Abstract
We present a new method to control a linear induction motor (LIM) using dynamic neural networks. First, we propose a neural identifier of triangular form; this neural model has the structure of a nonlinear block contrallable form (NBC). Then, a reduced order observer is designed in order to estimate the secondary fluxes. Finally, a sliding mode control is developed to track velocity and flux magnitude. Simulations are presented to illustrate the applicability of the proposed scheme.
Original language | English |
---|---|
Pages (from-to) | 4041-4046 |
Number of pages | 6 |
Journal | Proceedings of the American Control Conference |
Volume | 5 |
State | Published - 2003 |
Externally published | Yes |
Event | 2003 American Control Conference - Denver, CO, United States Duration: 4 Jun 2003 → 6 Jun 2003 |
Keywords
- Block control
- Dynamic neural networks
- Identification
- Nonlinear systems
- Variable structure systems