The Application of Artificial BP Neural Networks and Monte-Carlo Method for the Reliability Analysis on Frame Structure

Abstract:

In order to conduct the reliability analysis of frame structure, the limit state function was first fitted by artificial BP neural network. Then considering the orthogonal array method, sample data was arranged. After that an improved network modes was trained for the probabilistic analysis on a wide range data with the Monte-Carlo method. The mean and standard deviation for the limit state function was easily obtained and the reliability index on the structure can be also calculated. Finally, the example indicated that this method used in the reliability analysis for frame structure was feasible.

• Periodical: Applied Mechanics and Materials (Volumes 204-208)
• Pages: 3256-3259
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.204-208.3256
• Online since: October 2012
• Authors: Zhi Cheng Xue, Hai Jun Wang
• Keywords: BP Artificial Neural Network, Frame Structure, Monte-Carlo Method (MC-Method), Reliability

References

[1] Cybenko G: Approximation by Superpositions of a Sigmoidalfunction, Math Control Signal Systems Vol.2(1989).

[2] Weicai Liu, Wei Zhang: The Application of Approximate Method on Adaptive Neural Network, Engineering Mechanics Vol.14(1997),pp.91-94.

[3] Guofan Zhao,Structure Reliability Theory and its Application of Engineering(Dalian University of Technology Press,China,1996).