基于贝叶斯模型平均的航空发动机可靠性分析

针对不同失效模式的特点，利用航空发动机状态监测信息，分别采用Gamma过程、Wiener过程和Weibull分布建立可靠性分析模型.采用贝叶斯模型平均（BMA）方法，分析不同失效模式对可靠性影响，对航空发动机可靠性进行集成分析.通过算例跟踪某在翼航空发动机可靠性水平，基于贝叶斯模型平均的航空发动机可靠性模型的计算误差率低于单一可靠性分析模型，计算结果误差率低于1%.说明多模型集成可靠性技术在复杂系统运行可靠性分析中的有效性和准确性. 

Reliability analysis on aero-engine using Bayesian model averaging

Considering the characteristics of different failure modes with use of monitoring information, Gamma process, Wiener process and Weibull distribution were selected to build the reliability analysis models on aero-engine. Bayesian model averaging (BMA) was used to integrate different models into one framework for reliability analysis on aero-engines, given its function for dynamically evaluating different failure mode effects. The reliability of some onwing aero-engine was analyzed by numerical example, showing that the reliability analysis model using BMA had less error than single reliability analysis model, and computation error rate of BMA was less than 1%. The results demonstrate the effectiveness of the reliability analysis of multi-model integration.