Evaluate the performance of the artificial neural network model in hybrids with different preprocessors to forecast the streamflow time series

Abstract

Abstract In arid and semi-arid regions, accurate forecasting of hydrological processes and sustainable management of water resources is inevitable due to the floods and water crisis. In the present study, was the various data preprocessing techniques used to help understand the processes and increase the accuracy of the artificial neural network (ANN) model. To forecast streamflow from preprocessors including, discrete wavelet transform (DWT), empirical mode decomposition (EMD), complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), successive variational mode decomposition (SVMD), and multi-filter of the smoothing (MFS) in hybrid with the ANN model, were utilized. In general, the results showed that choosing the suitable approach to applying input data and the suitable preprocessor can have a highly favorable effect on the performance of hybrid models for daily streamflow forecasting. Overall, the results indicate that had the best performance the MFS-ANN model in short-term forecasting and the SVMD-ANN model in long-term forecasting.