CCF structural reliability estimation under statistical uncertainty

For most structures, dependent failure is a common feature. It will lead to considerable errors or even misleading conclusion by neglecting failure dependence and assuming independent failure when estimating the structural reliability. In this paper, the reliability models of CCF structures caused by the randomness of load are developed and the estimation methods in consideration of both physical uncertainty and statistical uncertainty are proposed. Firstly, according to the LSI model and conditional probability method, reliability models of CCF component, CCF series system, CCF parallel system and CCF k/n system are derived. Secondly, the reliability point estimates of CCF structures are presented according to the maximum likelihood estimation of unknown parameters. Finally, in consideration of the statistical uncertainty of load and strength, a Monte-Carlo method is presented. The reliability samples of units from their reliability confidence distribution and reliability estimation function of CCF system are combined in the method to simulate CCF structural operation and obtain the interval estimate of structural reliability. This method has little limitation to the size and complexity of systems.