A membrane-based flat-plate liquid desiccant dehumidification air-conditioning system

The enormous increase of energy consumption for building indoor environment, specifically for indoor thermal comfort has brought numerous state-of-art technologies as alternatives to the traditional vapour compression air-conditioning system. The membrane-based liquid desiccant dehumidification has gain wide attention for good feasibility, high energy efficiency and no desiccant carry-over problem. As a matter of fact, the membrane-based liquid desiccant system is still under research and development stage with limited number of literatures. The main aim of this thesis is to develop and evaluate a novel membrane-based liquid desiccant dehumidification air-conditioning system. The research scope mainly covers system design, numerical modelling, simulation programme development, experimental testing and system performance evaluation. The first step in the study is to develop a numerical model for a cross flow flat-plate type membrane-based dehumidifier and validate model by experimental data. Following this, a single regenerator with the similar structure as the dehumidifier is investigated numerically and experimentally. With models for single dehumidifier and regenerator, a numerical model for a more realistic complete membrane-based dehumidification system, which consists of two membrane-based heat and mass contractors, three heat exchangers, cooling water and hot water supply units is developed and validated experimentally. Finally, influences of using mixed LiCl-CaCl2 solution with different mixing ratios on the system performance are analysed. The results show that the number of heat transfer unit NTU and mass flow rate ratio m^* have the most significant effects on the performance of the dehumidifier and regenerator, and their effects are interacted with each other. The increasing gradients of dehumidifier and regenerator effectiveness hardly change when NTU and m^* exceed their critical values 〖NTU〗_crit and m_crit^*, which are 4 and 1 for the dehumidifier, and 4 and 2 for the regenerator. The dehumidifier sensible effectiveness is insensitive to both solution inlet temperature T_sol and concentration C_sol, while the latent effectiveness increases significantly with C_sol at a high solution temperature. The dehumidifier shows broad adaptability in different weather conditions by providing relative stable state supply air, in particular at high NTU, while neither T_(air,in) nor W_(air,in) has remarkable influence on the regenerator performance, though the solution re-concentration ability can be enhanced slightly by applying drier and warmer air. Regenerator can benefit from increased solution temperature as enhanced re-concentration (0.8% increase of moisture flux rate MFR) and cooling effects (13.7% increase of temperature decrease rate TDR). However this means more cooling energy is required for the high temperature desiccant solution. For the complete system, 〖NTU〗_de and m^* have the most considerable impact on the overall system effectiveness. The critical values of m^* vary under different NTUs, and it is preferable to keep m^* at or below the critical values as further increasing solution flow rate would reduce the system coefficient of performance COP. The latent cooling output Q_lat is normally two times higher than the sensible cooling output Q_sen under all circumstances meaning further cooling, such as indirect evaporative cooling is required after the dehumidification. Compared with the pure LiCl solution system, the mixed solution system COP can be raised up to 30.23% by increasing CaCl2 content. The optimum mixing ratio varies with the solution concentration. For the LiCl-CaCl2 solution, the highest COPs appear at the mixing ratios of 3:1, 2:1 and 1:1 for 20%, 30% and 40% concentrations respectively