| 학회 | 한국화학공학회 |
| 학술대회 | 2021년 가을 (10/27 ~ 10/29, 광주 김대중컨벤션센터) |
| 권호 | 27권 2호, p.1469 |
| 발표분야 | 공정시스템 |
| 제목 | Deep neural network-based cost optimization of wet flue gas desulfurization system using waste sea shells to solve the high-grade limestone depletion |
| 초록 | In this study, we proposed the cost optimal selection and blending ratio of waste sea shells as high-grade limestone substitutes through cost optimization using deep neural network (DNN)-based surrogate model. Derivation of the cost optimal blending ratio is proceeded to the following procedure. First the process model to generate dataset is developed and the dataset is gypsum purity according to blending ratio. To calculate the total annualized cost (TAC), the mathematical model is proposed and the TAC is added to the dataset. Second, the generated dataset is preprocessed based on the z-score normalization. Third, to predict the gypsum purity and TAC according to blending ratio, DNN-based surrogate model is developed. Finally, cost optimal selection and blending ratio is derived under the two constraints: gypsum purity and total SOx absorbent consumption. As a results, the cost optimal blending ratio derived to low grade limestone (80.86 %), oyster shell (10.78 %), scallop shell (0.216 %), cockle shell (0.323 %), clam shell (2.426 %) and mussel shell (5.391 %). Keywords: Wet flue gas desulfurization system, waste sea shell, cost optimization |
| 저자 | 임종훈, 정수환, 조형태, 김정환 |
| 소속 | 한국생산기술(연) |
| 키워드 | 공정모사 및 설계; 공정최적화; 인공지능 기반 공정기술 |
| 원문파일 | 초록 보기 |
