Forecasting model for the change of reservoir water level stage based on temporal pattern of reservoir water level

摘要

Reservoir water level forecasting is vital in reservoir operation and management.The output of the forecasting model can be used in reservoir decision support systems.This study demonstrates the application of Artificial Neural Network (ANN) in developing the forecasting model for the change of reservoir water level stage.In this study, sliding window technique has been used to extract the temporal pattern that represents time delays in the reservoir water level. The patterns are used as input to the ANN model.The results show that a model with 4 days of time delay has produced the acceptable performance with both low error rate and high accuracy.