Desiccant coated fin tube energy exchanger design optimization implementing KNN-ML tool and adsorption/desorption kinetics analysis using finite difference based transient model

Abstract

In air conditioning applications, to increase energy exchange capabilities and enhance indoor air quality, the desiccant coated fin tube energy exchanger (DCFTE) appears to be a capable substitute compared to traditional energy exchangers like an adsorber bed and rotary wheel. Hence, for enhancing the air conditioning system's performance, accurate prediction of the adsorption kinetics of DCFTE is crucial. Thus, a K-Nearest Neighbor-Machine Learning (KNN-ML) tool is employed for predicting the exit and design parameters of DCFTE. The impact of the operating parameters on the performance of DCFTE has been assessed by utilizing the KNN-ML tool. To analyze the performance of DCFTE, energy exchange and dehumidification capacity are selected as the performance indices. The optimum operating conditions for a specific operational range are also found using the KNN-ML tool. Moreover, a finite difference approach-based transient model is developed to assess the inlet parameters' effect on the performance characteristics and adsorption/desorption kinetics of DCFTE. The coefficient of performance and moisture removal capacity are selected as the performance indices. Furthermore, the effect of time on the outlet air specific humidity, outlet water temperature, air temperature, and desiccant concentration during the dehumidification and regeneration process is analyzed. The effect of the water vapor partial pressure ratio on adsorption capacity is examined. Moreover, Lewis and Stanton number's influence on the DCFTE performance is analyzed by employing the transient model. Silica gel is utilized as a desiccant.