Theoretical and experimental investigation on the application of solar water heater coupled with air humidifier for regeneration of liquid desiccant

Theoretical and experimental investigation on the application of flat plate solar water heater coupled with air humidifier for regeneration of liquid desiccant has been presented in this work. The heated water from the storage tank of the solar heating system is circulated in a finned tube air heater. Hot air from the air heater is blown through a packing of a honeycomb type for the purpose of regeneration of calcium chloride (CaCl2) solution. An experimental system has been designed and installed for this purpose. The system comprises a solar water heater with a storage tank connected to an air/water heat exchanger. Hot air from the heat exchanger is blown to the air humidifier, which functions in this study as a regenerator. Calcium chloride solution is applied as the working desiccant in this study. Solution concentration is determined at the end of regeneration process and the mass of evaporated water is evaluated. It is observed that the heating temperature varies, at day time, in a range of about 5 degrees C. This limited variation in hot water temperature demonstrates the importance of the storage tank to attain a nearly steady state operation of the system. Experimental results show that solution with 30% concentration can be regenerated up to 50% using solar energy. In the theoretical part of this study, a multiple-layer artificial neural network (ANN) model has been applied to study the performance of a solar liquid-desiccant dehumidification/regeneration system when calcium chloride solution is applied as the working desiccant. The experimental results of the present study are used to construct and test the ANN model. Then the model has been utilized to describe and analyze the effect of the inlet conditions of air on the regeneration process. Good agreement between the outputs from the ANN model and the corresponding results from the experimental data has been found. The proposed model can work well as a predictive tool to complement the experiments. (C) 2011 Elsevier Ltd. All rights reserved.