Abstract
This paper proposes an online tuning method for the parameters of a fuzzy neural network using variable structure systems theory. The proposed learning algorithm establishes a sliding motion in terms of the fuzzy neuro controller parameters, and it leads the error towards zero. The Lyapunov function approach is used to analyze the convergence of the weights for the case of triangular membership functions. Sufficient conditions to guarantee the convergence of the weights are derived. In the simulation studies, the approach presented has been tested on the velocity control of an electro hydraulic servo system in presence of flow nonlinearities and internal friction.
| Original language | English |
|---|---|
| Title of host publication | ASCC 2011 - 8th Asian Control Conference - Final Program and Proceedings |
| Number of pages | 6 |
| Publication date | 2011 |
| Pages | 617-622 |
| Article number | 5899143 |
| ISBN (Print) | 9788995605646 |
| Publication status | Published - 2011 |
| Externally published | Yes |
| Event | 8th Asian Control Conference - Kaohsiung, Taiwan, Province of China Duration: 15 May 2011 → 18 May 2011 |
Conference
| Conference | 8th Asian Control Conference |
|---|---|
| Country/Territory | Taiwan, Province of China |
| City | Kaohsiung |
| Period | 15/05/2011 → 18/05/2011 |
| Sponsor | Ministry of Education of the People's Republic of China, National Science Council, Bureau of Foreign Trade, National Yang Ming Chiao Tung University |