| 학회 | 한국화학공학회 |
| 학술대회 | 2017년 봄 (04/26 ~ 04/28, ICC 제주) |
| 권호 | 23권 1호, p.234 |
| 발표분야 | 공정시스템 |
| 제목 | 딥러닝기반 이차침천조의 슬러지 침강성 예측 및 분류 |
| 초록 | In biological wastewater treatment plants, the secondary settler is crucial part of plant. Thus, we formulate an efficient model for classification of secondary settler to achieve adequate effluent quality in WWTP. We propose time series prediction by new state of the art model, deep belief network (DBN) and deep neural network (DNN). The DNN has applied to predict the solid volume index of the settler. SVI provides a good estimation of the settling properties of sludge. To effectively monitor the features of secondary settler, DBN has used which are a probabilistic generative network composed of multiple layers of restricted Boltzmann machine (RBM). We use 2-layer deep network of RBMs to capture the feature of input space of time series data. The proposed method showed efficient classification results than neural network i.e. 87.34% predicted performance. The proposed method can effectively use in WWTP for simultaneously prediction of SVI value and classifying the current state (Normal, Bad, and Bulking state) of the secondary settler. This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIP) (No.2015R1A2A2A11001120). |
| 저자 | Usman Safder, 남기전, 이승철, 허성구, 유창규 |
| 소속 | 경희대 |
| 키워드 | 공정모델링; 공정최적화; 공정제어 |
| 원문파일 | 초록 보기 |
