北京航空航天大学学报 ›› 2020, Vol. 46 ›› Issue (2): 331-339.doi: 10.13700/j.bh.1001-5965.2019.0240

• 论文 • 上一篇    下一篇

航空发动机传感器与执行机构信息重构算法

孙浩, 郭迎清, 赵万里   

  1. 西北工业大学 动力与能源学院, 西安 710129
  • 收稿日期:2019-05-18 发布日期:2020-03-11
  • 通讯作者: 郭迎清 E-mail:yqguo@nwpu.edu.cn
  • 作者简介:孙浩,男,博士研究生。主要研究方向:航空发动机故障诊断与容错控制;郭迎清,男,博士,教授,博士生导师。主要研究方向:航空发动机先进控制与健康管理技术。

Information reconstruction algorithm of aero-engine sensors and actuators

SUN Hao, GUO Yingqing, ZHAO Wanli   

  1. School of Power and Energy, Northwestern Polytechnical University, Xi'an 710129, China
  • Received:2019-05-18 Published:2020-03-11

摘要: 为实现航空发动机传感器与执行结构在故障情形下的故障幅值估计及信息重构,缓解故障对发动机性能的影响,在已有故障检测和故障隔离算法的基础上,提出一种基于修正的广义似然比(GLR)方法的信息重构算法。针对某型民用涡扇发动机的传感器与执行机构发生恒偏差与漂移故障的情形下进行了仿真验证。结果表明:基于修正的GLR方法对传感器和执行机构恒偏差和漂移故障的故障幅值估计具有较高的精度,两种故障情形下故障幅值的估计值的均方根误差均不超过0.005,故障部件信息重构后故障对系统性能的影响得到有效缓解。

关键词: 航空发动机, 故障诊断, 故障幅值估计, 广义似然比(GLR), 信息重构

Abstract: In order to realize the fault magnitude estimation and information reconstruction of aero-engine sensor and actuator in fault condition and to accommodate the influence of fault on engine performance, based on the fault detection and fault isolation algorithms, a reconstruction algorithm based on a modified generalized likelihood ratio (GLR) method is proposed. Aimed at the constant deviation fault and drift fault of sensors and actuators of a certain type of civil turbofan engine, a simulation experiment was implemented. The simulation results show that the modified GLR method has higher accuracy for fault magnitude estimation of sensors and actuators with constant deviation and drift fault. The root mean square error of the fault magnitude estimation does not exceed 0.005 in both fault types. And after the information reconstruction of fault component, the effect of the fault on the system performance is effectively accommodated.

Key words: aero-engine, fault diagnosis, fault magnitude estimation, generalized likelihood ratio (GLR), information reconstruction

中图分类号: 


版权所有 © 《北京航空航天大学学报》编辑部
通讯地址:北京市海淀区学院路37号 北京航空航天大学学报编辑部 邮编:100191 E-mail:jbuaa@buaa.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发