Digital control using weighting sequence models

Hepburn, George Alan (1970) Digital control using weighting sequence models. PhD thesis, University of Glasgow.

Abstract

A brief review of identification techniques for linear systems is presented, special emphasis being given to the relatively new statistical techniques which yield an identification in terms of a weighting sequence model. It is noted that the use of such models in controller design has received little attention, the few exceptions to this being discussed, and it is concluded that further development in this field will prove useful. An outline proposal for a scheme to simultaneously identify and control a general multi-input, multi-output system whose paths may be represented by linearised models, is givan. The proposal envisages the implementation of a combination of feedforward and feedback controllers, both of which utilise weighting sequence models of the system. The method used in carrying out the identification is described, particular attention being given to the elimination of certain errors inherent in the basic technique. The effectiveness of the correction procedures is demonstrated by example. A feedforward controller is described which effectively eliminates the effect of measurable disturbance inputs in a system with one controlled output, subsequent considerations being limited to this case. The limitations of this controller are investigated by analytical techniques and examples of simulated operation. A feedback controller, using techniques similar to those devised for the feedforward controller is introduced to minimise the effects of unmeasurable inputs, and errors in the feedforward controller. An analysis of the stability of this system is given and the important aspects of its performance are demonstrated by means of simulation. The program used to simulate the operation of the system is presented, together with a general description. The extent to which it has been possible to implement the original proposals and the limitations which it has been found necessary to introduce at various stages, are summarised. In pctrticular, the failure to achieve satisfactory simultaneous identification and control is noted. Suggestions are made for a logical continuation of the work.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Additional Information: Adviser: H A Barker
• Keywords: Electrical engineering
• Date of Award: 1970
• Depositing User: Enlighten Team
• Unique ID: glathesis:1970-74153
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 23 Sep 2019 15:33
• Last Modified: 23 Sep 2019 15:33
• URI: https://theses.gla.ac.uk/id/eprint/74153