基于动态线性模型的民航发动机性能状态监控与在翼寿命预测  

A method of condition monitoring and on-wing life prediction for civil aviation aircraft engine based on dynamic linear model

作　　者：孙见忠[1] 左洪福[1] 刘鹏鹏[1] 朱磊[1] 袁风 

机构地区：[1]南京航空航天大学民航学院,南京210016 [2]东航工程技术公司,上海200335

出　　处：《系统工程理论与实践》2013年第12期3243-3250,共8页Systems Engineering-Theory & Practice

基　　金：国家自然科学基金与中国民航联合资助基金重点项目(60939003;61179058)

摘　　要：以民航发动机为对象.提出了一种基于动态线性模型的性能状态监控与在翼寿命预测方法.利用动态线性模型来描述性能参数偏差值序列.借助贝叶斯因子法来监测参数序列的异常.算例分析表明.该方法可操作性强.昔助统计方法能够在故障发生的早期及时预警;动态线性模型同时用来描述发动机的性能退化过程,能够考虑维修、水洗以及故障等使用因素对性能退化的影响.实例证明该方法更加符合航空公司实际运营情况,能够更加合理地预测下发时间.A method of condition monitoring and on-wing life prediction for civil aviation aircraft engine is proposed in this study. The dynamic linear model is used to describe the dynamics of the condition monitoring parameters and the Bayes factors are calculated to detect the abnormal of the parameter sequences. Case study on the simulation data set shows the feasibility of the proposed method for fault detection and a timely warning can be given once a fault occurs. The dynamic linear model can be utilized to describe the process of the engine performance deterioration, and further future evolution trend can be predicted within the Bayesian framework. The advantage of the proposed method is that the dynamic linear model can incorporate sudden changes in the performance introduced by the on-line maintenance, washing and fault. The case study on a real data set shows that the proposed method for performance prognostics can give a more reasonable prediction of the engine on-wing life since it takes the real operating conditions into account.

关 键 词：动态线性模型 贝叶斯状态估计与预测 贝叶斯因子 发动机状态监控 在翼寿命预测 

分 类 号：V37[航空宇航科学与技术—航空宇航推进理论与工程]