中国民航大学学报

• 民用航空 • 上一篇    下一篇

一种改进的灰色故障识别方法及其鲁棒性分析

王玉臣   

  1. (中国民航大学工程技术训练中心,天津300300)
  • 收稿日期:2014-12-15 修回日期:2015-01-05 出版日期:2015-08-22 发布日期:2015-09-24
  • 作者简介:王玉臣(1986—),男,河北沧州人,助教,硕士,研究方向为航空发动机持续适航以及航空排放.
  • 基金资助:

    中央高校基本科研业务费专项(3122014D010)

Improved grey fault recognition method and its robustness analysis

WANG Yu-chen   

  1. (Engineering Techniques Training Center,CAUC,Tianjin 300300,China)
  • Received:2014-12-15 Revised:2015-01-05 Online:2015-08-22 Published:2015-09-24

摘要:

灰色关联分析方法在机械设备故障诊断中广泛应用,以航空发动机的气路系统为研究对象,对典型的气路性能故障进行研究,提出一种改进的灰色关联分析方法,以PW4164/68 发动机的典型涡轮故障为例验证方法的有效性;考虑航空发动机性能参数测量中的影响因素,对灰色关联分析方法的鲁棒性进行分析,研究表明:在一定的测量偏差范围内,灰色关联分析方法具有较高的鲁棒性;改进后的灰色关联分析方法在性能参数测量的相对偏差为0.24 时还能准确地对发动机故障进行识别,因此改进措施有效地提高了发动机故障灰色识别方法的鲁棒性。

关键词: 航空发动机, 故障诊断, 灰色关联分析, 鲁棒性

Abstract:

Grey incidence analysis method is widely used in fault diagnosis of mechanical equipments. Taking aero-engine gas path system as the object,the typical faults of gas path performance are studied. An improved grey incidence analysis method is proposed, which is proved to be effective by analyzing typical turbine faults of PW4164/68 engine. Considering the influential factors of performance parameters measurement of aero-engine, robustness of the grey incidence analysis method is analyzed. The study shows that the grey incidence analysis method is of high robustness in a certain measuring deviation. The improved grey incidence analysis method can accurately identify the engine fault when relative measuring deviation of performance parameters is equal to 0.24,thus the improving measures effectively increase the robustness of grey engine fault recognition method.

Key words: aero-engine, fault diagnosis, grey incidence analysis, robustness

中图分类号: