Performance Design of a Cryogenic Air Separation Unit for Variable Working Conditions Using the Lumped Parameter Model
Jinghua Xu,Tiantian Wang,Qianyong Chen,Shuyou Zhang,Jianrong Tan
DOI: https://doi.org/10.1007/s11465-019-0558-6
IF: 4.063
2019-01-01
Frontiers of Mechanical Engineering
Abstract:Large-scale cryogenic air separation units (ASUs), which are widely used in global petrochemical and semiconductor industries, are being developed with high operating elasticity under variable working conditions. Different from discrete processes in traditional machinery manufacturing, the ASU process is continuous and involves the compression, adsorption, cooling, condensation, liquefaction, evaporation, and distillation of multiple streams. This feature indicates that thousands of technical parameters in adsorption, heat transfer, and distillation processes are correlated and merged into a large-scale complex system. A lumped parameter model (LPM) of ASU is proposed by lumping the main factors together and simplifying the secondary ones to achieve accurate and fast performance design. On the basis of material and energy conservation laws, the piecewise-lumped parameters are extracted under variable working conditions by using LPM. Takagi-Sugeno (T-S) fuzzy interval detection is recursively utilized to determine whether the critical point is detected or not by using different thresholds. Compared with the traditional method, LPM is particularly suitable for “rough first then precise” modeling by expanding the feasible domain using fuzzy intervals. With LPM, the performance of the air compressor, molecular sieve adsorber, turbo expander, main plate-fin heat exchangers, and packing column of a 100000 Nm3 O2/h large-scale ASU is enhanced to adapt to variable working conditions. The designed value of net power consumption per unit of oxygen production (kW/(Nm3 O2)) is reduced by 6.45%.