Date of Award

Spring 1-1-2012

Document Type


Degree Name

Master of Science (MS)

First Advisor

Michael Brandemuehl

Second Advisor

Jay Burch

Third Advisor

Gregor Henze


The purpose of this research is to determine an optimal control strategy for an air conditioner that simultaneously uses liquid desiccant dehumidification and indirect evaporative cooling. The advantages posed by liquid desiccant dehumidification are numerous: elimination of roughlysignificant reduction (80-90%) ofin electricity consumption, air pollutant removal, potential to use low-grade energy use for regeneration, and avoiding a need for excessive cooling and reheating of process air to achieve dehumidification. The National Renewable Energy Laboratory in Golden, CO has developed an innovative air conditioner that combines liquid desiccant dehumidification and indirect evaporative cooling, referred to as Desiccant-Enhanced Evaporative Air Conditioning, or DEVap. In the first stage of the device, process air is dehumidified with a liquid desiccant film and simultaneously cooled with an evaporatively-cooled airstream. In the second stage of the device, another evaporatively-cooled airstream removes sensible energy from the process air without changing the humidity of the process air. This second evaporatively-cooled airstream is siphoned off from the cool-dry process air exiting the device, providing a large cooling potential.

Latent and sensible cooling loads can be met independently by adjusting four control variables: mixed air flow rate ratio through the device (Ṙ_ma), first stage exhaust flow rate ratio in the first stage (R_e1), outdoor air fraction (OAF), and inlet liquid desiccant concentration (C_LD,in,DEVap). A wide range of outdoor and return air conditions and sensible heat ratio values were simulated with the intent of optimizing source coefficient of performance (COP_source) by changing control variable values. From these simulations, optimal control strategies were developed.

Control strategies are divided into three cases: (1) latent cooling only, (2) sensible cooling only, and (3) both sensible and latent cooling. A simple strategy concerning R_ma and C_LD,in,DEVap was developed for case (1). For case (2), it was determined that return air states had less influence than outdoor air states and that holding OAF and R_e1 constant while linearly increasing R_ma would be satisfactory for most cases. For case (3), optimal control was divided into two phases: a ramping phase where OAF and R_e1 are held constant while R_ma linearly increases and a final phase where R_ma is held at its maximum value and OAF slightly shifts value to provide large cooling capacities.