Volterra system-based neural network modeling by particle swarm optimization approach

In this paper, we propose a novel identification method for nonlinear discrete dynamic systems. A feedforward neural network with the structure of Volterra system is newly presented. This kind of mathematical model possesses more adjustable parameters than the original Volterra system to further enhance the modeling capacity. Tuning adjustable parameters inside the neural network is based on the particle swarm optimization (PSO) instead of the commonly used back-propagation method. The PSO algorithm is with multiple direction searches and can easily find out the global solution for the given optimization problem. This paper also develops the whole design steps for PSO-based feedforward neural network modeling for nonlinear discrete systems. Two kinds of examples are illustrated to validate the feasibility and efficiency of the proposed method. In addition, some examinations containing different initial conditions and population sizes, and presence of measurement noises are considered to evaluate the modeling performance.

[1]  Jie Ding,et al.  Time series AR modeling with missing observations based on the polynomial transformation , 2010, Math. Comput. Model..

[2]  Sheng-Fuu Lin,et al.  Reinforcement group cooperation-based symbiotic evolution for recurrent wavelet-based neuro-fuzzy systems , 2009, Neurocomputing.

[3]  Musa H. Asyali,et al.  Nonlinear system identification via Laguerre network based fuzzy systems , 2009, Fuzzy Sets Syst..

[4]  Anuj Narang,et al.  Reparametrized ARX models for predictive control of staged and packed bed distillation columns , 2010 .

[5]  B. Wayne Bequette,et al.  Nonlinear system identification and model reduction using artificial neural networks , 2003, Comput. Chem. Eng..

[6]  Héctor Pomares,et al.  Soft-computing techniques and ARMA model for time series prediction , 2008, Neurocomputing.

[7]  W. Chang,et al.  Design of a higher-order digital differentiator using a particle swarm optimization approach , 2008 .

[8]  Ricardo J. G. B. Campello,et al.  Optimal expansions of discrete-time Volterra models using Laguerre functions , 2003, Autom..

[9]  Heinz Koeppl,et al.  Minimum Mean-Square Error Equalization for Second-Order Volterra Systems , 2008, IEEE Transactions on Signal Processing.

[10]  Hamidreza Modares,et al.  Parameter estimation of bilinear systems based on an adaptive particle swarm optimization , 2010, Eng. Appl. Artif. Intell..

[11]  Long Li,et al.  A modified gradient-based neuro-fuzzy learning algorithm and its convergence , 2010, Inf. Sci..

[12]  L. Coelho,et al.  Nonlinear model identification of an experimental ball-and-tube system using a genetic programming approach , 2009 .

[13]  Hao Tang,et al.  Fault diagnosis approach based on Volterra models , 2010 .

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

[15]  W. Chang.  PID control for chaotic synchronization using particle swarm optimization , 2009 .

[16]  Jiashu Zhang,et al.  A novel adaptive bilinear filter based on pipelined architecture , 2010, Digit. Signal Process..

[17]  Feng Qian,et al.  Identification and control of nonlinear systems by a time-delay recurrent neural network , 2009, Neurocomputing.

[18]  Martin T. Hagan,et al.  Neural network design , 1995 .

[19]  James Kennedy,et al.  Particle swarm optimization , 2002, Proceedings of ICNN'95 - International Conference on Neural Networks.

[20]  Aydin Kizilkaya,et al.  ARMA model parameter estimation based on the equivalent MA approach , 2006, Digit. Signal Process..