Structural reliability analyis of elastic-plastic structures using neural networks and Monte Carlo simulation

Abstract This paper examines the application of Neural Networks (NN) to the reliability analysis of complex structural systems in connection with Monte Carlo Simulation (MCS). The failure of the system is associated with the plastic collapse. The use of NN was motivated by the approximate concepts inherent in reliability analysis and the time consuming repeated analyses required for MCS. A Back Propagation algorithm is implemented for training the NN utilising available information generated from selected elasto-plastic analyses. The trained NN is then used to compute the critical load factor due to different sets of basic random variables leading to close prediction of the probability of failure. The use of MCS with Importance Sampling further improves the prediction of the probability of failure with Neural Networks.

[1]  M. Papadrakakis,et al.  A computationally efficient method for the limit elasto plastic analysis of space frames , 1995 .

[2]  Pericles S. Theocaris,et al.  Neural networks for computing in fracture mechanics. Methods and prospects of applications , 1993 .

[3]  Yoshio Hirose,et al.  Backpropagation algorithm which varies the number of hidden units , 1989, International 1989 Joint Conference on Neural Networks.

[4]  L. Berke,et al.  Optimum Design of Aerospace Structural Components Using Neural Networks , 1993 .

[5]  John F. Abel,et al.  Yield surface applications in nonlinear steel frame analysis , 1982 .

[6]  Robert E. Melchers,et al.  Structural Reliability: Analysis and Prediction , 1987 .

[7]  Masanobu Shinozuka,et al.  Weighted Integral Method. II: Response Variability and Reliability , 1991 .

[8]  Manolis Papadrakakis,et al.  A simple and efficient solution method for the limit elasto-plastic analysis of plane frames , 1991 .

[9]  Prabhat Hajela,et al.  APPLICATIONS OF ARTIFICIAL NEURAL NETS IN STRUCTURAL MECHANICS , 1992 .

[10]  J. E. Pulido,et al.  Structural reliability using Monte-Carlo simulation with variance reduction techniques on elastic-plastic structures , 1992 .

[11]  Y. Ueda,et al.  The plastic node method: A new method of plastic analysis , 1982 .

[12]  Khan Ai. Topping Bhv. and Bahreininejad A..  Parallel training of neural networks for finite elementmesh generation , 1993 .

[13]  C. Allin Cornell,et al.  Adaptive hybrid conditional expectation approaches for reliability estimation , 1991 .

[14]  Prabhat Hajela,et al.  Minimizing distortion in truss structures—A Hopfield network solution , 1993 .

[15]  J. Hammersley.  SIMULATION AND THE MONTE CARLO METHOD , 1982 .

[16]  Ross B. Corotis,et al.  Reliability of Nonlinear Framed Structures , 1983 .

[17]  A.H-S. Ang,et al.  Advances in structural reliability , 1992 .

[18]  Y. Fujimoto,et al.  Fitting-Adaptive Importance Sampling in Reliability Analysis , 1991 .

[19]  Reuven Y. Rubinstein,et al.  Simulation and the Monte Carlo Method , 1981 .

[20]  M. Shinozuka.  Basic Analysis of Structural Safety , 1983 .

[21]  C. Bucher.  Adaptive sampling — an iterative fast Monte Carlo procedure , 1988 .

[22]  Yoh-Han Pao,et al.  Adaptive pattern recognition and neural networks , 1989 .