Premium
Linear, parameter‐varying control and its application to a turbofan engine
Author(s) -
Balas Gary J.
Publication year - 2002
Publication title -
international journal of robust and nonlinear control
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.361
H-Index - 106
eISSN - 1099-1239
pISSN - 1049-8923
DOI - 10.1002/rnc.704
Subject(s) - turbofan , control theory (sociology) , robustness (evolution) , nonlinear system , jacobian matrix and determinant , computer science , mathematics , engineering , control (management) , biochemistry , chemistry , physics , quantum mechanics , artificial intelligence , automotive engineering , gene
This paper describes application of parameter‐dependent control design methods to a turbofan engine. Parameter‐dependent systems are linear systems, whose state‐space descriptions are known functions of time‐varying parameters. The time variation of each of the parameters is not known in advance, but is assumed to be measurable in real‐time. Three linear, parameter‐varying (LPV) approaches to control design are discussed. The first method is based on linear fractional transformations which relies on the small gain theorem for bounds on performance and robustness. The other methods make use of either a single (SQLF) or parameter‐dependent (PDQLF) quadratic Lyapunov function to bound the achievable level of performance. The latter two techniques are used to synthesize controllers for a high‐performance turbofan engine. A LPV model of the turbofan engine is constructed from Jacobian linearizations at fixed power codes for control design. The control problem is formulated as a model matching problem in the ℋ ∞ and LPV framework. The objective is decoupled command response of the closed‐loop system to pressure and rotor speed requests. The performance of linear, ℋ ∞ point designs are compared with the SQLF and PDQLF controllers. Nonlinear simulations indicate that the controller synthesized using the SQLF approach is slightly more conservative than the PDQLF controller. Nonlinear simulations with the SQLF and PDQLF controllers show very robust designs that achieve all desired performance objectives. Copyright © 2002 John Wiley & Sons, Ltd.