Wavelet networks for nonlinear system modeling

This study presents a nonlinear systems and function learning by using wavelet network. Wavelet networks are as neural network for training and structural approach. But, training algorithms of wavelet networks is required a smaller number of iterations when the compared with neural networks. Gaussian-based mother wavelet function is used as an activation function. Wavelet networks have three main parameters; dilation, translation, and connection parameters (weights). Initial values of these parameters are randomly selected. They are optimized during training (learning) phase. Because of random selection of all initial values, it may not be suitable for process modeling. Because wavelet functions are rapidly vanishing functions. For this reason heuristic procedure has been used. In this study serial-parallel identification model has been applied to system modeling. This structure does not utilize feedback. Real system outputs have been exercised for prediction of the future system outputs. So that stability and approximation of the network is guaranteed. Gradient methods have been applied for parameters updating with momentum term. Quadratic cost function is used for error minimization. Three example problems have been examined in the simulation. They are static nonlinear functions and discrete dynamic nonlinear system.

[1]  Roberto Kawakami Harrop Galvão,et al.  LINEAR-WAVELET MODELS FOR SYSTEM IDENTIFICATION , 2002 .

[2]  Gérard Dreyfus,et al.  Training wavelet networks for nonlinear dynamic input-output modeling , 1998, Neurocomputing.

[3]  Marios M. Polycarpou,et al.  Adaptive wavelet control of nonlinear systems , 1997, Proceedings of the 36th IEEE Conference on Decision and Control.

[4]  Qinghua Zhang,et al.  Wavelet networks , 1992, IEEE Trans. Neural Networks.

[5]  Fei-Yue Wang,et al.  Modular structure of fuzzy system modeling using wavelet networks , 2005, Proceedings. 2005 IEEE Networking, Sensing and Control, 2005..

[6]  Roberto Muñoz,et al.  Multivariate calibration model from overlapping voltammetric signals employing wavelet neural networks , 2006 .

[7]  Marc Thuillard,et al.  A Review of Wavelet Networks, Wavenets, Fuzzy Wavenets and their Applications , 2002, Advances in Computational Intelligence and Learning.

[8]  Daming Shi,et al.  The construction of wavelet network for speech signal processing , 2006, Neural Computing & Applications.

[9]  Gérard Dreyfus,et al.  Initialization by selection for wavelet network training , 2000, Neurocomputing.

[10]  Kumpati S. Narendra,et al.  Identification and control of dynamical systems using neural networks , 1990, IEEE Trans. Neural Networks.

[11]  Fulei Chu,et al.  A Hierarchical Evolutionary Algorithm for Constructing and Training Wavelet Networks , 2002, Neural Computing & Applications.

[12]  Emine Dogru Bolat,et al.  Comparison of Wavenet and Neuralnet for System Modeling , 2005, KES.

[13]  Yagyensh C. Pati,et al.  Analysis and synthesis of feedforward neural networks using discrete affine wavelet transformations , 1993, IEEE Trans. Neural Networks.

[14]  Daniel W. C. Ho,et al.  Fuzzy wavelet networks for function learning , 2001, IEEE Trans. Fuzzy Syst..

[15]  Nalan Ozkurt,et al.  The implementation of nonlinear dynamical systems with wavelet network , 2006 .

[16]  Yasar Becerikli,et al.  On three intelligent systems: dynamic neural, fuzzy, and wavelet networks for training trajectory , 2004, Neural Computing & Applications.

[17]  Qinghua Zhang,et al.  Using wavelet network in nonparametric estimation , 1997, IEEE Trans. Neural Networks.

[18]  Yasar Becerikli,et al.  On a Dynamic Wavelet Network and Its Modeling Application , 2003, ICANN.