Particulate matter has become one of the major issues in environmental sustainability, and its accurate measurement has grown in importance recently. Low-cost sensors (LCS) have been widely used to measure particulate concentration, but concerns about their accuracy remain. Previous research has shown that LCS data can be successfully calibrated using various machine learning algorithms. In this study, for better calibration, dynamic weight was introduced to the loss function of the LSTM model to amplify the loss, especially in a specific band. Our results showed that the dynamically weighted loss function resulted in better calibration in the specific band, where the model accepts the loss more sensitively than outside of the band. It was also confirmed that the dynamically weighted loss function can improve the calibration of the LSTM model in terms of both overall performance and local performance in bands. In a test case, the overall calibration performance was improved by about 12.57%, from 3.50 to 3.06, in terms of RMSE. The local calibration performance in the band improved from 4.25 to 3.77. Such improvements were achieved by varying coefficients of the dynamic weight. The results from different bands also indicated that having more data in a band will guarantee better improvement.
Disinfection is an effective microorganism inactivation method that has contributed historically to decreasing waterborne diseases. It is necessary to improve hydraulic efficiency for ensuring disinfection ability without creating disinfection by-products. However, many hydraulic efficiency indices, which are a type of black-box analyses based on residence time distribution curves, have been used to assess short-circuiting and mixing. We propose a novel index (internal short-circuiting index, ISI) and analysis approach (internal short-circuiting estimation method, ISEM) based on computational fluid dynamics (CFD) modelling for understanding the local hydrodynamics. Then, we implement ISEM to quantify the effect of the clearwell configuration with the different length-to-width and shape ratios on the hydraulic efficiency. As the hydraulic efficiency surrogated by T10/T converges to the maximum value, the ISI values at inlet and outlet reduce rapidly, and the recirculation and dead zones shrink in the channel zones. Thus, the ISI curve changes from a V shape to U shape. The ISEM demonstrates that it is applicable under various conditions and it enables engineers to design clearwells for optimizing the disinfection ability. Furthermore, the CFD model in this study can be combined with machine learning techniques in future studies to extract knowledge for reducing the computational cost.
25 February 2023 in Journal of the Korean Society for Marine Environment & Energy
최근 기후변화와 환경오염 등이 원인이 되어 해파리 대량 발생이 지속적으로 증가하고 있다. 특히 노무라입깃해파리는 국내에 대량 발생으로 피해를 입히는 주요 종으로 지적되며 적절한 대응과 조기 경보의 필요성이 대두되었다. 본 연구는 부산 해역의 해파리 대량 발생에 영향을 주는 수온, 유속 등 다양한 시계열 데이터를 해양수질자동측정망과 해수관측부이 데이터로부터 수집하고 LSTM-AE 모델을 활용하여 노무라입깃해파리의 대량 발생을 사전에 감지하고자 하였다. 학습을 마친 LSTM-AE 모델은 조기 경보를 위한 복원 오차를 0.7869로 설정하였을 때 0.6162의 정밀도 및 재현율, 그리고 0.7469의 정확도를 보였다. 제시된 LSTM-AE 모델을 활용하여 해파리 대량 발생을 2주 이전에 감지가능한 것이 확인되어 추후 노무라입깃해파리의 대량 발생으로 인한 피해 저감에 기여할 수 있을 것으로 기대된다.