Author(s): Takuma Suzuki; Makoto Nakatsugawa; Yosuke Kobayashi

Linked Author(s):

Keywords: Dam inflow prediction; Sparse modeling method; Elastic Net; Clustering; Generalization of prediction methods

Abstract: This study proposes a generalization method for dam inflow prediction in Japan. Climate change is causing frequent floods, and the Japanese government is requiring pre-flood discharge at water conservancy dams. However, it is complex and time-consuming to develop individual models for many dams. Therefore, this study attempted to generalize the inflow prediction method using Elastic Net, one of the sparse modeling methods. Dams were clustered based on geographic information of the watershed, and a representative model was selected for each cluster.

DOI: https://doi.org/10.64697/978-90-835589-7-4_41WC-P1819-cd

Year: 2025