The Applications of Low-Order Model of Nonlinear Distillation

• Main Authors: Ming-Bang Liu, 劉明邦
• Other Authors: Shuh-Woei Yu
• Format: Others
• Language: zh-TW
• Published: 1993
• Online Access: http://ndltd.ncl.edu.tw/handle/84973693556130426587

碩士 === 國立中央大學 === 化學工程研究所 === 81 === Dual composition control of distillation column has been one of the most pursued research topics in the field of process control. In addition to the control diffculties caused by loop interaction, the large numbers of variables, process and measurement delays, the nonlinear characteristics of many high- purity distillations has to be considered as well. However, before nonlinear control algorithms can be readily implemented, linear control systems will continue to enjoy their popularity in nonlinear process applications basically because of their proven reliability and the availability of rigorous stability and performance evaluation techniques. To overcome the limitation of control performance of linear feedback controllers due to nonlinear dynamic behavior of the process, special strategies baesd on the conventional PI structure such as nonlinear PI and variable transformation have been proposed. The primary objective of this study is to compare control performance of these nonconventional strategies with diagonal PI and full PI using a low-order nonlinear model derived from a set of linear transfer function models of a simulated high- purity distillation column. Test results reveal the fact that a well designed diagonal PI is compable with nonlinear PI and variable transformation approach while full PI provides the best control.