The liquid desiccant dehumidification system is becoming a hotspot in air dehumidification due to its promising prospects. Many models have been reported in the literature, but they are rarely applicable to control design due to various reasons. In this paper, a simplified dynamic model of liquid desiccant dehumidifier is proposed from a control point of view based on the heat and mass transfer principles, where the spatial differentials of fluid properties are approximated by discretization and dynamic interaction along the flow direction. Unknown model parameters are first estimated from static experimental data through the Levenberg-Marquardt algorithm and then refined from dynamic experimental data through the extended Kalman filter. The proposed model performs well in the experimental validation and is expected to be applied in a future control design and fault diagnosis application. (C) 2016 Elsevier Ltd. All rights reserved.