PLS를 이용한 증류 공정의 국부적인 조성 추정 소프트 센서

홍선주; 허창구; 한종훈

HWAHAK KONGHAK, Vol.37, No.3, 445-452, June, 1999

Export Citation

PLS를 이용한 증류 공정의 국부적인 조성 추정 소프트 센서

Local Composition Soft Sensor in a Distillation Column using PLS

홍선주, 허창구, 한종훈

초록

본 연구는 증류 공정에서 원료 조성의 변화와 같은 외란의 영향, 정상 조업 영역이 넓은 경우, 또는 다양한 품질 규격의 추정(estimation) 등으로 인하여 조업 영역이 여러 개로 구분될 때, 각각의 조업 영역에 맞는 국부적인 조성 추정 소프트센서의 설계에 관한 것이다. 조업 영역의 분류는 linkage based clustering방법을 이용하여 구분하였으며. 국부 모델로써 partial least squares(PLS)를 이용하였다. 본 연구에서 제안하는 국부 PLS 소프트센서는 원료 조성의 변화를 갖는 모사된(simulated) 이성분 증류 공정의 에탄올 조성 추정과 정상 조업 영역이 넓은 산업체의 splitter증류 공정의 톨루엔(toluene) 조성 추정에 적용하였다. 국부 PLS 소프트센서는 global PLS, 신경회로망을 기반으로 하는 비선형PLS. 신경회로망을 이용한 소프트센서와 비교하였을 때, 두 사례 연구에서 모두 뛰어난 조성 추정 성능을 보여주었다.

This paper discusses local composition soft sensors in a distillation column with regards to different operation modes, which are caused by effect of load disturbance such as feed composition change, wide operation regions, estimation of various product specification, etc. Linkage_based clustering method and PLS are used for the classification of various operation regions, modeling of local soft sensor respectively. The proposed method was illustrated using estimation of ethanol in a binary distillation column simulated in simulator, HYSYS^TM and toluene in an industrial splitter column within a wide operation window. In simulation, process data for sub-operation regions have been collected to reflect the changes of feed composition that is of frequent occurrence in real distillation column. Local PLS soft sensors were compared to global PLS, nonlinear PLS based on neural network and neural network. and shown better performance than other methods.

Keywords:Distillation Column;Different Operation Modes;Local Soft Sensor;Linkage Based Clustering;PLS

[References]

Tham MT, Montague GA, Morris AJ, Lant PA, J. Process Control, 1, 3 (1991)
Shin J, Lee M, Park S, Korean J. Chem. Eng., 15(6), 667 (1998)
Znar A, "Nonlinear Time Series Models for Multivariate Dynamic Processes," in web site of Ali Znar (1996)
Kresta JV, Marlin TE, Macgregor JF, Comput. Chem. Eng., 18(7), 597 (1994)
Fujii H, Lakshminarayanan S, Shah SL, "Application of the PLS Technique to the Estimation of Distillation Tower Top Composition," IFAC ADCHEM, 529 (1997)
Mejedell T, Skogestad S, "Output Estimation Using Multiple Secondary Measurements: High-Purity Distillation,", 39, 10 (1993)
Geladi P, Kowalski BR, Anal. Chim. Acta, 185(1) (1986)
Kresta JV, Marlin TE, Macgregor JF, Comput. Chem. Eng., 18(7), 597 (1994)
Martens H, Naes T, "Multivariate Calibration," John Wiley & Sons (1989)
Qin SJ, McAvoy TJ, Comput. Chem. Eng., 16(4), 379 (1992)
Willett P, "Similarity and Clustering in Chemical Information Systems," John Wiley & Sons (1987)
Duda RO, Hart PE, "Pattern Classification and Scene Analysis," John Wiley & Sons (1973)

[Referenced By]

[1] Tham MT, Montague GA, Morris AJ, Lant PA, J. Process Control, 1, 3 (1991)

[2] Shin J, Lee M, Park S, Korean J. Chem. Eng., 15(6), 667 (1998)

[3] Znar A, "Nonlinear Time Series Models for Multivariate Dynamic Processes," in web site of Ali Znar (1996)

[4] Kresta JV, Marlin TE, Macgregor JF, Comput. Chem. Eng., 18(7), 597 (1994)

[5] Fujii H, Lakshminarayanan S, Shah SL, "Application of the PLS Technique to the Estimation of Distillation Tower Top Composition," IFAC ADCHEM, 529 (1997)

[6] Mejedell T, Skogestad S, "Output Estimation Using Multiple Secondary Measurements: High-Purity Distillation,", 39, 10 (1993)

[7] Geladi P, Kowalski BR, Anal. Chim. Acta, 185(1) (1986)

[8] Kresta JV, Marlin TE, Macgregor JF, Comput. Chem. Eng., 18(7), 597 (1994)

[9] Martens H, Naes T, "Multivariate Calibration," John Wiley & Sons (1989)

[10] Qin SJ, McAvoy TJ, Comput. Chem. Eng., 16(4), 379 (1992)

[11] Willett P, "Similarity and Clustering in Chemical Information Systems," John Wiley & Sons (1987)

[12] Duda RO, Hart PE, "Pattern Classification and Scene Analysis," John Wiley & Sons (1973)