The following information was submitted:
Transactions: INTERNATIONAL JOURNAL of ENERGY and ENVIRONMENT
Transactions ID Number: 19-145
Full Name: Amir Alikhani
Position: Assistant Professor
Age: ON
Sex: Male
Address: 11, 4th floor No 27 Street No 16 Street Amirabad Ave, Tehran, Iran
Country: IRAN
Tel:
Tel prefix:
Fax:
E-mail address: dramiralikhani@yahoo.com
Other E-mails:
Title of the Paper: Combination of neuro fuzzy and wavelet model usage in river engineering
Authors as they appear in the Paper: Amir Alikhani
Email addresses of all the authors: dramiralikhani@yahoo.com
Number of paper pages: 13
Abstract: The sediment load transported in river is the most complex hydrological and environmental phenomenon due to the large number of obscure parameters such as spatial variability of basin characteristics and river discharge patterns. Wavelet analysis, which give information in both the time and frequency domains of the signal, give considerable knowledge about the physical form of the data. Neuro-fuzzy modeling is another method that refers to the approach of applying deferent learning algorithms developed in the neural network literature to fuzzy modeling or a fuzzy inference system (FIS). Combination of neuro fuzzy and wavelet model usage in river engineering were used in this paper. Three models were investigated and compared with each other. In this research suspended sediment load (SSL) prediction in a gauging station in the USA by neuro-fuzzy (NF), conjunction of wavelet analysis and neuro-fuzzy (WNF) and conventional sediment rating curve (SRC) models were inves!
tigated. In the proposed WNF model, observed time series of river discharge and SSL were decomposed at different scales by wavelet analysis. As input to the NF model for prediction of SSL in one day ahead summed effective time series of discharge and SSL were imposed. The results showed that the WNF model performance gave better in prediction rather than the two other models. The WNF model produced reasonable predictions for the extreme values. Furthermore, the cumulative SSL estimated by this technique was closer to the actual data than the others one. This model is able to simulate hysteresis phenomenon, where the SRC model has not this ability. In the best of the author's knowledge, this paper is the first application of wavelet-neuro-fuzzy hybrid model for prediction of SSL. The results of this paper illustrated the advantage of WNF model to NF approach in simulation of suspended sediment time series. Reasonable contributions are presented to the water resources and env!
ironmental engineering literature by this article.
Keywords: Hysteresis, modellings, neuro-fuzzy, Time series, Wavelet.
EXTENSION of the file: .doc
Special (Invited) Session: Wavelet and neuro-fuzzy combination model for predicting suspended sediment transport in rivers
Organizer of the Session: 618-534
How Did you learn about congress:
IP ADDRESS: 87.107.33.251
