Multivariable gas turbine engine controller design using quantitative feedback theory
Abstract
Complex thermo-fluid systems pose many difficult engineering problems in the area of automatic control system design. Typically, thermo-fluid systems exhibit nonlinear behavior, a high degree of process uncertainty, as well as transport lag and other time delay behaviors. The jet aircraft engine is a good example of one such system where the linearized dynamical behavior varies widely over the ranges of power code, altitude and flight Mach number. The customer demands high system performance levels, efficiencies, and safety margins with the aid of automatic control despite the high level of system complexity and uncertainty. The development of suitably powerful controller design techniques for this type of problem is therefore of considerable importance. There exist many distinct controller design techniques capable of producing control laws that are robust in the sense that they are immune to variations in the linearized models used to characterize non-linear system dynamical behavior. Quantitative Feedback Theory (QFT) is one such technique that seeks to break down multiple-input multiple-output system control problems into a series of single-input single-output (SISO) problems. QFT relies on a graphical synthesis approach to controller design, converting the performance and stability objectives into gain-phase bounds on the magnitudes of the individual SISO loop transmission functions. It is in this spirit that this work is undertaken, replacing much of this heuristic development in favor of a more thorough, mathematically rigorous approach, while retaining the powerful graphical design method. This thesis contains a motivation for the use of robust control within the frame-work of turbo-mechanical systems, a review of the historical development of multivariable aircraft engine control, a concise mathematical description of the control problem, and a mathematically rigorous mechanism for carrying out the design of robust control laws. An application of the theory is then brought to bear on a particular multivariable engine control problem; that of the General Electric GE16 after-burning, gas turbofan aircraft engine.
Degree
Ph.D.
Advisors
Nwokah, Purdue University.
Subject Area
Mechanical engineering|Automotive materials
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.