| 학회 |
한국화학공학회 |
| 학술대회 |
2022년 봄 (04/20 ~ 04/23, 제주국제컨벤션센터) |
| 권호 |
28권 1호, p.126 |
| 발표분야 |
[주제 2] 기계학습 |
| 제목 |
Inverse Modeling of Rheological Parameter of Dense Suspension Using Artificial Neural Networks |
| 초록 |
For suspension flow modeling, it is critical to determine rheological parameters using measurements. While a rheometer can be used to determine the rheological parameters, the rheology may change during subsequent flow after the measurement. This study proposes work an inverse model for predicting rheological parameters from pressure drops measured in a dense suspension pipe flow system. The flow data were generated using a power-law fluid and the Weissenberg-Rabinowitsch-Mooney-Schofield (WRMS) method, which establishes a relationship between pressure drop and flow rate for generalized Newtonian fluids. The flow data was used to train a direct neural network model. The pressure drop is calculated directly from the rheological parameters using this direct model. By minimizing the difference between the measured pressure drop and the direct model output, the inverse model determines rheological parameters. As a result of optimization, the rheological parameter, n, was successfully determined. |
| 저자 |
오혜정, 신준섭, 최석훈, 이신제, 박나연, 남재욱, 이종민
|
| 소속 |
서울대 |
| 키워드 |
공정시스템(Process Systems Engineering) |
| E-Mail |
|
| 원문파일 |
초록 보기 |
