基于包络解调的离心式压缩机弱故障信号增强方法研究

在离心式压缩机使用要求不断提高下,为了增强故障诊断精确性,提出基于包络解调的非平稳工况下离心式压缩机弱故障信号增强方法。将小波包分析和独立分量分析结合,通过小波包分析法对含有噪声的混合信号进行降噪,根据 FastICA 算法分离降噪后的混合信号,对分离出的信号采用收缩函数实行频段内的去噪操作,完成多源故障信号分离去噪。在故障信号分离的基础上,考虑到被分离出的信号伴随着微弱噪声,进一步通过包络解调随机共振实现弱故障信号增强。对多源信号分离结果进行包络解调操作,并对包络信号实行变尺度随机共振输出处理,实现故障特征信号增强,达到故障诊断的目的。通过实验分别对此方法的信号去噪增强效果和故障诊断精确性进行验证,实验结果表明,该方法不仅弱故障信号增强效果显著,且故障诊断鲁棒性强,精度高,具有可实践性。

Research on Weak Fault Signal Enhancement Method of Centrifugal Compressor Based on Envelope Demodulation

With the increasing requirements of the use of centrifugal compressors , in order to enhance the accuracy of fault diagnosis , a weak fault signal enhancement method for centrifugal compressors under non-stationary operating conditions based on envelope demodulation is proposed.Combining wavelet packet analysis and independent component analysis , the mixed signal containing noise is denoised by wavelet packet analysis , according to FastICA algorithm , the mixed signals after denoising are separated , and the contracted function is used for the separated signal de-noise operation in the frequency band which is implemented to complete multi-source fault signal separation and de-noise.On the basis of fault signal separation , considering that the separated signal is accompanied by weak noise , the weak fault signal is enhanced by envelope demodulation stochastic resonance.The results of multi-source signal separation are demodulated by envelope , and the envelope signal is output by variable scale stochastic resonance to enhance the fault characteristic signal and achieve the purpose of fault diagnosis.The experimental method is used to verify the signal denoising enhancement effect and the accuracy of fault diagnosis.The experimental results show that the proposed method not only enhances the weak fault signal significantly , but also has strong robustness , high accuracy , and practicality.