A Methodology for the Modeling of Forced Dynamical Systems From Time Series Measurements Using Time-Delay Neural Networks

The main goal of this research was to develop and present a general, efficient, mathematical, and theoretical based methodology to model nonlinear forced-vibrating mechanical systems from time series measurements. A system identification modeling methodology for forced dynamical systems is presented based on a dynamic system theory and a nonlinear time series analysis that employ phase space reconstruction (delay vector embedding) in modeling dynamical systems from time series data using time-delay neural networks. The first part of this work details the modeling methodology, including background on dynamic systems, phase space reconstruction, and neural networks. In the second part of this work, the methodology is evaluated based on its ability to model selected analytical lumped-parameter forced-vibrating dynamic systems, including an example of a linear system predicting lumped mass displacement subjected to a displacement forcing function. The work discusses the application to nonlinear systems, multiple degree of freedom systems, and multiple input systems. The methodology is further evaluated on its ability to model an analytical passenger rail car predicting vertical wheel/rail force using a measured vertical rail profile as the input function. Studying the neural modeling methodology using analytical systems shows the clearest observations from results, providing prospective users of this tool an understanding of the expectations and limitations of the modeling methodology.

[1]  Yoshihiro Suda,et al.  Neural network based estimation of friction coefficient of wheel and rail , 1997, Proceedings of IEEE International Conference on Intelligent Engineering Systems.

[2]  David R. Martinelli,et al.  Optimization of railway operations using neural networks , 1996 .

[3]  Derong Liu,et al.  Cloning template design of cellular neural networks for associative memories , 1997 .

[4]  John Zolock,et al.  Application of Time Series Analysis and Neural Networks to the Modeling and Analysis of Forced Vibrating Mechanical Systems , 2003 .

[5]  Irwin W. Sandberg,et al.  Uniform approximation of multidimensional myopic maps , 1997 .

[6]  Fraser,et al.  Independent coordinates for strange attractors from mutual information. , 1986, Physical review. A, General physics.

[7]  A. Mees,et al.  Determining the Minimum Embedding Dimensions of Input–Output Time Series Data , 1998 .

[8]  David S. Broomhead,et al.  Delay Embeddings for Forced Systems. II. Stochastic Forcing , 2003, J. Nonlinear Sci..

[9]  Wlidzimierz Choromański.  Application of Neural Network for Intelligent Wheelset and Railway Vehicle Suspension Designs , 1996 .

[10]  L. Cao.  Practical method for determining the minimum embedding dimension of a scalar time series , 1997 .

[11]  J. Stark,et al.  Delay Embeddings for Forced Systems. I. Deterministic Forcing , 1999 .

[12]  Gary G. Yen.  Health monitoring of vibration signatures in rotorcraft wings , 2004, Neural Processing Letters.

[13]  Brauer,et al.  3D-to-2D transition by Cs intercalation of VSe2. , 1993, Physical review letters.

[14]  Tommy W. S. Chow,et al.  Feedforward Neural Networks Based Input-Output Models for Railway Carriage System Identification , 2004, Neural Processing Letters.

[15]  Kennel,et al.  Method to distinguish possible chaos from colored noise and to determine embedding parameters. , 1992, Physical review. A, Atomic, molecular, and optical physics.