Abstract
This paper deals with the regulation and stabilization of lightly damped systems under plant uncertainty. A controller consisting of a neural-network-based mode separator together with a modular passband tuning regulator is proposed. The mode separator generates the in-phase components of the plant utilized by the passband tuning regulator to maintain closed-loop stability and asymptotic regulation. This combined controller is robust and requires minimal plant information to operate. A synthesis procedure is outlined in this paper to summarize the design steps in a systematic manner. Finally, the application of the passband controller to the stabilization of a magnetic leadscrew is considered. Resonance is shown to be completely eliminated. The stabilization time of the transient response is also significantly reduced, confirming the feasibility of the controller.
Original language | English (US) |
---|---|
Pages (from-to) | 2169-2176 |
Number of pages | 8 |
Journal | IEEE Transactions on Industrial Electronics |
Volume | 55 |
Issue number | 5 |
DOIs | |
State | Published - May 2008 |
All Science Journal Classification (ASJC) codes
- Control and Systems Engineering
- Electrical and Electronic Engineering
Keywords
- Lightly damped systems
- Mode separation
- Neural network
- Passband control
- Regulation
- Stabilization
- Tuning regulator
- Vibration control