Optimization of parameters for air dehumidification systems including multilayer fixed-bed binder-free desiccant dehumidifier

In air dehumidification systems, the optimization of operating conditions is as important as the improvement of the dehumidifier itself under varying loads. Therefore, a comprehensive parameter analysis and optimization study of a novel air dehumidification system, which consists of a multilayer fixed-bed binder-free desiccant dehumidifier and a water cooling and heating device, is proposed herein. First, a mathematical model to predict the transient heat and mass transfer in the air dehumidification system is established and validated by comparing with the experimental results. Second, the cyclic performance of the system is analyzed and the influences of each parameter, including operating, structural, and adsorption and transport parameters, are analyzed. Third, the optimization of operating parameters is conducted based on the combination of a backpropagation (BP) neural network and genetic algorithms (GAs). The operating parameters are optimized by maximizing the energy efficiency and dehumidification performance and a trade-off relation is identified between them. For three typical load conditions, it is successfully demonstrated that the optimum energy efficiency and dehumidification performance can be determined from the Pareto front obtained by an elitist non-dominated sorting GA (NSGA Ⅱ) within the requirement specifications and the optimum operating parameters can also be determined by the combination of a BP neural network and a NSGA Ⅱ.