Find Plant Zeros: Can We Determine Stability from Output Y?

  • Thread starter Thread starter ohmatlab
  • Start date Start date
  • Tags Tags
    Plant
Click For Summary
SUMMARY

This discussion focuses on determining the stability of an unknown plant H using only output data Y. The original poster raises the challenge of identifying the zeros of plant H without prior knowledge of its characteristics. It is established that if the zeros are not strictly located in the left-hand side s-plane, the plant is classified as non-minimum phase and potentially unstable. The poster proposes using the Fast Fourier Transform (FFT) on the output data Y to identify these zeros, although they encounter difficulties in this approach.

PREREQUISITES
  • Understanding of control systems and plant stability concepts
  • Familiarity with the Fast Fourier Transform (FFT) technique
  • Knowledge of s-plane analysis in control theory
  • Experience with signal processing and data analysis
NEXT STEPS
  • Research methods for identifying zeros and poles in control systems
  • Learn about non-minimum phase systems and their implications
  • Explore advanced FFT applications in signal processing
  • Investigate alternative techniques for system identification from output data
USEFUL FOR

Control engineers, signal processing specialists, and researchers in system identification who are looking to analyze plant stability from limited output data.

ohmatlab
Messages
2
Reaction score
0
Good afternoon. I want to ask a question. If we knew the data X through the plant H, and the output is Y. All we have is measurement data Y. Can we find the zeros of the plant H from the output data Y?
Thanks and have a nice weekend.
01/09/2009

__________________________________

Hi! Friends, I apologize for unclear descriptive, and thanks for remind me.
I just consider a engineer design question. For a plant, if the numerator have the zeros not strictive located on the left hand side s-plane ( maybe on the right hand side s-plane or just on jw axis ), we can said this plant is non-minimum phase. And it will be unstable. Therefore, for normal situation, from the knwledge of the plant, we can define this plant is stable or not.

Now I consider another problem, sometimes we can not get the plant details. Maybe I just only get the measurement data Y. Y is the output data from unknow plant or channel " H ". I assumed the input data X through the unknown plant or channel H, and only knowkedge we have is just measurement data Y. Because we don't have any knowledge of the plant H, therefore, we can not define if this plant stable or unstable from its poles and zeros. Only information we have is just output data Y.

Now, my question is that if I only have data Y. I just knew data Y is the output from a unknow plant H. reagrding this limited information, do any betters know how to find the zeros of the plant to define the plant atable or not?

I thought one immature idea. If I " fft " the Y. Setting Yfft=fft( Y). Then I set Y1=abs(Yfft). Therefore, every value of the Y1 must be greater or equal the " 0 ". Only value=0 is when meeting the plant zeros. So I can find zeros of the plant. I tried to do this, but I meet difficulties. Do any betters have any better idea?

Thanks Brewong and Danger reply my message. Have a nice weekend.

01/10/2009
 
Last edited:
Engineering news on Phys.org
I don't really understand the question. Could you be a bit more descriptive? Are you talking about a manufacturing process with its input and output?
 
I'm not as articulate are Brewnog, so I'm just going to say :confused:
 

Similar threads

  • · Replies 11 ·
Replies
11
Views
3K
Replies
2
Views
2K
Replies
6
Views
2K
  • · Replies 2 ·
Replies
2
Views
6K
  • · Replies 2 ·
Replies
2
Views
6K
Replies
31
Views
4K
  • · Replies 4 ·
Replies
4
Views
2K
Replies
1
Views
2K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K