A self-optimising hybrid control system using parameter identification and frequency-response matching

This thesis deals with a type of control system that could adequately control non-stationary processes. The type of non-stationary process considered is typified by an aircraft, in which the pitch rate performance can vary considerably at a very fast rate from one set of flight conditions to another. controlling such a process with a conventional fixed-component control system would be inadequate, so that a fully adaptive control system has to be considered. In this thesis adaptation has been achieved by having a controller with a number of variable coefficients the settings of which are selected by a separate adaptive loop. This separate adaptive loop consists of two parts, the first of which identifies the performance of the process, and the second which calculates the controller settings that should produce the desired overall performance from the process. The bulk of the work described in this thesis was devoted to this adaptive loop, especially with regard to increasing its speed of adaptation and giving it in-built stability. [Continues.]