City Research Online

Robust control design using quantitative feedback theory

Nandakumar, R. (2008). Robust control design using quantitative feedback theory. (Unpublished Doctoral thesis, City University London)

Abstract

This thesis presents results in the area of robust control design using the Quantitative Feedback Theory (QFT) methodology. The thesis outlines the main philosophy and the various stages of this design approach and develops computational tools for carrying out a systematic design of uncertain feedback control systems in this framework using techniques of graphical design and computational geometry. Further, the thesis develops optimisation-based control design methods which can be carried out within a QFT computer-aided-design environment, with main emphasis on automatic loop-shaping. Two main design algorithms are proposed. The first involves the robust design of uncertain systems using fixed-structure controllers (PID, phase-lead/lag etc) which are widely used in practice. The second method is based on linear programming, and attempts to design the optimal controller in the frequency domain, subject to robust stability and performance specifications, augmented by additional realisability constraints based on the Bode gain/phase integral relationship. The proposed methods are tested via simple design examples and a detailed case-study involving the design of a non-linear hydraulic actuator. Simulation results of the closed-loop system demonstrate the applicability of the proposed techniques for the effective design of uncertain complex systems.

Publication Type: Thesis (Doctoral)
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Departments: Doctoral Theses
School of Mathematics, Computer Science & Engineering > Computer Science
Doctoral Theses > School of Mathematics, Computer Science and Engineering
URI: http://openaccess.city.ac.uk/id/eprint/19631
[img]
Preview
Text - Accepted Version
Download (15MB) | Preview

Export

Downloads

Downloads per month over past year

View more statistics

Actions (login required)

Admin Login Admin Login