Reliability calculation of concrete filled steel tubular arch bridge based on neural network
Abstract：Since the performance function existing in the reliability analysis of large complex structures can not be expressed explicitly,the reliability of long-span concrete filled steel tubular arch bridge structure was studied with the response surface method based on neural network.The experimental sample points were determined with uniform design method,and the deterministic finite element analysis was carried out.The neural network training was conducted with the obtained training samples,and the structure and parameters of neural network were determined.In addition,in combination with the conventional methods,the reliability analysis was performed on the neural network response surface.The calculated results show that the neural network with introducing the uniform design method can well simulate the real response of complex structures.The present method is feasible in the reliability analysis of long-span concrete filled steel tubular arch bridge structure,and has higher efficiency and accuracy.
Keywords：neural networkconcrete filled steel tubulararch bridgearch ribreliabilityuniform designresponse surface methodseries system
Publication Date：2011-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)