| 학회 | 한국화학공학회 |
| 학술대회 | 2002년 봄 (04/26 ~ 04/27, 강원대학교) |
| 권호 | 8권 1호, p.361 |
| 발표분야 | 공정시스템 |
| 제목 | 인공신경망 및 부분최소자승법을 이용한 공기압축시스템의 모델링 |
| 초록 | Most of chemical plants are equipped with compression systems of various types and capacities for pressurizing air or gases. In this study, the performance prediction model is developed using Artificial Neural Network (ANN) algorithm and Partial Least Squares (PLS) method for accurately predicting the overall efficiency and the actual power consumption/generation of a multistage compression system that consists of the multistage compressor, electric motor/generator, and the multistage expander/turbine as a driver. Both PLS and ANN showed good modeling performances, but the ANN was slightly better than the PLS since the ANN is able to capture the nonlinear relationships between operating variables. The developed performance prediction model helps the operators to move the operating condition to the optimal one and can be used as a core model of an optimization framework for compression systems. |
| 저자 | 한인수, 한종훈, 노의철, 이우창, 이경훈 |
| 소속 | 포항공과대 |
| 키워드 | Modeling; PLS; Neural Network; Compressor; Expander; Efficiency |
| , , , , | |
| VOD | VOD 보기 |
| 원문파일 | 초록 보기 |
