Optimal Design for Cryogenic Structured Packing Column Using Particle Swarm Optimization Algorithm

Bin Wang,Shanshan Shi,Shunhao Wang,Limin Qiu,Xiaobin Zhang
DOI: https://doi.org/10.1016/j.cryogenics.2019.102976
IF: 2.134
2019-01-01
Cryogenics
Abstract:Large scale cryogenic air separation is the most efficient and cost-effective approach to produce high-purity air products by present. Structured packing columns (SPC) are widely focused and applied due to their characteristics of high efficiency and energy saving in the cryogenic distillation process. The optimal design of the SPC is to reduce the energy consumption and initial investment, while it is a highly nonlinear and multivariable problem. The coexistence of real variables and integer variables, such as the flow rates and the positions of materials at the inlets/outlets, makes the optimization become a typical mixed integer nonlinear programming (MINLP) problem. The purpose of this paper is to study the optimal design method for the cryogenic SPC using the particle swarm optimization (PSO) algorithm. A modified PSO for handling the MINLP problem (MI-PSO) is proposed. A multi-objective optimal design for the SPC in cryogenic air separation plant with the capacity of 17,000 Nm(3)/h is investigated as a instance. By MI-PSO algorithm, the total exergy loss theoretically reduces 36.3% and the main condenser heat load decreases 5.4% after optimization, which can provide test prediction for the cryogenic distillation experiment.
What problem does this paper attempt to address?