Optimal design of adsorption chillers based on a validated dynamic object-oriented model

Abstract

The design of adsorption chillers is usually based on experience and high experimental effort. Experimental effort can be reduced by using dynamic models. In the present study, a dynamic model is validated with a modular adsorption chiller test bed and then used to optimize design and process parameters to gain maximum cooling power. The modularity of the test bed enables the exchange of single components without changing the remaining setup. This modular structure is also reflected in the object-oriented dynamic model. Model calibration is based on the heat flows of all components. This measure allows the gain of deep insight into the system behavior and a quantitative comparison of model accuracy. The calibrated model is validated by predicting the system behavior for different operating conditions and also changed adsorbent materials. Adsorbent materials silica gel 123 and zeolite 13X are investigated. Operating points vary in cycle time, as well as temperatures of evaporation, adsorption, and desorption. The model exhibits excellent prediction capability for the coefficient of performance and for the cooling power. The modular setup of the model is then used for targeted optimization of the adsorption system; the cycle time and the sizing of the heat exchangers are rigorously optimized, leading to adsorption chillers with maximum cooling power.