Date of Award
Spring 2022
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Electrical Engineering (ENAS)
First Advisor
Narendra, Kumpati
Abstract
The study of adaptive systems concerns the identification and control of dynamicalsystems in the presence of uncertainties. In the past seven decades, research on adaptive systems has primarily centered around the identification and control of linear dynamical systems with constant parameters. The use of multiple models to judiciously combine information from various sources has proven to be a powerful technique for improving adaptive behavior (in terms of speed, accuracy, and robustness) in these systems. However, the main interest of adaptive structures has always been in the identification and control of systems with unknown time-varying parameters. Situations which call for adaptation in such time-varying environments are increasing across many nontechnical and technical fields. These include medical emergencies, trading on the stock market, production processes, and aerospace systems. The principal objective of this thesis is to investigate the adaptive identification and control of such systems. Since time varying parameters can change in infinite ways, it is extremely unlikelythat one can find a universal solution to such problems. A more realistic way to tackle this problem is to classify time variations into subclasses and discuss each subclass separately. This thesis addresses identification and control problems in three main subclasses of time-varying systems: systems with periodically varying parameters, systems with piecewise constant parameters, and systems with continuously varying parameters in a finite region. For each subclass, the application of the multiple models concept is investigated and subsequently, new adaptive identification and control strategies are proposed. Finally, the stability of the overall system when using the proposed structures is investigated theoretically and simulation discussions are provided to demonstrate the effectiveness of the proposed structures.
Recommended Citation
Esfandiari, Kasra, "Adaptive Identification and Control of Time-varying Systems Using Multiple Models" (2022). Yale Graduate School of Arts and Sciences Dissertations. 748.
https://elischolar.library.yale.edu/gsas_dissertations/748
