Comparison between NARX parameter estimation methods with Binary Particle Swarm Optimization-based structure selection method

This paper compares between several parameter estimation methods (Classical Gramm-Schmidt (CGS), Modified Gramm-Schmidt (MGS), Householder Transform (HT), and Givens Rotation (GR)) for Nonlinear Autoregressive with Exogenous Inputs (NARX) system identification of a DC motor model using Binary Particle Swarm Optimization (BPSO) by (Kennedy and Eberhart, 1997) as the structure selection method. First, we describe the application of BPSO for model structure selection, by representing its particles' solutions as probabilities of change in a binary string. The binary string was then used to select a subset of regressor columns from the regressor matrix. The parameters (linear least squares solution) were then estimated using CGS, MGS, GT and GR. One-Step Ahead (OSA) and correlations tests performed on the DC motor dataset show that: 1) The BPSO-based selection method has the potential to become an effective method to determine parsimonious NARX model structure, and 2) The CGS, HT and GR algorithms were the best choices for parameter estimation of the model.

[1]  David Rees,et al.  Nonlinear gas turbine modeling using NARMAX structures , 2001, IEEE Trans. Instrum. Meas..

[2]  Sheng Chen,et al.  Extended model set, global data and threshold model identification of severely non-linear systems , 1989 .

[3]  Stephen A. Billings,et al.  An alternative solution to the model structure selection problem , 2001, IEEE Trans. Syst. Man Cybern. Part A.

[4]  Russell C. Eberhart,et al.  A discrete binary version of the particle swarm algorithm , 1997, 1997 IEEE International Conference on Systems, Man, and Cybernetics. Computational Cybernetics and Simulation.

[5]  O. Weck,et al.  A COMPARISON OF PARTICLE SWARM OPTIMIZATION AND THE GENETIC ALGORITHM , 2005 .

[6]  Mohd Nasir Taib,et al.  Particle Swarm Optimization for NARX structure selection — Application on DC motor model , 2010, 2010 IEEE Symposium on Industrial Electronics and Applications (ISIEA).

[7]  N. C. Van de Giesen,et al.  Monthly streamflow prediction in the Volta Basin of West Africa: A SISO NARMAX polynomial modelling , 2008 .

[8]  Russell C. Eberhart,et al.  Recent advances in particle swarm , 2004, Proceedings of the 2004 Congress on Evolutionary Computation (IEEE Cat. No.04TH8753).

[9]  M. Clerc,et al.  The swarm and the queen: towards a deterministic and adaptive particle swarm optimization , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).

[10]  Niels Kjølstad Poulsen,et al.  Neural Networks for Modelling and Control of Dynamic Systems: A Practitioner’s Handbook , 2000 .

[11]  Shen Ping,et al.  NDEKF Neural Network Applied to Electronically Controlled Fuel Injection System , 2007, 2007 2nd IEEE Conference on Industrial Electronics and Applications.

[12]  Jianwen Yang,et al.  Dynamic Modeling Using Cascade-Correlation RBF Networks for Tilt Rotor Aircraft Platform , 2005, 2005 International Conference on Neural Networks and Brain.

[13]  Robert E. Kearney,et al.  A least-squares parameter estimation algorithm for switched Hammerstein systems with applications to the VOR , 2005, IEEE Transactions on Biomedical Engineering.

[14]  Robert E. Kearney,et al.  NARMAX representation and identification of ankle dynamics , 2003, IEEE Transactions on Biomedical Engineering.

[15]  Huijun Li,et al.  A Nonlinear Predictive Model Based on Multilayer Perceptron Network , 2007, 2007 IEEE International Conference on Automation and Logistics.

[16]  Yue Shi,et al.  A modified particle swarm optimizer , 1998, 1998 IEEE International Conference on Evolutionary Computation Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98TH8360).

[17]  P. Caminal,et al.  Model identification of the neural control of the cardiovascular system using NARMAX models , 1991, [1991] Proceedings Computers in Cardiology.

[18]  Xingsheng Gu,et al.  A fuzzy identification algorithm for NARMAX model and its application to the estimation of oxygen concentration , 2004, Fifth World Congress on Intelligent Control and Automation (IEEE Cat. No.04EX788).

[19]  Takuji Nishimura,et al.  Mersenne twister: a 623-dimensionally equidistributed uniform pseudo-random number generator , 1998, TOMC.

[20]  Yan Bai,et al.  A New Nonlinear System Identification Method Using Gene Expression Programming , 2007, 2007 International Conference on Mechatronics and Automation.

[21]  R. Eberhart,et al.  Comparing inertia weights and constriction factors in particle swarm optimization , 2000, Proceedings of the 2000 Congress on Evolutionary Computation. CEC00 (Cat. No.00TH8512).