Performance analysis of liquid desic cant dehumidification system using artificial neural networks

The heart of the liquid desiccant cooling system is the dehumidification process which is influenced by many parameters. Different types of dehumidification equipment have been developed and a variety of analytical models have been employed to analyze the dehumidification process. The dehumidification process involves simultaneous heat and mass transfer and reliable transfer coefficients are required in order to analyze the system. This has been proved to be difficult and many assumptions are made to simplify the analysis. Artificial Neural Network (ANN) is widely used as an innovative way to tackle complex and ill-defined problems. The present research proposes the use of ANN based model in order to simulate the relationship between inlet parameters and the performance of the dehumidifier. For the analysis, randomly packed dehumidifier is chosen since the packed tower facilitates high mass transfer by providing a large surface area in a relatively small volume. Lithium chloride is selected as the liquid desiccant due to its stability with high performance. A multilayer ANN is used to investigate the performance of dehumidifier. For training ANN models, data is obtained from analytical equations. The training process implies adjustment of connection weights and biases so that the differences between ANN output and the desired output are minimized. Eight parameters are used as inputs to the ANN, namely: air and desiccant flow rates, air and desiccant inlet temperatures, air inlet humidity, the desiccant inlet concentration, dimensionless temperature ratio, and the inlet temperature of the cooling water. The output of the ANN is the water condensation rate. The predicted water condensation rate by the ANN is validated with experimental data and the value of R² is found to be 0.9251. Results and the performance of the developed system are presented in this chapter.