Modeling of liquid desiccant cooling and dehumidification system based on artificial neural network

Abstract

Liquid desiccant dehumidification system (LDDS) has emerged as an energy-efficient approach for air dehumidification. In this paper, a simple model for the liquid desiccant cooling and dehumidification air conditioning (LDCDAC) system is proposed. The model is built by using artificial neural network (ANN) to describe the cooling, dehumidification and regeneration performance of the LDCDAC system. The system outlet parameters, such as chilled water temperature, air temperature and humidity, can be calculated directly from the inlet parameters. A multilayer neural network is adopted, and the ANN model is trained by the experimental data collected under different operating conditions. The model predictions of the heat and mass transfer rates are compared with the experimental values. The results indicate that the model predicting errors are within ±8%. The proposed model can be used in control and optimization applications of the LDCDAC system.