基于谱峭度的齿轮箱故障特征提取

发布时间：2018-02-27 19:06

本文关键词： 齿轮箱谱峭度包络分析阶比跟踪特征提取　出处：《昆明理工大学》2011年硕士论文　论文类型：学位论文

【摘要】：在机械设备中,齿轮箱作为改变转速和传递动力的通用零部件,在各行各业中得到了广泛运用。齿轮箱工作环境非常恶劣,齿轮箱部件容易受到损坏而出现故障。因此以齿轮箱作为机械设备状态监测与故障诊断的研究对象具有较好的现实意义。基于谱峭度的信号处理方法是近年来所提出来的故障特征提取新途径,目前在故障诊断领域正被广泛研究。但其在实际应用中还存在各种急需解决的问题,如：齿轮箱故障信号常常湮没于强背景噪声中；基于谱峭度的包络分析方法虽然可以自适应的确定包络参数,但计算量大；如何将此方法应用到齿轮箱升降速过程的故障诊断中等。为了解决以上问题,本论文利用AR模型对齿轮局部故障信号进行预白化,来增强齿轮冲击故障信号。并提出了一种基于复平移Morlet小波和谱峭度的齿轮故障诊断的改进包络分析方法,该方法利用相邻不同级滤波器组中具有交叠频带的滤波器滤波和谱峭度计算结果的相关性,提出了一种改进的滤波器构建方法,该改进方法继承了可根据谱峭度值自适应确定共振解调最佳带通滤波器参数的优点,同时又明显的减少所构建带通滤波器的数量,显著降低了相关计算量,提高了计算效率。齿轮故障仿真结果验证了本方法的可行性。同时,通过对旋转机械变速运行工况的齿轮箱振动研究,提出了一种基于谱峭度的滚动轴承故障包络阶比跟踪分析方法。该方法利用旋转机械运行过程中滚动轴承故障引起的冲击性振动会激起其周围结构共振的原理,应用谱峭度方法自适应地确定优化的共振解调带通滤波中心频率和滤波带宽,进而通过共振解调算法获得包含轴承故障初始阶段振动特征的包络信号,进而通过阶比跟踪技术获得消除了频率模糊的阶比谱,实现对旋转机械变速运行工况下的滚动轴承故障诊断。仿真和测试试验结果验证了本方法的有效性。
[Abstract]:In mechanical equipment, gearbox is widely used in various industries as a universal component to change rotational speed and transfer power. The gearbox working environment is very bad, The gearbox parts are liable to be damaged and malfunction, so it is of great practical significance to take the gearbox as the research object of mechanical equipment condition monitoring and fault diagnosis. The signal processing method based on spectral kurtosis is a new method of fault feature extraction proposed in recent years, which is widely studied in the field of fault diagnosis. However, there are still many problems that need to be solved in practical application. For example, the gearbox fault signal is often buried in the strong background noise, the envelope analysis method based on spectral kurtosis can determine the envelope parameter adaptively, but the calculation is heavy. How to apply this method to the fault diagnosis of gear box speed rise and down process is moderate. In order to solve the above problems, this paper uses AR model to prewhiten the local fault signals of gears. An improved envelope analysis method for gear fault diagnosis based on complex translation Morlet wavelet and spectral kurtosis is proposed. Based on the correlation between the filter filter with overlapping frequency band and the result of spectral kurtosis calculation, an improved filter construction method is proposed in this method. The improved method inherits the advantages of adaptively determining the optimal parameters of the resonant demodulation bandpass filter according to the spectral kurtosis value, at the same time, it obviously reduces the number of the bandpass filters constructed, and significantly reduces the amount of calculation. The simulation results of gear faults verify the feasibility of this method. At the same time, the vibration of the gearbox under the variable speed operation condition of the rotating machinery is studied. Based on spectral kurtosis, a fault envelope order tracking analysis method for rolling bearings is proposed, which makes use of the principle that the impact vibration caused by rolling bearing faults during the operation of rotating machinery will arouse the resonance of its surrounding structures. The spectral kurtosis method is used to adaptively determine the center frequency and the filtering bandwidth of the optimized resonance demodulation bandpass filter, and then the envelope signal which includes the vibration characteristics of the initial stage of the bearing fault is obtained by the resonance demodulation algorithm. By using order tracking technique, the order spectrum of frequency ambiguity is eliminated, and the fault diagnosis of rolling bearing under variable speed operation condition of rotating machinery is realized. The simulation and test results verify the effectiveness of this method.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH165.3

【引证文献】