Evaluation for Parachute Reliability Based on Fiducial Inference and Bayesian Network
Reliability is an important indicator for evaluating the performance of a parachute system.Verification of reliability often requires a large number of successful systemlevel airdrop tests,which is often difficult to accept in practice.This paper proposes a reliability analysis method based on Fiducial inference and Bayesian network.The method firstly decomposes the parachute into each group according to the system composition,and uses the layered BN to represent; secondly,based on the Fiducial inference method,the mean value and the lower confidence limit of the strength test data of each group are calculated.And apply the stress-strength model to calculate the probability that the intensity is greater than the stress,and as the reliability of each component; third,considering the conduction relationship between the nodes of each layer,respectively calculate the node conduction probability.The joint probability of the node with conduction relationship is calculated by the fusion node probability based Bayesian.Finally,based on the prior probability of the node,the posterior probability of each node is calculated and the parachute groups(parts)system are obtained.The analysis of the example shows that the calculation process of the method can fuse the reliability information of each layer of the system,which improve the accuracy of the system reliability.It can be used for the reliability analysis of the parachute system in the design stage,also can provide reference for inspection and acceptance of parachute system.
parachute reliability Bayesian network fiducial inference stress-strength model
Zhanfeng Wang Jianbo Wang Pengfei Meng Fuyong Wang Wenbao Dai
Airborne/Airdrop Technical Support Battalion,95829 PLA Troops,Hubei,China,432011
国际会议
上海
英文
124-127
2018-12-22（万方平台首次上网日期，不代表论文的发表时间）