Potential Flood Prediction at Downstream of Hydropower System based on ANN and Fuzzy

Authors

  • Nurul Najwa Anuar, M. Reyasudin Basir Khan, Aizat Faiz Ramli

Abstract

Due to heavy raining season, hydropower scheme experience problems such as high flooding particularly at downstream zone. Hence, hydropower station required a discharge forecasting alongside flood prediction to prevent flooding at downstream area. Moreover, the discharge forecasting is important for optimization of power generation through water regulation. This paper discusses the use of Artificial Neural Network (ANN) to predict the downstream discharge pattern of a cascade hydropower station. Meanwhile, the fuzzy inference system was used to determine the potential of flood risk based on predicted discharge pattern. This study starts with data collection at a selected hydropower station. Data collected are Forebay Elevation (FBE), inflow and discharge that were used as input parameters for prediction algorithm. The Elman Neural Network architecture was used in this study for discharge predictions. Next, the optimum number of hidden neurons and training algorithm were identified. The performance of model was assessed using performance metrics such as Root Mean Square Error (RMSE), Mean Square Error (MSE), Mean Absolute Error (MAE) and Sum Square Error (SSE). The result shows ANN exhibits high performance discharge forecasting through minimal error values. Finally, Fuzzy model was used to identify the flood risk based on discharge and water level.

Downloads

Published

2020-05-10

Issue

Section

Articles