Nonlinear dynamic system identification using Volterra series: Multi-objective optimization approach

In this paper, system identification of the non-linear dynamic system based on optimized Volterra model structure is considered. Model structure selection is an important step in system identification, which involves the selection of variables and terms of a model. The important issue is choosing a compact model representation where only significant terms are selected among all the possible ones beside good performance. An automated algorithm based on multi-objective optimization is proposed. The developed model should fulfil two criteria or objectives namely good predictive accuracy and optimum model structure. Genetic algorithm is applied to search the significant Volterra kernels among all possible candidate model combinations. The result shows that the proposed algorithm is able to correctly identify the simulated examples and adequately model the nonlinear discrete dynamic system.

[1]  Kalyanmoy Deb,et al.  Running performance metrics for evolutionary multi-objective optimizations , 2002 .

[2]  O. Nelles.  Nonlinear System Identification: From Classical Approaches to Neural Networks and Fuzzy Models , 2000 .

[3]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[4]  S. Fakhouri.  Identification of the Volterra kernels of nonlinear systems , 1980 .

[5]  S. Billings,et al.  Correlation based model validity tests for non-linear models , 1986 .

[6]  Leon O. Chua,et al.  Fading memory and the problem of approximating nonlinear operators with volterra series , 1985 .

[7]  Leehter Yao.  Genetic algorithm based identification of nonlinear systems by sparse Volterra filters , 1999, IEEE Trans. Signal Process..

[8]  S. Billings,et al.  Analysis of nonlinear oscillators using Volterra series in the frequency domain , 2009 .

[9]  M. J. Korenberg,et al.  A robust orthogonal algorithm for system identification and time-series analysis , 1989, Biological Cybernetics.

[10]  Hazem M. Abbas,et al.  Volterra system identification using adaptive genetic algorithms , 2004, Appl. Soft Comput..

[11]  J. D. Schaffer,et al.  Some experiments in machine learning using vector evaluated genetic algorithms (artificial intelligence, optimization, adaptation, pattern recognition) , 1984 .

[12]  Wagner Caradori do Amaral,et al.  CHOICE OF FREE PARAMETERS IN EXPANSIONS OF DISCRETE-TIME VOLTERRA MODELS USING KAUTZ FUNCTIONS , 2005 .

[13]  Sayed Mohammad Reza Loghmanian,et al.  Multi-objective optimization of neural network structure for system identification using genetic algorithm , 2009 .

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

[15]  David T. Westwick,et al.  Factors Affecting Volterra Kernel Estimation: Emphasis on Lung Tissue Viscoelasticity , 2004, Annals of Biomedical Engineering.

[16]  Malcolm J. Hawksford,et al.  Identifica-tion of Discrete Volterra Series Using Maximum Length Sequences , 1996 .

[17]  Gary G. R. Green,et al.  Calculation of the Volterra kernels of non-linear dynamic systems using an artificial neural network , 1994, Biological Cybernetics.

[18]  André Monin,et al.  New method for identifying finite degree Volterra series , 2008, Autom..

[19]  Kiyoshi Tanaka,et al.  Analysis of NSGA-II and NSGA-II with CDAS, and Proposal of an Enhanced CDAS Mechanism , 2009, J. Adv. Comput. Intell. Intell. Informatics.

[20]  Hazem M. Abbas,et al.  An adaptive evolutionary algorithm for Volterra system identification , 2005, Pattern Recognit. Lett..

[21]  R. K. Ursem.  Multi-objective Optimization using Evolutionary Algorithms , 2009 .

[22]  Daniel E. Rivera,et al.  Control Relevant Model Reduction of Volterra Series Models , 1998 .

[23]  Kalyanmoy Deb,et al.  Muiltiobjective Optimization Using Nondominated Sorting in Genetic Algorithms , 1994, Evolutionary Computation.

[24]  Pier Andrea Traverso,et al.  Non-linear dynamic system modelling based on modified Volterra series approaches , 2003 .

[25]  M. Tummala,et al.  Identification of Volterra systems with a polynomial neural network , 1992, [Proceedings] ICASSP-92: 1992 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[26]  Tony J. Dodd,et al.  A NEW SOLUTION TO VOLTERRA SERIES ESTIMATION , 2002 .