Reliability Analysis Of Composite Stiffened Structure With Delaminating Using Bootstrap Based Maximum Entropy Method

Composite material is used in a wide engineering field,which can improve the structural capacity and reduce the weight as well.At present,composite material is applied to the main bearing members of the aircraft structure,which has become one of the key technologies affecting the performance of the aircraft.Delaminating is the major damage form of the composite laminate during the process of manufacturing and service,and the bearing capacity analysis of the composite structure with delaminating is very important to the aircraft design.This paper is focused on reliability analysis of composite stiffened structure with delaminating using Bootstrap based maximum entropy methodFirstly,the research background and significance of this paper are introduced.The analytical methods of composite structure with delaminating in recent years are investigated,and the status of the reliability analysis of composite structure is summarized.For some problems with small samples,the basic idea and technical process of Bootstrap resample technology are introduced.Next,the analysis model of the composite stiffened plate with a elliptical delaminating is set up by ABAQUS.The buckling eigenvalue is obtained by the linear buckling analysis,which provides the initial deflection in post-buckling analysis.Hashin-Ye strength criterion and Tan,Gu rules are used in the progressive damage analysis to get the load-displacement curve under uniaxial compression loading.The effects of strength failure and delaminating extension on the structural ultimate bearing capacity are compared.In fact,some parameters of the composite structure are stochastic,such as material property,geometry size and loading,which cause a deviation in structural design and strength evaluation,so reliability analysis is needed.Considering the randomness of material property parameters,the failure probability of the ultimate bearing capacity is obtained by using Improved Dimension Reduction Method based Maximum Entropy Method,and the results are compared with those by Monte Carlo simulation.Finally,considering that statistical moments are calculated based on the small samples,causing the randomness of the results,Bootstrap resampling technology is used to solve the problem of small samples by producing a new augmented sample.The maximum entropy method is performed to obtain the probability density function and failure probability,with the mean value of the statistical moments calculated by the new the Bootstrap samples.The effects to the probability density function and failure probability of different original samples and sampling number are researched.A linear combination sampling method is proposed which can significantly improve the precision of the calculated results.The Bootstrap technology based MEM is applied in the reliability analysis of the composite stiffened structure with delaminating.