【摘要】 提出了一种基于短时间样本的故障诊断方法,通过频谱校正提高频谱精度。首先对原始信号进行小波降噪,提高信噪比;然后进行经验模态分解,获取信号的各阶本征模态函数;分别对各阶本征模态函数进行希尔伯特解调分析,获得包含系统故障特征信息的调制信号;接着采用校正算法对调制信号进行频谱校正,频谱变换后获得精确的频谱;最后根据校正结果进行系统故障判别。实践表明,此方法具有速度快、精度高的特点,适合于设备的在线快速诊断。更多还原

【Abstract】 This paper proposes a new fault diagnosis method based on short-time signal sample,which adopts spectrum correction algorithm to improve spectrum accuracy for the purpose of accurate diagnosis.Firstly,wavelet de-noising is applied to improve signal-noise ratio,and empirical mode decomposition(EMD) is used to obtain the intrinsic mode functions(IMF) of the signal,then Hilbert transform is utilized to demodulate the IMFs to get the modulation signal that contains system fault information.Spectrum correction algorithm is employed to get accurate frequencies and amplitudes of the signal spectrum.Finally,the fault of the tested system is evaluated based on the corrected spectrum.Practice proves that this method features quick response and high accuracy,and can be used in on-line real-time fault diagnosis.更多还原