Performance evaluation of packed tower liquid desiccant dehumidifier based on LSSVM

In this paper, an empirical model based on least square support vector machine (LSSVM) method to predict the output air conditions in a packed tower liquid desiccant dehumidifier is developed. By analysis of the coupled heat and mass transfer between the process air and desiccant solution, six variables are used as the inputs of the LSSVM model, namely: desiccant solution and air flow rates, desiccant solution and air inlet temperature, desiccant concentration, and air relative humidity. Meanwhile, outlet air temperature and relative humidity related with the performance of the dehumidifier are considered as the outputs of the LSSVM model. Compared with the existing theoretical models, the present one is very simple, yet accuracy, and does not need complex theoretical analysis. The experimental results illustrate the effectiveness of the proposed model on performance predicting in a packed tower liquid desiccant dehumidifier. This developed model is expected to have widely applications in performance evaluation, operational monitoring, fault detection and diagnosis.