Amin Tamuzi - Computer Aided Process Engineering (CAPE) Laboratory, School of Chemical, Petroleum and Gas Engineering, Iran University of Science and Technology, Narmak, Tehran, Iran
Norollah Kasiri - Computer Aided Process Engineering (CAPE) Laboratory, School of Chemical, Petroleum and Gas Engineering, Iran University of Science and Technology, Narmak, Tehran, Iran
Amirhossein Khalili-Garakani - Chemistry and Process Engineering Department, Niroo Research Institute, Tehran, Iran

Implementation of innovative distillation systems in multicomponent distillation design is a complex task because of multitude design variables. Operating pressure is one of the most prominent and effective variables in the distillation columns, which affects capital and operating costs directly. Many heuristic and optimization based methods are presented to find optimal operating conditions of distillation columns. Since the natural gas liquids, NGL, fractionation process is a costly and an energy demand intensive process, the design and operation of these units may affect many important petrochemicals supply chain and whole natural gas processing plant. Herein a comparison has been made between an easy to use heuristic design method and a stochastic based optimization method with genetic algorithm to design the simple and complex multicomponent distillation columns sequences for NGL fractionation processes. The results demonstrate the heuristic method is faster but in complex distillation systems, is inaccurate. In the studied case of the NGL fractionation process, the calculated column pressure by a heuristic method showed up to ۴۰% different in comparisons against stochastic optimization results. This error leads to a ۳% increase of the total annual costs in the heuristic method, which may have a significant impact on the final design and change the evaluation distillation scenarios because of cumulative error effects.

NGL fractionation, Operating Pressure, Multicomponent distillation, Process optimization, Heuristic