Comparison of Stochastic Methods with Respect to Performance and Reliability of Low-Temperature Gas Separation Processes Nassim Tahouni, 1 * Robin Smith 2 and M. Hassan Panjeshahi 1 1. Department of Chemical Engineering, University of Tehran, P.O Box 11365/4563, Tehran, Iran 2. Centre for Process Integration, School of Chemical Engineering and Analytical Science, University of Manchester, Manchester M60 1QD, UK In this paper, the performances of two popular stochastic methods, the genetic algorithm (GA) and simulated annealing (SA), are verified in the optimization of low-temperature gas separation processes. While the feasibility of GA optimization of low-temperature processes has recently been addressed, our work studied the quality of GA solutions. Having optimized the solutions of three different case studies, it was observed that SA is more robust and reliable than GA when applied to such systems, and by adjusting the key parameters in the SA method, the optimization process can avoid pre-mature convergence and is able to give the best near-global results. Dans cet article, les rendements de deux m´ ethodes stochastiques populaires, l’algorithme g´ en´ etique (AG) et l’algorithme du recuit simul´ e (ARS) sont v´ erifi´ es dans l’optimisation de processus de s´ eparation de gaz ` a basse temp´ erature. Bien que la faisabilit´ e de l’optimisation de l’AG pour les processus ` a basse temp´ erature ait ´ et´ e r´ ecemment trait´ ee, notre travail permettait d’´ etudier la qualit´ e des solutions de l’AG. Ayant optimis´ e les solutions de trois ´ etudes de cas diff´ erentes, il a ´ et´ e observ´ e que l’ARS est plus robuste et plus fiable que l’AG lorsqu’il appliqu´ e` a de tels syst` emes et en ajustant les param` etres cl´ es dans la m´ ethode AG, le processus d’optimisation peut ´ eviter une convergence pr´ ematur´ ee et est en mesure de donner les meilleurs r´ esultats quasi globaux. Keywords: sub-ambient processes, optimization, gas separation processes, genetic algorithm, simulated annealing INTRODUCTION Design and Optimization of Sub-Ambient Processes—Difficulties and Complications I n the chemical process industry, many processes, such as natural gas liquefaction, gas separation and ethylene production, operate partially or totally below ambient temperature. They require heat removal through refrigeration systems that must be integrated with the process streams on the site, with heat exchanger networks, or with external utilities to remove the heat extracted from separation processes to available heat sinks. For large-scale systems, multiple levels of refrigeration, cascaded systems, and mixed refrigerants are used. This, coupled with a high degree of heat integration, makes the design of such systems enormously complicated due to the complex interactions that arise (Smith, 2005). The design of a sub-ambient process involves the design of (1) the core process (mostly distillation), (2) the heat exchanger networks, and (3) the refrigeration system. These three design tasks are interdependent, and modification in any one of the components impacts the other two. It should be noted that the cost of running such processes is usually dominated by the power required for refrigeration. In low-temperature liquefaction and gas separation processes, it is desirable to develop a conceptual methodology for ∗ Author to whom correspondence may be addressed. E-mail address: ntahuni@yahoo.co.uk Can. J. Chem. Eng. 88:256–267, 2010 © 2010 Canadian Society for Chemical Engineering DOI 10.1002/cjce.20265 Published online 8 March 2010 in Wiley InterScience (www.interscience.wiley.com) | 256 | THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING | | VOLUME 88, APRIL 2010 |