Thesis

Nonlinear feedback controller design methods for actuator and sensor performance limited systems

Creator
Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2014
Thesis identifier
  • T13771
Qualification Level
Qualification Name
Department, School or Faculty
Abstract
  • This work describes the development of high performance, nonlinear controller design methodologies which aim to achieve high performance control with systems that have limited or low performance actuators and sensors. The first class of actuator constraints are actuators or sensors with slow dynamics with respect to the desired controller response. In order to achieve high performance control with these systems a nonlinear control algorithm was to be designed that provided more robust performance without compromising the controller's speed of response. The second class of problem involves actuators or sensors that are constrained by absolute limits. These are prevalent in almost all real systems and can be categorised as rate and amplitude limits. A control algorithm was to be designed that is able to operate for prolonged periods on either rate or amplitude limits without performance degradation. This was achieved by using a dynamic, controller output limiting design which aims to prevent the controller output from exceeding specified rate or amplitude limits. These controller designs were applied to control problems in aerospace and energy systems. Specifically, automatic flight control and automatic internal climate control. Both case studies involve control systems that are highly constrained by their actuation systems. In the case of internal climate control the actuation systems have very large inertias and there can be significant sensor delays. For flight control, the power of the actuator's driving control surfaces is heavily constrained as well as the physical deflection limits of the surfaces. By employing these control algorithms, the control performance and robustness of these systems can be significantly improved. This was demonstrated by simulations of heating and ventilation control of a modern office building and a missile flight control system. The contribution to knowledge, detailed in this thesis, is the development of nonlinear controller design methodologies which provide targeted solutions to some of the most widespread control problems encountered across a wide range of applications; namely, the problem of achieving high performance control with low performance, or limited, actuation or sensor systems.
Resource Type
DOI

Relations

Items