基于图像处理的航空发动机自适应频谱报警技术研究

Research on Aero-engine Adaptive Spectrum Fault Alarming Technology Based on Image Processing

在飞行试验中,作为飞行器直接的动力来源,航空发动机的状态监测和故障报警技术是保障飞行试验安全的重要手段,而传统的报警技术存在误报警,准确率不高等问题;针对传统频域报警方法不能包容工况变化,局部特征与全局特征无法兼顾的问题,研究了基于图像处理的航空发动机自适应频谱报警方法,包括包络生成方法和频段自动划分方法,将两者应用于均化频谱的报警中,并通过对比频谱幅值及频段能量值实现故障报警;利用某型航空发动机故障数据对方法的有效性进行了验证,结果表明,该方法能够在发动机不同工况下有效划分频谱高能量和低能量区,兼顾了对振动信号局部特征与全局特征,提高了故障报警的自动化水平和准确率。

In the flight test, as a direct power source of the aircraft, the condition monitoring and fault alarming technology of the aero-engine is an important method to ensure the safety of the flight test. However, the traditional alarm technology has the problem of false alarm and low accuracy. The traditional frequency domain alarm method cant accommodate the change of working conditions, and the problem that the local characteristics and global characteristics cant be taken into account. Therefore the improved envelope alarm method and frequency band alarm method are studied and the two are applied to the homogenized spectrum to get fault alarming by comparing spectral amplitude and band energy value. The effectiveness of the method is validated by using aero-engine fault data. The results show that this method can effectively divide the high-energy and low-energy regions of the spectrum under different operating conditions of the engine, taking into account the local and global characteristics of the vibration signal, and improved Failure alarm automation level and accuracy.