricio Cortes for MPCC.

I have developed these models in Simulink as an matlab function.

Thank you for your time. Hope you'll also get new ideas from these. And, do let me know if you find the solution. Thanks again!
 
So looks like this is Direct Torque Control (DTC) method applied here..

Hum...Let me study the reference that you share it is indeed a good document ...

No worries thats the right place to post...

Are you sure about system/model parameters congiguration on Matlab

A video support or a screenshot at least on the MatLAB PROGRAM woudl be welcome before giving any other recommandations/advises

Cheers
ControlsGuy25.
 
I have a read on the document that you mentionned ...

In Matlab code for FOC it is describing a insertion of a saturation block at output of PI controller in order to keep the amplitude of Torque reference to be in the limits of the simulated system/machine /pump/motor...

And for sure anti windup may be also added..

I dont see a saturation block at the output of the PI can you confirm...
Think also about antiwindup ...

Have a good read on Appendix B page 213 all is well described...

Any time!

James
 
What is purpose of Pl(z) ON THE BLCOK DIAGRAM ...

On the document it is showing a MATLAB CODE WITH ORQUE ESTIMATION FUNCTION ..PAGE 219

Did you notice this and take into account in the studied system

I imagine that MATLAB simulink got excatly what you need in their library

Will have a search on that library ..and come back asap..
 
I have a read on the document that you mentionned ...

In Matlab code for FOC it is describing a insertion of a saturation block at output of PI controller in order to keep the amplitude of Torque reference to be in the limits of the simulated system/machine /pump/motor...

And for sure anti windup may be also added..

I dont see a saturation block at the output of the PI can you confirm...
Think also about antiwindup ...

Have a good read on Appendix B page 213 all is well described...

Any time!

James
I do have anti-windup. Limited the Torque reference between 0 and 30 Nm
 
What is purpose of Pl(z) ON THE BLCOK DIAGRAM ...

On the document it is showing a MATLAB CODE WITH ORQUE ESTIMATION FUNCTION ..PAGE 219

Did you notice this and take into account in the studied system

I imagine that MATLAB simulink got excatly what you need in their library

Will have a search on that library ..and come back asap..
It's the PI controller. To track speed.
 
MPCC is similar to FOC but instead of SVM prediction algorithm of currents used. Prediction equation is obtained by discretising the Induction Motor equations. But in this model weighting factors are not included hence distortions in currents were more it may be improved by adding weighing factor or best prediction equation. Inverter Model used is 3 level NPC.
 
MPCC is similar to FOC but instead of SVM prediction algorithm of currents used. Prediction equation is obtained by discretising the Induction Motor equations. But in this model weighting factors are not included hence distortions in currents were more it may be improved by adding weighing factor or best prediction equation. Inverter Model used is 3 level NPC.
Yeah, instead of FOC I've used CSCC. And instead of SVM it's MPCC

For current tracking, a weighting factor is not necessary as it's the only constraint I'm keeping on the system. The flux constraint is taken care of by the CSCC.
 
Top