摘要为了从复杂的轴承振动信号中提取微弱的故障信息,将相关峭度引入滚动轴承故障诊断领域,结合奇异值分解(Singular Value Decomposition,SVD)和相关峭度,提出了一种新的滚动轴承故障特征提取方法。该方法首先利用SVD对轴承振动信号进行分解,然后根据相关峭度选取SVD分解后的分量,提取出滚动轴承的弱故障信号。通过对轴承内验证了该方法的有效性。
In order to extract the faint fault information from complicated vibration signal of bearing, the correlated kurtosis is introduced into the field of rolling bearing fault diagnosis. Combined with SVD and correlated kurtosis, a feature extraction method is proposed. According to the method, by SVD processing a group of component signals are obtained, then the component signals with equal correlated kurtosis are selected to be added together, and the weak fault signal is clearly extracted. The effectiveness of the method is demonstrated on both simulated signal and actual data.
张永祥;王孝霖;张 帅;朱杰平. 基于奇异值分解和相关峭度的滚动轴承故障诊断方法研究[J]. , 2014, 33(11): 167-171.
ZHANG Yong-xiang;WANG Xiao-lin;ZHANG Shuai;ZHU Jie-ping. Research on rolling element bearing fault diagnosis based on singular value decomposition and correlated kurtosis. , 2014, 33(11): 167-171.