Parametric studies of silica gel and molecular sieve desiccant wheels: Experimental and modeling approaches

In this paper, six desiccant wheels with diameters of 44 cm, 55 cm, and 77 cm using silica gel and molecular sieve desiccants under different conditions were experimentally evaluated. Process outlet temperature, dehumidification effectiveness, moisture removal capacity, sensible energy ratio, and regeneration specific heat input were examined as functions of regeneration temperature, volume flow rate and, rotation speed at different wheel diameters and materials. The results show that there are optimum operating conditions for each wheel, characterized by a specific diameter and desiccant material. To present a robust, reliable, accurate, and fast model based on the experimental data, a hybrid approach-adaptive neuro fuzzy inference system (Hybrid-ANFIS) is developed to precisely calculate the parameters mentioned above. The outcomes of the developed model are further compared with a genetic algorithm-least square support vector machine approach (GA-LSSVM) to identify the capability of the suggested Hybrid-ANFIS model. It was found that the predictability of the suggested Hybrid-ANFIS model outperforms the GA-LSSVM, confirming that the Hybrid-ANFIS technique represents a simple tool that can be easily implemented in dynamic simulation tools for the optimization and control of desiccant cooling systems.