MIMO system identification with extended MADALINE neural network trained by Levenberg-Marquardt Method

Presented in this paper is an extended version of the Multi-ADAptive LINear Element (MADALINE) neural network, termed EMADALINE, for On-line System identification of Multi-Input Multi-Output (MIMO) linear time-varying (LTV) systems Trained by Levenberg-Marquardt Method. A sliding window on the data set is used in the learning algorithm for the purpose of improving convergence speed during training and thus better tracking system parameters. Based on the input output polynomial model, which can be easily transformed into the row canonical state space model, Tapped delay lines are introduced, so the EMADALINE becomes recurrent in nature and thus is suitable for parameter estimation of such systems. The EMADALINE can then be setup under the assumption that the system structure is known in advance. The estimated parameters are obtained as the weights of trained individual neurons of the EMADALINE. The method is implemented in MATLAB and simulation study was then performed on a few well known examples. Simulation results show that the algorithms offer impressive results on convergence speed improvement. This work is based on author's previous work on MIMO systems' identification (Wenle Zhang, 2010).

[1]  Rahmat A. Shoureshi,et al.  Neural networks for system identification , 1990 .

[2]  Lennart Ljung,et al.  System Identification: Theory for the User , 1987 .

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

[4]  Wenle Zhang,et al.  MADALINE neural network for parameter estimation of LTI MIMO systems , 2010, Proceedings of the 29th Chinese Control Conference.

[5]  A. Gretton,et al.  Support vector regression for black-box system identification , 2001, Proceedings of the 11th IEEE Signal Processing Workshop on Statistical Signal Processing (Cat. No.01TH8563).

[6]  Wenle Zhang,et al.  An extended ADALINE neural network trained by Levenberg-Marquardt method for system identification of linear systems , 2013, 2013 25th Chinese Control and Decision Conference (CCDC).

[7]  Ferenc Szeifert,et al.  System Identification Using Delaunay Tessellation of Self-Organizing Maps , 2003 .

[8]  Bernard Widrow,et al.  30 years of adaptive neural networks: perceptron, Madaline, and backpropagation , 1990, Proc. IEEE.

[9]  Harpreet Singh,et al.  Single layer neural networks for linear system identification using gradient descent technique , 1993, IEEE Trans. Neural Networks.

[10]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[11]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[12]  John J. Hopfield,et al.  Neural networks and physical systems with emergent collective computational abilities , 1999 .

[13]  Gonzalo Joya,et al.  Gray box identification with hopfield neural networks , 2004 .

[14]  S Z Qin,et al.  Comparison of four neural net learning methods for dynamic system identification , 1992, IEEE Trans. Neural Networks.