Abstract
An identifying algorithm for the structure and parameters of fuzzy model is proposed based on an orthogonal least square. First of all, fuzzy relation matrix of fuzzy model is confirmed by orthogonal least square. The validity of fuzzy rules is made certain by means of analyzing the efforts of orthogonal vectors in fuzzy model, and then the number of fuzzy rules and fuzzy rules could be acquired. In addition, the conclusion parameters of fuzzy model are confirmed by the orthogonal least square in order to optimize the structure and the parameters of fuzzy model. To demonstrate the validity of the paper, the proposed method is successfully used to build Box-Jenkins's model of gas furnace.
| Original language | English |
|---|---|
| Pages (from-to) | 1143-1146 |
| Number of pages | 4 |
| Journal | Jisuanji Xuebao/Chinese Journal of Computers |
| Volume | 27 |
| Issue number | 8 |
| State | Published - Aug 2004 |
Keywords
- Fuzzy competitive learning
- Fuzzy model
- Orthogonal least square
- System identification
Fingerprint
Dive into the research topics of 'Fuzzy identification of nonlinear system based on orthogonal least square'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver