| Paper Details: | Downloads: 766 |
| Serial Number: | P1110906635
|
| Title: | Optimal T-S models for identification of nonlinear systems from input-output data
|
| Authors: | Hamid Ouakka and Ismail Boumhidi
|
| Abstract: | The main problem for identification algorithms, based on fuzzy clustering, is that the optimal number of clusters must be known and fixed in advance . In general, for each clustering aim, the correct number is determined using knowledge of the process or by some data driven techniques.
In this paper, we present a new approach for automatically generate the optimal number of cluster needed for properly approximating an unknown nonlinear black box system with Takagi-Sugeno models directly from the input-output data structure. An approximation model of the system is built by fitting data to a polynomial equation, a preliminary decomposition of the data is realized based on detection turning points, maxima and minima,of the function. Then a merging technique is proposed to optimize the identified number of clusters.
The performance of the proposed method is evaluated for both quality of clustering and fuzzy modeling with first order Takagi-Sugeno systems using GK fuzzy clustering and identification algorithm .
|
| Keywords: | Fuzzy clustering, optimal clusters number, nonlinear system, polynomial interpolation, Takagi-Sugeno models, GK algorithm
|
| Journal/Conference: | International Journal of Automatic Control and System Engineering
|
| Volume: | 9
|
| Issue: | 1
|
| Submission Date: | 2/5/2009 12:00:00 AM
|
| Review Date: | 2/17/2009 12:00:00 AM
|
| Publishing Date: | 3/12/2009 12:00:00 AM
|
| Article Downloads: | 766
|
| Download: |
|