| 학회 |
한국화학공학회 |
| 학술대회 |
2013년 가을 (10/23 ~ 10/25, 대구 EXCO) |
| 권호 |
19권 2호, p.1349 |
| 발표분야 |
공정시스템 |
| 제목 |
석탄화력발전소 NOx 모델의 온라인 업데이트 : LSSVM 및 output bias 업데이트 선정기준 |
| 초록 |
In this paper, a new model update scheme is applied to Least Squares Support Vector Machines to build its online version. For long term predictions, this online model is equipped with recursive LSSVM update along with output bias update. Two update schemes, scheme-I and scheme-II, are applied to NOx emission process data from Taean Power Plant, South Korea. These schemes make use of NOx emission’s behavior to select the proper update method from LSSVM update and output bias update at the instance of time when new measurement arrives to update or calibrate the model. Prediction result obtained from proposed online LSSVM model is compared with online PLS model developed by authors (Ahmed et. al.; 2009) . Online LSSVM predicts NOx outstandingly and also outperforms online PLS model. |
| 저자 |
파이잘 아메드, 조현준, 김태형, 성노욱, 여영구
|
| 소속 |
한양대 |
| 키워드 |
|
| E-Mail |
|
| 원문파일 |
초록 보기 |
