报告题目：Machine Learning Approaches for Hydrological Modeling and Forecasting
      报  告 人：Prof. Sunmin Kim

邀  请 人：金钟硕教授，陈杰 教授

时      间：2021年11月26日（星期五）下午3:30-4:30

地      点：国家重点实验室学术报告厅（农水楼一楼）

              Zoom会议（ID:  873 912 5358）

              密码：123456

报告人简介：  

Graduated with Bachler and Master degrees in ChungNam National University, and Doctoral degree in Kyoto University. Now Dr.Kim is associate professor, Dept. of Civil & Earth Resources Eng, Kyoto University. Research topics are focusing on realtime flood forecasting as well as short-term rainfall forecasting. Recent topics are based on artificial neural networks and deep learning algorithms. Prof.Kim’s professional affiliations have included the followings:  Japan Society of Civil Engineers (JSCE), Japan Society of Hydrology and Water Resources (JSHWR), Korea Water Resources Association (KWRA), American Geophysical Union (AGU). And he has been granted for Pioneering Research as principal investigator by JSPS between 2014-2019, to name a few.

报告简介：

Machine learning approaches are gaining lots of attentions in hydrological modeling and forecasting, thanks to the drastic improvements and recommendation of new techniques such deep learning algorithms that can utilize the accumulated digital datasets and improved computing power. Machine learning approaches can be a plausible alternative in practical applications, especially in cases where physical modeling approach poses a significant challenge in terms of mathematical complexity, insufficient reliability of initial or boundary conditions and the various assumptions that need to be made in term of their model parametrization.

This presentation aims to explore the most recent progress made in terms of machine learning approaches, especially with artificial neural networks (ANN), applied in hydrological modeling and forecasting based on several application cases. In the presentation, the basic concept of ANN is introduced and several application examples, such as river discharge forecasting and rainfall forecasting, will be introduced. Considerations and limitations in ANN applications will be also discussed.

欢迎相关专业教师和研究生的光临！