【作者】 巩晓赟； 左长青； 韩捷； 雷文平；

【Author】 Gong Xiaoyun,Zuo Changqing,Han Jie,Lei Wenping

【机构】 郑州大学振动工程研究所；

【摘要】 针对旋转机械故障诊断中非线性振动信息分析的不全面性,提出了一种小波包一矢双谱分析方法。首先对同源双通道信息分别采用小波包进行适当层次的分解,实现对高频部分去噪,减少大量的随机噪声以提高信噪比,并对提取的信号进行重构。然后结合全矢谱技术进行双谱分析,以克服单通道双谱分析方法的不足,在抑制高斯噪声的同时获得比能量谱更丰富的信息。利用全矢谱技术进行融合的小波包一矢双谱分析方法,不仅继承了双谱分析方法的优势,而且更加全面地反映出了信号的真实性。最后通过实例对其算法的可行性进行了验证,也进一步表明了该方法在故障诊断中的有效性。更多还原

【Abstract】 According to limitation of nonlinear vibration signal analysis in fault diagnosis of rotating machinery,a new method for combining wavelet package with vector - bispectrum is proposed. Firstly,vibration signals from two different directions are respectively decomposed by wavelet packet transform,and recomposed the abstracted signals.Wavelet packet de-noising can reduce large random noise and increase signal to noise ratio.Then,to overcome the faults of traditional fault diagnosis method based on single source data,vector - bispectrum technology is selected. Compared to traditional method,the new approach is more effective than others because its inheritance and all - sided feature.Finally,data fusion of example vibration signals is considered as evidences for the feasibility of the new technique.Furthermore,experimental results are also given to verify the validity of the proposed methods in the field of diagnosing fault.更多还原