ETRI-Knowledge Sharing Plaform



논문 검색
구분 SCI
연도 ~ 키워드


학술대회 Spatiotemporal Algal Bloom Prediction of Geum River, Korea Using the Deep Learning Models in Company With the EFDC Model
Cited 0 time in scopus Download 3 time Share share facebook twitter linkedin kakaostory
장미영, 김재영, 서동일, 김지용
Summer Simulation Conference (SummerSim) 2020, pp.1-11
20HB1700, 직독식 수질복합센서 및 초분광영상 기반 시공간 복합 인공지능 녹조 예측 기술, 권용환
Harmful algae bloom in river or lake is an annual phenomenon in many countries. A simulation is a powerful tool that can provide spatiotemporal analysis of the algal bloom based on the comprehensive interpretation of complex water quality interactions. Meanwhile, deep learning-based water quality predictions show remarkable performance with dense sensor data, but they are only applicable to specific points where we can monitor at all times. In this paper, we propose a spatiotemporal algal bloom prediction model by taking advantage of both the water quality simulation and deep learning models. The EFDC(Environmental Fluid Dynamics Code) model generates water quality data for the study area by simulating the measured boundary data. Then, we utilize the generated water quality data for a CNN based learning algorithm to predict a Chl-a concentration. We prove that our approach provides reliable performance in terms of prediction accuracy of Chl-a concentration.
Algal bloom prediction, Deep learning, EFDC simulation model, Water quality modeling
KSP 제안 키워드
Algal blooms, Bloom prediction, Chl-a, Comprehensive interpretation, EFDC model, Environmental fluid dynamics code, Geum river, Harmful algae bloom, Learning-based, Prediction accuracy, Quality data