Simulation And Parameter Estimation Of Air Separation Lower Column

The air separation unit(ASU)is a complex process that provides high-purity products of oxygen, nitrogen, and argon, which are widely used in metallurgical, chemical, petrochemical, aerospace and other fields. The main distillation column is the most critical equipment in the air separation process, whose fundamental part is the lower column. So it’s particularly important to study on the lower column. Work in this thesis is based on mechanism analysis of the air separation lower tower and developing of HYSYS-Matlab data communication interface. In this thesis, a steady process model of air separation lower column is built respectively using design data and process data, and updated by parameter estimation on stage efficiency. Among them, the simulation of trace components in streams of high purity and the lack of measurement points in air separation plant are two difficult issues to solve in this thesis.The main research contents and contributions are as follows:1. Process mechanism model, including thermodynamic model and distillation model of the air separation lower tower is researched, which provides a theoretical basis for the tower simulation; features and interface of HYSYS simulation software are researched, and HYSYS-Matlab data communication interface is developed for real-time data communication between HYSYS and Matlab, which provides a technical basis for the tower simulation and parameter estimation. At last, software architecture of simulation and parameter estimation is given.2. A steady process model of air separation lower column is built using design data and updated by parameter estimation on stage efficiency. In the process of parameter estimation, several error evaluation methods are researched and compared. A new error evaluation method based on effective number of bits is given. The property of air separation lower column is researched based on its model.3. A steady process model of air separation lower column is built using process data and updated by parameter estimation on stage efficiency. The problem of lacking of measurement points in air separation plant is solved by process mechanism analysis and sensitivity analysis. The result of model building and parameter estimation is satisfactory with this solution. Model is built and updated at different loads based on process data cluster analysis. The relationship between result of parameter estimation on stage efficiency and different loads is researched. The relationship between accuracy of parameter estimation and selection methods of different measurement points is researched and the best selection methods of measurement points is obtained.