ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Hybrid Models Performance Assessment to Predict Flow of Gamasyab River

Journal: International Journal of Advanced Biological and Biomedical Research (Vol.2, No. 5)

Publication Date:

Authors : ; ; ; ;

Page : 1837-1846

Keywords : Hybrid model; Wavelet - Artificial Neural Network; Wavelet - Adaptive Neural Fuzzy Inference System; Gamasyab River; Monthly flow prediction;

Source : Download Find it from : Google Scholarexternal

Abstract

Awareness of the level of river flow and its fluctuations at different times is one of the significant factor to achieve sustainable development for water resource issues. Therefore, the present study two hybrid models, Wavelet- Adaptive Neural Fuzzy Interference System (WANFIS) and Wavelet- Artificial Neural Network (WANN) are used for flow prediction of Gamasyab River (Nahavand, Hamedan, Iran). For this purpose, original time series using wavelet theory decomposed to multi time sub-signals, then these decomposed sub-signals as in input data are used in Adaptive Neural Fuzzy Interference System (ANFIS) and Artificial Neural Network (ANN) for monthly flow prediction. The obtained result shows that WANFIS model has better performance than WANN and can be used for short term and long term flow prediction. One of the weaknesses of fuzzy models is the model estimation error in minimum and maximum points. Which this problem can solve by using hybrid models of wavelet - fuzzy inference system.Also based on results of hybrid model of wavelet- network, it can be concluded that to achieve accurate estimation of the number of different intermediate layers are examined and using one intermediate layer in all conditions is not enough to achieve the best results. Generally, hybrid model of wavelet - Adaptive Neural Fuzzy Interference System have better performance in estimation of the extent points and it is better method for prediction of Gamasyab River flow.

Last modified: 2019-07-05 05:02:07