Dynamic Modeling and Advanced Control of Air Conditioning and Refrigeration Systems

Abstract

This dissertation makes contributions on both fronts and can be divided into two distinct parts. The first portion of the dissertation presents the development, simulation, and experimental validation of a first principles modeling framework that captures the dynamics of a variety of vapor compression cycles in a form amenable to controller design. These models are highly nonlinear, and require a nonlinear control strategy to attain high performance over the entire operating envelope. To this end, a gain-scheduled control approach based on local models and local controllers is presented that uses endogenous scheduling variables. This comprises the second portion of the dissertation, where a theoretical framework for designing gain scheduled controllers, tools for analyzing the stability of the nonlinear closed loop system, and experimental evaluation of advanced control strategies for vapor compression systems is presented. These results demonstrate that while linear control techniques offer significant advantages versus traditional a/c control systems over small ranges, the gain-scheduled approach extends these advantages over the entire operating regime.