基于SVD和熵优化频带熵的滚动轴承故障诊断研究

摘要

针对在奇异值分解(Singular Value Decomposition,SVD)中,随机噪声对各阶的贡献几乎相等,导致单一SVD降噪效果不理想的问题,提出了基于SVD和频带熵(Frequency Band Entropy,FBE)相结合的轴承故障特征提取方法.针对基于FBE的带通滤波器的阶数和带宽需经验确定的问题,提出了基于信息熵最小值原则的参数优化方法.首先,对原始振动信号在相空间重构Hankel矩阵并利用SVD进行降噪处理,采用奇异值相对变化率来确定模型的阶次;然后,对降噪后的信号进行基于FBE的带通滤波,并采用基于信息熵最小值原则的优化方法确定带通滤波器的阶数和带宽.最后,对滤波信号进行包络谱分析,提取轴承故障特征频率,并用峭度指标证明了带通滤波器的有效性.通过数值仿真和实际轴承故障数据分析,证明了该方法提取轴承故障特征频率的有效性.%According to the problem that in singular value decomposition (SVD),the contributions of random noise to each order are almost equal,which results in the unsatisfactory effect of noise reduction using SVD alone,a fault feature extraction method based on SVD and frequency band entropy (FBE) is proposed.Aiming at the order and bandwidth of the FBE-based band-pass filter which need to be determined by experience,a novel method of parameters optimization based on the principle of information entropy minimum is proposed.Firstly,the Hankel matrix is reconstructed from the original vibration signal in the phase space and the SVD method is used to reduce the noise.The singular value relative change rate is used to determine the order of the model.Then,FBE-based band-pass filtering is performed on the noise-reduced signal,and an optimization method based on the principle of information entropy minimum is used to determine the order and bandwidth of band-pass filter.Finally,the filtered signal is subjected to envelope analysis to extract the characteristic frequency of the bearing fault,and the effectiveness of the band-pass filter is proved by the kurtosis index.Through the numerical simulation and the analysis of the actual bearing fault data,the effectiveness of the method to extract the characteristic frequency of the bearing fault is validated.