Investigating the Accuracy of Hybrid Models with Wavelet Transform in the Forecast of Watershed Runoff

In the hydrological cycle, runoff precipitation is one of most significant and complex phenomena. order to develop improve predictive models, different perspectives have been presented in its modeling. Hydrological processes can be confidently modeled with help artificial intelligence techniques. this study, Leilanchai watershed was simulated using neural networks (ANNs) M5 model tree methods their hybrid wavelet transform. Seventy percent data used train state thirty test were collected from 2000 2021. addition daily monthly scales, observed results compared within each scale. Initially, rainfall time series divided into multiple sub-series transform combat instability. The resultant subheadings then utilized as input for an ANN tree. demonstrated that models improved model's accuracy by 4% 26%. It also tree's 41%. wavelet-M5 does not diminish same degree wavelet-ANN (WANN) forecast horizon lengthens. Consequently, has a relatively stable behavior pattern. Finally, conjunction transform, accuracy.