0100479010

open access

Version of Record

Optimizing River Discharge Forecasts with Machine Learning for Japanese Public Dams Operation

English (英語)

国民経済雑誌

227(2)

45-64

神戸大学経済経営学会

2023-02-10

2024-03-04

Despite a climate and topology favorable to hydropower(HP) generation, HP only accounts for 4¥％ of today’s Japanese primary energy consumption mix. In recent years, calls for improving the efficiency of Japanese HP have emerged from prominent voices in the Ministry of Land, Infrastructure, Transport and Tourism. Among potential optimizations, data-driven dam operation policies using accurate river discharge forecasts have been advocated for. Meanwhile, Machine Learning(ML) has recently made important strides in hydrological modeling, with forecast accuracy improvements demonstrated on both precipitation nowcasting and river discharge prediction. We are interested in the convergence of these societal and technological contexts : our goal is to provide scientific evidence and actionable insights towards the implementation of more efficient dam operation policies using ML-based river discharge forecasts on a national scale. Towards this goal, this work presents a framework that simulates dam operation using uncertain river discharge forecasts. This framework aims to jointly quantify river discharge forecast accuracy, and the impact of forecast errors on dam operation efficiency. We conduct a preliminary study of ML-based discharge forecast on a dataset of 127 Japanese public dams we have assembled, and attempt to quantify the impact of different forecast error components on operation efficiency.

Dam Operation

紀要論文