[CCoE Notice] PhD Defense Announcement - Shirazi

[Khator, Suresh]

LINEAR PARAMETER VARYING CONTROL AND LMI OPTIMIZATION
OF NONLINEAR SYSTEMS

Farzad A. Shirazi

Committee members:
Dr. Karolos M. Grigoriadis, Dr. Matthew A. Franchek, Dr. David Zimmerman,
Dr. Gangbing Song, Dr. Daniel Viassolo and Dr.  Javad Mohammadpour

November 15th at 2 pm, Large Conference Room,
Mechanical Engineering Department
Abstract
The core idea in this dissertation is to explore state-of-the-art applications of LPV systems in controlling nonlinear plants and the performance comparison with conventional control methods. Nonlinear systems can be cast into LPV framework and well-established LPV controller synthesis techniques are applied to them using linear matrix inequalities (LMIs) which can be solved efficiently using polynomial-time algorithms. We investigate the stability and performance analysis of micro-electromechanical system (MEMS) actuators, smart base isolation of structures using magneto-rheological (MR) dampers and multi-input-multi-output (MIMO) control of large-scale horizontal axis wind turbines.

An H∞ inverse controller based on the modified LuGre model, and a dynamic output-feedback LPV controller based on the modified Bingham model are proposed and experimented for seismic protection of a two-story model building. A lumped model is developed for wind turbines and validated in time and frequency domains using FAST (Fatigue, Aerodynamics, Structures, Turbulence) code. In the end, a novel approach towards integrated structure and LPV controller design of wind turbines is formulated and solved by iterative LMI problems to improve the closed-loop performance in terms of minimizing structural fatigue loads and maximizing power capture.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://Bug.EGR.UH.EDU/pipermail/engi-dist/attachments/20111115/ebcbe5af/attachment.html