Optimization of operation parameters of refrigeration cycle using particle swarm and NLP techniques

In this paper, the two mixed refrigerant refrigeration cycles were proposed to be replaced by pure ethylene cycle in the olefin plant of the Tabriz petrochemical complex. Both these components composition of refrigerant and the compressor operations pressures are the key design parameters in the mixed refrigerant refrigeration systems. The purpose of the paper is to present a systematic method based on a combination of mathematical methods and thermodynamic viewpoint to optimize mixed refrigerant cycles parameters. Particle swarm optimization and non-linear programming techniques were employed to optimize the parameters of cycles. Results show that the particle swarm optimization is superior to the NLP optimization techniques in finding the values of optimizing variables.