支持向量机在航空发动机故障诊断中的应用  

Applications of Support Vector Machines to Aeroengine Fault Diagnosis

作　　者：尉询楷[1] 陆波[1] 汪诚[1] 路建明[1] 李应红[1] 

机构地区：[1]空军工程大学工程学院,陕西西安710038

出　　处：《航空动力学报》2004年第6期844-848,共5页Journal of Aerospace Power

摘　　要：提出了一种基于支持向量机的航空发动机故障诊断方法。该方法在利用恰当的工具或方法提取到发动机故障的特征信息之后,采用支持向量机对特征信息与故障模式进行关联,实现故障模式的分类。该方法专门针对小样本集合设计,能够在小样本情况下获得较大的推广能力,将其应用于某型发动机的故障诊断中,取得了较好的效果。该方法尤其对于新型发动机的故障诊断具有一定的参考价值。An aeroengine fault diagnosis method based on support vector machines is presented. After extracting fault symptoms via proper equipment or methods, support vector machines were adopted to set up correlation between fault symptoms and fault patterns and to realize the classification of faults. Support vector machines are specially designed for good generalization ability under the condition that only small samples set is available. Applying this method to a certain engine, the diagnostic results are satisfactory, and it is referable to develop diagnostic system for newly-developed engines.

关 键 词：航空、航天推进系统 航空发动机 故障诊断 特征提取 支持向量机 

分 类 号：V235.13[航空宇航科学与技术—航空宇航推进理论与工程] TP18[自动化与计算机技术—控制理论与控制工程]