An adaptive fuzzy wavelet neural network with gradient learning algorithm for nonlinear function approximation

Author

Oysal, Y. ; Yilmaz, Sabri

Author_Institution

Comput. Eng. Dept., Anadolu Univ., Eskisehir, Turkey

fYear

2013

fDate

10-12 April 2013

Firstpage

152

Lastpage

157

Abstract

In this paper a new adaptive fuzzy wavelet neural network (AFWNN) model is proposed for nonlinear function approximation problems. The AFWNN model is a Takagi-Sugeno-Kang (TSK) fuzzy system in which the membership functions of fuzzy rules are replaced with wavelet basis functions, which are known to have time and frequency localization properties. The AFWNN model is trained using a gradient-based optimization algorithm for certain types of nonlinear time series, for instance fractal processes and the simulation results are found to be substantially more accurate than alternative methods.

Keywords

function approximation; fuzzy neural nets; fuzzy systems; gradient methods; learning (artificial intelligence); nonlinear functions; optimisation; time series; wavelet transforms; AFWNN model; TSK fuzzy system; Takagi-Sugeno-Kang fuzzy system; adaptive fuzzy wavelet neural network; fractal process; frequency localization properties; fuzzy rules membership function; gradient learning algorithm; gradient-based optimization algorithm; nonlinear function approximation problem; nonlinear time series; time localization properties; wavelet basis functions; Adaptation models; Autoregressive processes; Computational modeling; Input variables; Predictive models; Time series analysis; Training; ANFIS; Fuzzy Systems; Time Series Prediction; Wavelet Neural Networks;

fLanguage

English

Publisher

ieee

Conference_Titel

Networking, Sensing and Control (ICNSC), 2013 10th IEEE International Conference on

Conference_Location

Evry

Print_ISBN

978-1-4673-5198-0

Electronic_ISBN

978-1-4673-5199-7

Type

conf

DOI

10.1109/ICNSC.2013.6548728

Filename

6548728