基于小波分析的高速牵引电机轴承故障诊断研究

发布时间：2018-01-02 12:26

本文关键词：基于小波分析的高速牵引电机轴承故障诊断研究　出处：《北京交通大学》2011年硕士论文　论文类型：学位论文

【摘要】：近年来,高速列车的出现对高速牵引电机轴承的工作状态提出了严格的要求。国外机械设备状态监测和故障诊断技术已经进入实用化阶段。我国故障诊断技术也有20多年的发展,不论在故障诊断理论和方法上,还是在工程实践及监测诊断产品的研发中,都需要加速发展。本文采用小波的希尔伯特变换、小波包变换、EMD经验模态分解方法对高速牵引电机轴承故障诊断的时频分析及故障识别方法进行了深入研究,并采用Lab VIEW与Matlab相结合的方法开发了诊断平台。 (1)小波和Hilbert变换相结合的高速牵引电机滚动轴承故障诊断方法的研究。首先利用Daubechies小波对故障信号进行小波的分解,然后提取出包含故障特征信息的分解层。对选取的小波层进行快速傅里叶变换并提取出特征频率。与理论上的特征频率进行对比,有效判定轴向线性裂纹故障类型。 (2)小波包方法在高速牵引电机滚动轴承故障诊断中的应用研究。提出了在小波包分解后的节点进行以能量为依据的最佳节点方法。并采取均值与方差和为阈值的小波包去噪方式对信号去噪和重组。选取小波包进行快速傅里叶变换,提取出特征故障频率,能简便有效地判定轴向线性裂纹故障类型。 (3)将小波包和EMD(Empirical Mode Decomposition)分解二者有机结合,探索研究其在高速牵引电机轴承故障诊断中的应用。将小波包、EMD分解应用到高速牵引电机轴承故障诊断中。在诊断前,首先进行小波包去噪,接着进行EMD分解,然后选择与原始信号相关系数最大的一层再次进行的小波包分解,并选取合适的小波包进行频谱分析,最后进行故障状态识别。该方法虽能诊断出轴向线性裂纹故障特征,但达到分解平衡耗时较长,不易与本文程序结合。 (4)试验台组合以及实验研究。利用实验台、数据采集设备,并根据小波分析理论进行实验验证。分别采用小波的Hilbert变换、小波包分解和EMD经验分解的方法对轴向线性裂纹故障轴承诊断进行实用分析,总结出本文最有效的诊断方法—小波包分解。 (5)基于Lab VIEW和Matlab编程的轴承诊断软件平台开发。结合Matlab良好的数据处理功能,应用yulewalk多通带滤波器对频域信号进行滤波,采用虚拟仪器技术设计开发出故障诊断软件平台,最终诊断出故障特征。
[Abstract]:In recent years , the occurrence of high - speed train has put forward strict requirements for the working state of high - speed traction motor bearings . The state monitoring and fault diagnosis technology of mechanical equipment abroad has entered the practical stage . The fault diagnosis technology in China has been developed in more than 20 years . In this paper , the time - frequency analysis and fault identification method of high - speed traction motor bearing fault diagnosis are studied in detail in the theory and method of fault diagnosis . ( 1 ) The research of fault diagnosis method of high speed traction motor rolling bearing combining wavelet and Hilbert transform . Firstly , Daubechies wavelet transform is used to decompose the fault signal , then the decomposition layer containing fault feature information is extracted . The selected wavelet layer is fast Fourier transformed and the characteristic frequency is extracted . Compared with the theoretical characteristic frequency , the fault type of axial linear crack is effectively determined . ( 2 ) The application of wavelet packet method in fault diagnosis of high speed traction motor rolling bearing . In this paper , the optimal node method based on energy is put forward , which is based on energy . The wavelet packet denoising method is used to denoise and recombine the signal . The wavelet packet is selected for fast Fourier transform , and the characteristic fault frequency is extracted . It can be used to determine the fault type of axial linear crack simply and effectively . ( 3 ) Combining the decomposition of the wavelet packet with EMD ( Empirical Mode Decomposition ) , this paper explores its application in fault diagnosis of high - speed traction motor bearing . The wavelet packet and EMD are applied to the fault diagnosis of high - speed traction motor bearings . Before the diagnosis , wavelet packet de - noising is firstly carried out , then EMD is decomposed , then the wavelet packet with the largest correlation coefficient of the original signal is decomposed , and the proper wavelet packet is selected for frequency spectrum analysis , and finally the fault state identification is carried out . ( 4 ) Test bench combination and experimental study . This paper makes a practical analysis of the diagnosis of axial linear crack by means of Hilbert transform , wavelet packet decomposition and EMD ' s empirical decomposition , and summarizes the most effective diagnosis method - wavelet packet decomposition . ( 5 ) Development of bearing diagnosis software platform based on Lab VIEW and Matlab programming . Combined with the good data processing function of Matlab , the frequency domain signal is filtered using yulewalk multi - pass band filter , and the fault diagnosis software platform is developed by using the virtual instrument technology design , and finally the fault feature is diagnosed .

【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH133.33;TH165.3

【引证文献】