遺伝的プログラミングを活用した洪水予測技術の開発～草木ダム及び佐波川におけるケーススタディ～

抄録

Promotion of flood forecasting system is crucial for reducing flood disasters urgently and effectively, especially in medium- and small-scale rivers. The authors propose a new method of flood forecasting using Genetic Programming (GP) and GMDH. Traditional method based on physical model takes time to analyze the basin and river channel characteristics, but the new method has potential to make a rainfall-runoff (or rainfall-water level) model automatically by learning the past data of river water level or dam inflow and rainfall.

Case studies were conducted for the dam inflow prediction at the Kusaki Dam and the water-level prediction at Shinbashi point of the Saba River. The results from both the case studies were encouraging to promote the new method, because the inflow or water-level predictions with 6-hour lead time were relatively good. Some results suggested the necessity of the direct incorporation of spatial distribution of rainfall in the upstream area for better accuracy.