Optimal synthesis of mixed-refrigerant systems for low-temperature processes

Using mixtures as refrigerants in the design of refrigeration systems offers significant opportunities in the search for more energy-efficient and compact designs. However, the design of mixed-refrigerant systems is extremely difficult, and few successful design methods are available. As a result, many existing operations can be far from the optimal conditions. In this paper, a novel method for the selection of refrigerant compositions has been proposed. Using this approach, a systematic synthesis method for the complete design of mixed-refrigerant systems has been developed. This approach combines the power of thermodynamics and nonlinear programming (NLP) techniques. Whereas NLP can determine the optimal choice of process operating conditions, thermodynamics simultaneously provides insights into and confidence in the solution. More importantly, the use of thermodynamic analysis in this approach helps overcome the complexity in NLP optimization for determining key design variables, including refrigerant composition, operating pressures, and flow rates. Case studies demonstrate that the method is effective and that mixed-refrigerant systems can significantly save energy and improve process efficiency.