Forecasting Method of Recession Time Constant of River Flow Rate into a Dam after Rainfall

Abstract

This paper describes an application of neural network for forecasting of the recession time constant of river flow rate into a dam after the rainfall. We proposed that the used data to forecast the recession time constant is classified by cluster analysis. A neural network system is developed through a case study on a dam for hydropower plant located the upper district of the Yahagi River in Central Japan. It is found from our investigations that the forecasted results of the recession time constant of river flow rate are improved by classifying the rainfalls based on cluster analysis.