涡轴发动机分布式控制系统传感器故障诊断技术

Sensor Fault Diagnosis Technology for Distributed Control System of Turboshaft Engine

关键词:  分布式控制系统  故障诊断  总线通信时延  等价空间法  Kalman滤波器组

Key words:  Distributed control system  Fault diagnosis  Bus communication delay  Parity space method  Bank of Kalman filters

摘要:

为满足航空发动机分布式控制系统的故障诊断需求，重点解决分布式控制系统中总线通信时延对传统故障诊断措施效果的影响，提出了一种新的航空发动机分布式控制系统故障诊断方法。针对航空涡轴发动机，搭建了包含随机时延的分布式控制系统数学模型，利用等价空间法（PSM）设计故障诊断滤波器组。并将改进的故障诊断方法与传统的基于Kalman滤波器组的方法进行对比仿真验证。结果表明，基于PSM算法所设计的故障诊断滤波器组，在对包含随机时延的涡轴发动机分布式控制系统进行传感器故障诊断中，能有效降低时延所引起的故障指示信号跳变幅度。证明此方法能够减少通信时延导致故障误判的情况，提高发动机分布式控制系统故障诊断的准确率。

Abstract:

In order to meet the needs of fault diagnosis of aero-engine distributed control system， a new fault diagnosis method was proposed， which focused on the influence of bus communication delay to the effect of traditional fault diagnosis measures. Aiming at the aero turboshaft engine， a mathematical model of distributed control system with random delay was established， and a bank of filters optimized by Parity Space Method（PSM） was used to detect fault. The improved fault diagnosis method was compared with the traditional method based on a bank of Kalman filters. The simulation results prove that the bank of proposed fault diagnosis filters based on PSM algorithm could effectively reduce the jump amplitude of fault indication signal caused by delay in sensor fault diagnosis of turboshaft engine distributed control system with random delay. Consequently， this method can reduce the misjudgment caused by communication delay and improve the accuracy of fault diagnosis of engine distributed control system.