RADFRAC Distillation Column Feasible Region Identification and Application in Realistic Operation under Uncertainty

• Main Authors: Yu-Hung Chang, 張育紘
• Other Authors: Vincentius Surya Kurnia Adi
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/jk2nhc

碩士 === 國立中興大學 === 化學工程學系所 === 106 === Traditionally most of the process design is based on economic efficiency as an assessment. Unfortunately, uncertainty is inevitable in a realistic environment. The operating parameters may vary in extreme conditions that the system may not be able to operate normally. Therefore, the operation feasibility of a process design should be ensured under uncertainty while simultaneously consider the capital cost. The RADFRAC distillation column module in Aspen Plus® is rigorous but lack of feasibility assessment feature. To address this problem, a feasible region identification algorithm based on MATLAB-Aspen Plus interface is developed. An algorithm with combination of random line search and bisection method is developed to identify the boundaries of the feasible region in a RADFRAC distillation column. The feasible area can be obtained by Delaunay triangulation strategy and the volume flexibility index is calculated accordingly. In the case studies, the ethanol and water binary system is utilized to show the effectiveness of the proposed method. The changing effects on feasible region from various design variables (i.e. feed stage location, total number of stages, and design specifications) are also discussed. The trade off on economic evaluation and flexibility analysis is also addressed for choosing the ideal design condition. Case studies on different homogeneous binary systems will also be discussed to show the applicability of the proposed method. The analyses on RADFRAC feasible region will provide users better understanding on adding uncertainty in the initial stage of equipment design.