A Comparison of Multivariable Control Design Techniques for a Turbofan Engine Control
Author(s) -
S. Watts,
Sanjay Garg
Publication year - 1995
Publication title -
volume 5: manufacturing materials and metallurgy; ceramics; structures and dynamics; controls, diagnostics and instrumentation; education; igti scholar award
Language(s) - English
Resource type - Conference proceedings
DOI - 10.1115/95-gt-258
Subject(s) - multivariable calculus , linear quadratic gaussian control , control theory (sociology) , robustness (evolution) , control engineering , turbofan , jet engine , control system , robust control , actuator , computer science , linear system , transfer function , engineering , control (management) , mathematics , automotive engineering , mathematical analysis , biochemistry , chemistry , electrical engineering , artificial intelligence , gene , aerospace engineering
This paper compares two previously published design procedures for two different multivariable control design techniques for application to a linear engine model of a jet engine. The two multivariable control design techniques compared were the Linear Quadratic Gaussian with Loop Transfer Recovery (LQG/LTR) and the H–Infinity (H∞) synthesis. The two control design techniques were used with specific previously published design procedures to synthesize controls which would provide equivalent closed loop frequency response for the primary control loops while assuring adequate loop de-coupling. The resulting controllers were then reduced in order to minimize the programming and data storage requirements for a typical implementation. The reduced order linear controllers designed by each method were combined with the linear model of an advanced turbofan engine and the system performance was evaluated for the continuous linear system. Included in the performance analysis are the resulting frequency and transient responses as well as actuator usage and rate capability for each design method. The controls were also analyzed for robustness with respect to structured uncertainties in the unmodeled system dynamics. The two controls were then compared for performance capability and hardware implementation issues.© 1995 ASME
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom