| 학회 |
한국화학공학회 |
| 학술대회 |
2018년 가을 (10/24 ~ 10/26, 대구 EXCO) |
| 권호 |
24권 2호, p.1229 |
| 발표분야 |
공정시스템(Process Systems Engineering) |
| 제목 |
Bayesian parameter estimation for the model development of the water lean amine solvent CO2 capture process |
| 초록 |
The accurate evaluation of KHU-B, which is the water lean amine solvent and developed by our research team, is impossible without the development of the process model. Inaccuracy in experimental results and the incomplete knowledge of model parameters are problems which have a crucial effect on the reliability of process models. Here, we solve their uncertainty problems and develop the CO2 capture process model for KHU-B. To detect non-influential parameters and their interactions (parameter subset selection), we carry out the global sensitivity analysis using Sobol' indices in CO2 capture model of ASPEN plus. After that, we make the quadratic hyper-surface surrogate model which reflects the interaction between selected parameters. The calibration of the proposed model using experimental data is conducted to estimate the parameters from Bayesian inference using Markov Chain Monte Carlo (MCMC) with Metropolis-Hastings algorithm. |
| 저자 |
김정남1, 나종걸1, 김훈식2, 이현주1, 이 웅1
|
| 소속 |
1한국과학기술(연), 2경희대 |
| 키워드 |
공정모델링 |
| E-Mail |
|
| VOD |
VOD 보기 |
| 원문파일 |
초록 보기 |
