Static and dynamic novelty detection methods for jet engine health monitoring
Static and dynamic novelty detection methods for jet engine health monitoring
复制标题
DOI：
10.1098/rsta.2006.1931
复制
发表时间：
2007-02-15
期刊：
影响因子：
5
通讯作者：
Tarassenko, Lionel
中科院分区：
文献类型：
--
作者：
Hayton, Paul;Utete, Simukai;Tarassenko, Lionel
新奇检测需要从已知为正态的训练数据学习正态模型。在本文中考虑的第一个模型是一个静态模型训练，以检测新的事件与从喷气发动机记录的振动谱的变化。我们描述了如何跨旋转轴的谐波的能量分布可以学习的支持向量机模型的正常性。第二个模型是使用基于期望最大化的方法从数据中部分学习的动态模型。该模型使用卡尔曼滤波器来融合性能数据，以表征正常的发动机行为。使用来自卡尔曼滤波器的归一化新息平方来检测与正常操作的偏差。
Novelty detection requires models of normality to be learnt from training data known to be normal. The first model considered in this paper is a static model trained to detect novel events associated with changes in the vibration spectra recorded from a jet engine. We describe how the distribution of energy across the harmonics of a rotating shaft can be learnt by a support vector machine model of normality. The second model is a dynamic model partially learnt from data using an expectation maximization-based method. This model uses a Kalman filter to fuse performance data in order to characterize normal engine behaviour. Deviations from normal operation are detected using the normalized innovations squared from the Kalman filter.