基于小波分析的压气机失速故障检测

Detection of compressor stall based on wavelet analysis

当发动机发生失速时 ,会引起压气机内压力信号的变化 ,这种压力信号的变化可以看作信号的奇异点。利用小波分析所具有的时 频特性和多分辨分析的特点 ,结合对压气机内压力信号频率成分的研究 ,通过检测模极大值的方法来确定信号的奇异点 ,进而探测压气机的失速故障。结果表明 。

Stall in engine compressor will result in variation of pressure in its compressors, which is regarded as a singularity of signal. Considering the good performance of wavelet analysis in time and frequency domain, a method based on multi-scale wavelet transform was proposed to detect the compressor stall, which is based on the fact that the wavelet transform of a signal will take its modulus maximum at its singular point in transform domain. As a result, the engine stall is detected precisely through the method developed in this paper.