【摘要】 针对飞机整体驱动发电机可靠性分析手段缺乏、故障样本小且维修策略相对保守的问题，提出了以历史故障数据为驱动的贝叶斯和马尔可夫链融合的可靠性分析方法。首先，利用蒙特卡洛方法对故障数据预处理以增大样本空间，并通过最大信息熵法求解飞机整体驱动发电机的先验分布；其次，利用马尔可夫链方法对复杂后验分布进行解算；最后，以可靠性为中心给出飞机IDG的维修建议。经过数值仿真软件计算后得IDG的累计失效函数参数估计误差分别为0.121 3和0.001 3,误差较小。仿真结果表明，提出的可靠性分析方法适用于小样本空间的飞机IDG可靠性分析，并根据结果给出了维修建议。更多还原

【Abstract】 In view of the lack of reliability analysis methods, small fault samples and relatively conservative maintenance strategies of aircraft integral drive generators, a reliability analysis method based on Bayesian and Markov chain fusion driven by historical fault data was proposed. Firstly, the Monte Carlo method was used to preprocess the fault data to increase the sample space, and the prior distribution of the aircraft overall drive generator was solved by the maximum information entropy method. Secondly, the Markov chain method was used to solve the complex posterior distribution. Finally, the maintenance suggestions of the aircraft IDG(integral drive generator) were given based on the reliability. The cumulative failure function parameter estimation errors of IDG calculated by numerical simulation software are 0.121 3 and 0.001 3 respectively, with small errors. The simulation results show that the proposed reliability analysis method is suitable for aircraft IDG reliability analysis in small sample space, and maintenance suggestions were given according to the results.更多还原