Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/87045
Title: Control and optimization of dedicated outdoor air-chilled ceiling systems using liquid desiccant dehumidification
Authors: Ge, Gaoming
Degree: Ph.D.
Issue Date: 2011
Abstract: Dedicated outdoor air-chilled ceiling (DOAS-CC) system, as an alternative air conditioning manner, has attracted world-wide interests in recent years. It can achieve independent control of indoor temperature and humidity, provide more effective ventilation, prevent virus and bacteria transmission among different zones, etc; therefore it can improve indoor thermal comfort and indoor air quality. Considering the huge energy consumption for treating outdoor air in hot and humid regions like Hong Kong, total heat recovery and independent dehumidification technologies are desirable for energy saving in air conditioning. A DOAS-CC system adopting a novel membrane-based total heat exchanger and liquid desiccant dehumidification is proposed and studied in the research. Robust control is prerequisite for reliable operation of air-conditioning systems besides proper designs and maintenance. It also has significant impacts on energy use as well as occupants' comfort, healthy and productivity. However, control issues in the DOAS-CC system and the liquid desiccant dehumidification process have been rarely concenned. Therefore, the aim of this research is to develop robust local control methods and optimal control strategies for the DOAS-CC integrated system to achieve the desired indoor thermal comfort and indoor air quality with minimum energy consumption. The aim is reached through addressing the following scientific issues: (1) performance study of the membrane-based total heat exchanger; (2) control characteristics of the liquid desiccant dehumidification system; (3) system approach based global optimization of air-conditioning systems considering multiple objectives and multiple variables. In addition, control strategy for preventing condensation on the surface of chilled ceiling is also developed. The main works of this research are shown as follows. Firstly, a comprehensive mathematical simulator for the whole DOAS-CC system has been developed on the TRNSYS platfrom. Mathematical models of major components including the membrane-based total heat exchanger, the liquid desiccant dehumifier and regenerator are developed based on the simplified heat and mass transfer analysis method and the principle of mass and energy conservation. Main components of the system are validated by experimental data. Performance of the system can be studied by the simulator.
Secondly, control strategies for the liquid desiccant dehumidification system are developed to achieve independent control of the supply air temperature and humidity. Control methods for the air dehumidification and cooling process in the dehumidifier side as well as desiccant solution regeneration process in the regenerator side are proposed and evaluated. Control characteristics of the liquid desiccant dehumidification system are also investigated. Thirdly, control methods for the multi-zone dedicated outdoor air system are developed to decouple the indoor air temperature, humidity and ventilation controls. The performance of optimization control methods, such as supply air humidity ratio reset and demand controlled ventilation strategy, are evaluated by simulation tests in the system. Fourthly, a model-based global optimal control strategy is developed to reduce the energy consumption of the DOAS-CC system and provide comfort indoor environment. The indoor air temperature, indoor air humidity and total energy consumption are considered simultaneously. The supply air temperature and humidity of the DOAS and the supply water temperature in the system are optimized using the Genetic Algorithm (GA) based on the trade-off among the indoor air temperature, indoor air humidity and total system energy consumption in the format of a system response-based cost function. Finally, a predictive condensation control strategy is developed for the DOAS-CC system to prevent condensation occurring at the start-up moment of chilled ceiling system. Neural network (NN) based models are developed to predict the temperature on the surface of chilled ceiling and the indoor air dew-point temperature at the start-up period, hence to predict whether condensation will occur or not. Another model is developed to predict the optimal prior operation time for DOAS to prevent condensation. The performance of this predictive condensation strategy is evaluated.
Subjects: Air conditioning -- Control -- Mathematical models.
Humidity -- Control -- Mathematical models.
Drying agents.
Hong Kong Polytechnic University -- Dissertations
Pages: xxi, 184 leaves : ill. ; 30 cm.
Appears in Collections:Thesis

Show full item record

Page views

38
Last Week
0
Last month
Citations as of Mar 24, 2024

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.