|Institution:||University of Illinois – Urbana-Champaign|
|Full text PDF:||http://hdl.handle.net/2142/44360|
Air Conditioning and Refrigeration (AC&R) systems are ubiquitous in modern society since they perform the key engineering function of transporting thermal energy from one physical location to another. In doing so they are able to change the condition of a defined spatial environment to prescribe a particular temperature and humidity. Increased economic and environmental concerns have placed greater emphasis on the energy efficiency of these AC&R systems. These concerns necessitate better component and system design as well as better operation of existing systems using advanced control techniques. The use of advanced control techniques, particularly for transient system operation, is the focus of this dissertation. Two significant challenges always exist in transient control of the AC&R systems: (i) developing control-oriented models that can capture the complex nonlinear thermodynamic behavior while balancing model simplicity with accuracy; and (ii) implementing control strategies that can achieve high performance and efficiency over a wide range of operating conditions. This dissertation makes contributions to these two fronts and is divided into two distinct parts. The first part of this dissertation introduces the development of a first-principles switched modeling framework, and presents simulation and experimental validation results in various AC&R system applications. These results show the validity of the modeling approaches to describe system transients under mode switching operations, such as cooling/heating mode switching and on/off cycling operation. An optimal operating strategy with on/off mode switching is illustrated as an example to demonstrate the effectiveness of the presented modeling tools in control design. To achieve high system performance under refrigerant phase transition conditions, a switching control strategy based on local models and local controller is introduced. This comprises the second part of this dissertation, where a first-principles invariant-order switched system with different operating models is formulated. Tools for designing controllers and analyzing the stability of the closed-loop switched system are presented. Simulation results demonstrate improved performance and efficiency with the presented control strategy compared to conventional control approaches in handling the nonlinear refrigerant phase transitions over a wide range of operating envelope.