Adaptive neural networks for model updating of structures

A model updating methodology based on an adaptive neural network (NN) model is proposed in this study. The NN model has a feedforward architecture and is first trained off-line using some training data that are obtained from finite-element analyses and contain modal parameters as inputs and structural parameters as outputs. To reduce the number of training data while maintaining the data completeness, the variation of structural parameters is arranged using an orthogonal array. This NN model is then adaptively retrained on-line during the model updating process in order to eliminate the difference between the measured and the predicted modal parameters. A modified back-propagation algorithm is developed, in which the learning rate is dynamically adjusted once every few iterations. A jump factor is introduced to overcome the numerical difficulty caused by the saturation of the sigmoid function in order to improve the convergence performance of the NN model. The current adaptive NN updating procedure is applied to a suspension bridge model and verified both numerically and experimentally. The results indicate that by adaptively training the NN model and iteratively adjusting the structural parameters, it is possible to reduce the differences between the measured and the predicted frequencies from a maximum of 17% to 7% for the first eight vertical modes.

[1]  C. Fritzen,et al.  DAMAGE DETECTION BASED ON MODEL UPDATING METHODS , 1998 .

[2]  M. Baruch.  Optimal correction of mass and stiffness matrices using measured modes , 1982 .

[3]  John E. Mottershead,et al.  Finite Element Model Updating in Structural Dynamics , 1995 .

[4]  Hamid Ahmadian,et al.  Parameter Selection Strategies in Finite Element Model Updating , 1997 .

[5]  Prabhat Hajela,et al.  Neurobiological computational models in structural analysis and design , 1991 .

[6]  Jon D. Collins,et al.  Statistical Identification of Structures , 1973 .

[7]  U. S. Fernando,et al.  Modern Practice in Stress and Vibration Analysis , 1993 .

[8]  Donald Grant Collins,et al.  Laminar Viscous-Inviscid Interactions at Transonic Speeds , 1974 .

[9]  J Bowe,et al.  Total quality management. , 1992, Contemporary longterm care.

[10]  James H. Garrett,et al.  Use of neural networks in detection of structural damage , 1992 .

[11]  Daniel J. Inman,et al.  Solution of the model correction problem via inverse methods , 1991 .

[12]  A. H,et al.  REGULARISATION METHODS FOR FINITE ELEMENT MODEL UPDATING , 1998 .

[13]  R. Daniel VanLuchene,et al.  INTEGRATED ASSESSMENT OF SEISMIC DAMAGE IN STRUCTURES , 1994 .

[14]  J. D. Collins,et al.  Statistical Identification of Structures , 1973 .

[15]  Roger C. E. Tan,et al.  Some acceleration methods for iterative computer of derivatives of eigenvalues and eigenvectors , 1989 .

[16]  Mauro J. Atalla,et al.  Model updating using neural networks , 1996 .

[17]  S. Masri,et al.  Identification of Nonlinear Dynamic Systems Using Neural Networks , 1993 .

[18]  Ri Levin,et al.  DYNAMIC FINITE ELEMENT MODEL UPDATING USING NEURAL NETWORKS , 1998 .

[19]  Daniel J. Inman,et al.  ON MODEL UPDATING USING NEURAL NETWORKS , 1998 .

[20]  Martin A. Riedmiller,et al.  A direct adaptive method for faster backpropagation learning: the RPROP algorithm , 1993, IEEE International Conference on Neural Networks.

[21]  George C. Lee,et al.  A Structural Damage Neural Network Monitoring System , 1994 .

[22]  H. Luo,et al.  Dynamic Learning Rate Neural Network Training and Composite Structural Damage Detection , 1997 .

[23]  John E. Mottershead,et al.  REGULARISATION METHODS FOR FINITE ELEMENT MODEL UPDATING , 1998 .

[24]  François M. Hemez,et al.  Locating and identifying structural damage using a sensitivity-based model updating methodology , 1993 .