Abstract
In this article, an adaptive control approach - Adaptive Generic Model Control (AGMC) for a class of nonlinear time-varying processes with input time delay is proposed. First, a nonlinear state predictor (NSP) is introduced, which extends the conventional generic model control (GMC) to a class of nonlinear processes with input time delay. Then a class of nonlinear time-varying processes with input time delay is further considered. A modified strong tracking filter (MSTF) is adopted to estimate the time-varying parameters of the nonlinear processes, and the state estimates are then utilized to update the plant models used in the NSP and MSTF, this results in an adaptive generic model control scheme for a class of nonlinear time-varying processes with input time delay. A modified mathematical model of a three-tank-system is used for computer simulations, the results show that the proposed AGMC algorithm is satisfactory, and it has definite robustness against model/plant mismatch in the measurement noise. © 2003 Elsevier Ltd. All rights reserved.
| Original language | English |
|---|---|
| Pages (from-to) | 517-531 |
| Number of pages | 15 |
| Journal | Journal of Process Control |
| Volume | 14 |
| Issue number | 5 |
| Early online date | 4 Nov 2003 |
| DOIs | |
| Publication status | Published - Aug 2004 |
| Externally published | Yes |
Bibliographical note
This work was mainly supported by the NSFC (Grant No. 60025307, 60234010, 60228001), partially supported by the national 863 program, RFDP (Grant No. 20020003063) and the national 973 program (Grant No. 2002CB312200) of China.Keywords
- Adaptive control
- Generic model control
- Modified strong tracking filter
- Nonlinear processes
- State predictor
- Time-varying processes