Optimization of Vapor Compression-Absorption Cascade Refrigeration System

MUSSATI, SERGIO F.; MUSSATI, MIGUEL C.; MOROSUK, TATIANA

Berlin

Conferencia; ISHPC2021 - Int. Sorption Heat Pump Conf. 2021; 2021

Technische Universität Berlin (TU Berlin))

This paper focuses on the optimization of a vapor compression-absorption cascade refrigeration system (VCACRS). The absorption cycle consists of a series flow double-effect system using saturated steam as the hot source and H2O-LiBr as the refrigerant-absorbent pair. The compression cycle is a conventional one but considering the possibility to use one of two candidate working fluids: R717 or R134a. A binary variable is used to mathematically model the selection of the working fluid in the compression cycle, resulting in a mixed-integer nonlinear mathematical programming (MINLP) model, which is implemented in the GAMS environment and solved using an optimization algorithm based on the generalized reduced gradient (GRG) method. The minimization of the total annual cost (TAC) - which accounts for capital investment and operating expenditures - is proposed as the objective function. The refrigeration temperature and capacity are the only parameters of the model (fixed values) and the mass flow rate, pressure, temperature and composition of all process streams are simultaneously optimized (decision variables). Also, the model is implemented in such a way that allows to systematically remove the high-pressure condenser and the two solution heat exchangers ? these three components can be optionally selected or removed as appropriate. As a result, an optimal structure, optimal component sizes, and optimal operating conditions of the cascade system are obtained. A novel system reported in the literature involving a different VCACRS configuration operated with H2O-LiBr in the absorption cycle and R717 in the compression cycle is used as a reference case for comparison purposes in order to elucidate which of the two systems is preferred when the total investment cost is considered. The comparison of the results shows that the optimal solution obtained from the superstructure-based model is much less costly than the reference case and represents the main implication of the obtained optimized system. In fact, it was found that TAC is by around 37% lower with respect to the reference case. Also, the high-pressure condenser is removed from the optimal solution. Whereas R134a is selected as optimal refrigerant in the compression cycle. If R134a is replaced by R717, the TAC is still by around 2% lower with respect to the reference case.