This thesis describes the application of H∞ design techniques to the control of high performance aero-engines. The design study presented is practical and realistic, the work being motivated by problems that arise naturally in real engineering situations. The aero-engine is multivariable and highly nonlinear: the dynamics vary considerably with the thrust being produced, and with the altitude and forward speed of the aircraft. Moreover, there are operational constraints that must never be violated for reasons of safety: certain engine variables should always be limited to safe vales. Furthermore, not all the engine parameters to be controlled are directly measurable; instead a number of related measurements are available. A methodology is presented to choose from the available measurements, those that are preferable for feedback control. Different techniques of model reduction using balanced realizations are considered. Two illustrative examples are presented, and the methods compared in detail. Explicit state-space formulae for an H∞-based two degrees-of-freedom robust controller are derived in discrete time. The controller provides robust stability with respect to coprime factor perturbations, and a degree of robust performance in the sense of making the closed-loop system match an ideal reference model. Special attention is paid to the structure of the controller. It is shown that the controller consists of a plant observer, the reference model, and a generalized state feedback law associated with the plant and model states. Multi-mode control logic is developed to ensure that safety limits are never violated. Actual engine test results are presented for sea-level static conditions. All the different modes of operation are tested. Full flight envelope evaluation of the controller is carried out using a nonlinear engine simulation. The robust performance of the controller is demonstrated and comparisons made with existing engine control systems.

Robust multi-mode control of high performance aero-engines

In this paper, a 2 degrees-of-freedom multimode controller design for the Rolls Royce Spey turbofan engine is presented. The controller is designed via discrete time H∞-optimization; it provides robust stability against coprime factor uncertainty, and a degree of robust performance in the sense of making the closed-loop system match a prespecified reference model. Multimode control logic is developed to preserve structural integrity of the engine by limiting engine variables to specified safe values. A simple strategy for antiwindup and bumpless transfer between controllers, based on the Hanus anti-windup scheme (1987, “Conditioning Technique, A General Anti-Windup and Bumpless Transfer Method,” Automatica, 23(6), pp. 729–739) and the observer-based structure of the controller, is presented. The structure of the overall switched controller is described. Actual engine test results using the Spey engine test facility at Pyestock are presented. The controller is shown to perform a variety of tasks, its multimode operation is illustrated and improvements offered on existing engine control systems are discussed.

Design and Implementation of a Digital Multimode H∞ Controller for the Spey Turbofan Engine

In this thesis the main theme is to demonstrate the potential performance improvements of gas turbine engines that are brought about by using multivariable control systems. Particular emphasis is on designing such control systems using the well-established engine thermodynamic models since these models are considered as the true representations of engine thermodynamic process and enable engine variable geometry features to be easily incorporated and their effects studied.

Modelling and Control of Aircraft Gas Turbine Engines

Gas turbine engines and its related technologies represent one of the most efficient forms of propulsion and power generation. Their applications range from land-based power generation, ground-based vehicle propulsion, on-board power and propulsion sources for marine ships, to aircraft propulsion and auxiliary power systems.

Introduction to Aeroengine Controls

Advanced control often assumes implicitly that an aeroengine model is available. Modeling for both engine and control systems plays an important role in engine research and development: accelerate schedules while reducing cost and risks. For example, on control system design phase, models can be used for validating the corresponding control functions; on controller software and hardware design phase, models are used in fast prototyping, hardware-in-loop simulation, and semi-physical testing; on control system implementation phase, on-board models can be used for advanced control and fault detection, e.g. sensor analytical redundancy, flight/propulsion integration, direct propulsion control etc.

Aeroengine Nonlinear Modeling

This paper describes the design of a multivariable control for an aeroplane turbofan engine based on working line control. The method starts with the identified engine. Controller performance is illustrated by means of engine test results. The emphasis is on the control structure, which must include the essential features of a complete flight-worthy system, such as all the necessary limiters and a pilot's lever-driven range control, using scheduled gains capable of accommodating parameter variations across the flight envelope. Less importance is attached, at this stage, to the system response attributable to the design of the individual multivariable compensators. In the first instance they were designed as simple P+I controllers. Any changes in the design of the individual compensators can be accommodated within the overall scheme.<
 >

A.E. Sutton

Papers

Application of multivariable control to a turbofan engine