| 基于在线Kriging模型的隔板塔优化方法 |
| |
| 引用本文： | 赵克凡,贾胜坤,罗祎青,袁希钢. 基于在线Kriging模型的隔板塔优化方法[J]. 化工学报, 2022, 73(1): 332-341. DOI: 10.11949/0438-1157.20210961 |
| |
| 作者姓名： | 赵克凡 贾胜坤 罗祎青 袁希钢 |
| |
| 作者单位： | 天津大学化工学院,天津300350;天津大学化学工程研究所,天津300350;天津大学化工学院,天津300350;天津大学化学工程研究所,天津300350;化学工程联合国家重点实验室,天津300350 |
| |
| 基金项目： | 国家自然科学基金项目(91834303,21978203)； |
| |
| 摘 要： | 隔板塔是一种新型的精馏节能技术，但其优化设计属于混合整数非线性规划问题，优化求解困难。本文提出在线Kriging代理模型优化方法，将其应用于隔板塔年度总费用优化。与传统离线代理模型相比，本文方法采用少量的隔板塔严格模拟数据点，即可得到优化问题的最优解。计算表明，较传统离线代理模型，本文提出的在线代理模型显示了更高的准确性和有效性，可以克服传统离线代理模型优化耗时、计算量大的缺点。
|
| 关 键 词： | 代理模型 Kriging模型 隔板塔 计算机模拟 优化 |
| 收稿时间： | 2021-07-11 |
Optimal design for dividing wall column using online Kriging surrogate model-based optimization method |
| |
| ZHAO Kefan,JIA Shengkun,LUO Yiqing,YUAN Xigang. Optimal design for dividing wall column using online Kriging surrogate model-based optimization method[J]. Journal of Chemical Industry and Engineering(China), 2022, 73(1): 332-341. DOI: 10.11949/0438-1157.20210961 |
| |
| Authors: | ZHAO Kefan JIA Shengkun LUO Yiqing YUAN Xigang |
| |
| Affiliation: | 1.School of Chemical Engineering and Technology, Tianjin University, Tianjin 300350, China;2.Institute of Chemical Engineering, Tianjin University, Tianjin 300350, China;3.State Key Laboratory of Chemical Engineering, Tianjin University, Tianjin 300350, China |
| |
| Abstract: | Divided wall column (DWC) is a new energy-saving distillation technology. However, its optimal design is a mixed integer nonlinear programming (MINLP) problem, and it is difficult to optimize the solution. In this paper, an online Kriging surrogate model-based optimization method is proposed and applied to the total annual cost optimization of DWC. Through the proposed method, the optimal solution of the optimization problem can be obtained by a small number of rigorous simulation data points. The solution of an illustrating example shows that, compared with the traditional offline surrogate model-based optimization method, the proposed method is more advantageous in terms of accuracy and effectiveness, being able to overcome the shortcomings of the traditional offline surrogate based-model optimization, such as time-consuming and large amount of calculation. |
| |
| Keywords: | surrogate model Kriging model divided wall column computer simulation optimization |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《化工学报》浏览原始摘要信息 |
|
点击此处可从《化工学报》下载免费的PDF全文 |
|
