Journal of Propulsion Technology ›› 2017, Vol. 38 ›› Issue (1): 191-198.

Previous Articles     Next Articles

Aero-Engine Gas Path Fault Diagnosis Based on Broyden Algorithm

  

  1. Nanjing University of Aeronautics and Astronautics,Jiangsu Province Key Laboratory of Aerospace Power System, Nanjing 210016,China,Nanjing University of Aeronautics and Astronautics,Jiangsu Province Key Laboratory of Aerospace Power System, Nanjing 210016,China and Nanjing University of Aeronautics and Astronautics,Jiangsu Province Key Laboratory of Aerospace Power System, Nanjing 210016,China
  • Published:2021-08-15

基于Broyden算法的航空发动机气路故障诊断

潘 阳,李秋红,王 元   

  1. 南京航空航天大学 能源与动力学院,江苏省航空动力系统重点实验室,江苏 南京 210016,南京航空航天大学 能源与动力学院,江苏省航空动力系统重点实验室,江苏 南京 210016,南京航空航天大学 能源与动力学院,江苏省航空动力系统重点实验室,江苏 南京 210016
  • 作者简介:潘 阳,男,硕士生,研究领域为系统仿真与控制。

Abstract: To solve the problems of slow response speed and poor accuracy in multi-fault diagnosis and off-design point fault diagnosis which arise in aero-engine gas path fault diagnosis based on Kalman filter,an equation-solving fault diagnosis method using the improved Broyden algorithm was proposed. For turbo-shaft engine,the gas path fault diagnosis equations were built. They include 3 equations based on the principle that the model outputs should consist with the engine outputs and 2 balance equations from engine model. The improved Broyden algorithm was used to solve the equations to get the component performance degradation factor and the model guess value. Simulation results show that the max steady-state fault diagnosis error based on the proposed method is less than 0.35% in the envelop for both single fault diagnosis and multi-fault diagnosis,and the maximum operation time of single step is less than 2ms. The result is much better than that of Kalman filter,which shows the advantage of the proposed method.

Key words: Aero-engine;Gas path fault diagnosis;Broyden algorithm;Kalman filter;Turbo-shaft engine

摘要: 针对基于Kalman的故障诊断算法响应速度慢、多故障诊断及非设计点诊断精度低的问题,提出一种基于改进Broyden算法求解方程组的航空发动机气路故障诊断方法。针对涡轴发动机,以模型输出跟踪发动机输出为准则确定3个方程,结合发动机模型中的2个平衡方程,构建气路故障诊断方程组,通过改进Broyden算法求解方程组以获得部件性能退化因子及模型猜值。数字仿真结果表明,所提出的基于Broyden算法求解方程组的航空发动机气路故障诊断方法,在包线内的单故障和多故障诊断稳态误差均小于0.35%,且诊断过程算法单步运行最大耗时小于2ms,具有良好的实时性,远优于Kalman滤波方法,验证了算法的先进性。

关键词: 航空发动机;气路故障诊断;Broyden算法;Kalman滤波器;涡轴发动机