Model reduction of linear and nonlinear systems using balancing methods

Abstract

This thesis presents a review of the existing body of knowledge pertaining to model reduction using balancing techniques. A simple linear system is studied in a noisy environment with noisy sensors and linear balancing techniques are performed. Following, a more complex system in that of a spring-mass system with nonlinear damping is studied using the analysis set forth by J.Scherpen [11] and A.Newman [8]. The connection between linear and nonlinear balancing techniques is established and possible existant methods that reduce the complexity of the analysis involved are presented. We consider nonlinear filtering theory in the context of 2-point Vortex motion with inroads made towards the prospect of data fusion in regards to the data provided by a number of Lagrangian tracers.