基于阶次分析技术的变工况行星齿轮箱故障诊断方法研究

发布时间：2019-01-30 18:15

【摘要】：行星齿轮箱的主要构成部件有齿圈、太阳轮、行星轮、中心转轴以及行星轴承等,这些主要部件都被紧凑地安放在较小的箱体内。这些紧凑独特的结构特点,使得行星齿轮箱可以传递较大的力和功率,也使得行星齿轮箱在车辆、船舶、航空航天、风电能源等领域被应用广泛。然而在实际工程中,恶劣时变的工作环境使得行星齿轮箱经常承受动态重载作用力,这将会导致行星齿轮箱产生零部件严重磨损、内部齿轮和轴承发生故障等问题,大大降低行星齿轮箱的使用效率。一旦行星齿轮箱产生损坏,将会导致不可估量的人身、经济损失,因此针对服役在时变工作环境下的行星齿轮箱传动系统进行实时有效的状态监测以及故障诊断具有重要意义。由于异常复杂的行星齿轮箱振动信号是限制行星齿轮箱故障诊断技术发展的瓶颈问题,因此如何有效地降低振动信号的复杂度就成为一个重要的研究方向,这也是本文的研究课题。本文所涉及的降低行星齿轮箱振动信号复杂度的方法分为两种:一种是利用阶次分析技术对采集到的振动信号整体进行简化,其中计算阶次分析方法和Vold-Kalman阶次分析方法被选取;另外一种是根据行星齿轮箱自身振动信号的独特构成,只提取出振动信号的相位信息进行研究,从而降低振动信号复杂度。综上所述,本文应用两种行星齿轮箱非稳态振动信号简化方法,对时变工况下行星齿轮箱的状态监测及故障诊断技术展开深入研究,论文的主要工作包括:(1)为了解决Vold-Kalman阶次分析方法的带宽瓶颈问题,本文利用计算阶次图谱和统计指标作为辅助手段,对Vold-Kalman滤波器带宽进行筛选优选,提出了一种基于阶次谱的Vold-Kalman滤波带宽优选方法。(2)利用优化带宽后的Vold-Kalman阶次分析方法与计算阶次分析方法相结合,提出了一种基于阶次分析方法的行星齿轮箱信号简化策略,对传感器采集到的行星齿轮箱振动信号进行整体的信号简化和相应的特征提取。(3)对行星齿轮箱振动信号中的相位信息部分进行深入研究。结合Hilbert变换技术,对振动信号相位信息进行提取,提出了一种基于余弦相位角信号的行星齿轮箱故障诊断方法,从而消去时变工况引起的调幅影响,进而达到简化振动信号的目的,以便更加有效地对行星齿轮箱进行状态监测和故障诊断。通过仿真模型与实验数据的分析对比,对文中所提出的方法的可行性以及有效性进行解释和验证,为故障诊断理论与方法的应用与发展做出了原创性的贡献。
[Abstract]:The main components of the planetary gearbox are gear ring, solar wheel, planetary wheel, center shaft and planetary bearing. These compact and unique structural features make the planetary gearbox transmit large force and power, and make the planetary gearbox widely used in the fields of vehicle, ship, aerospace, wind energy and so on. However, in the actual engineering, the severe and time-varying working environment often causes the planetary gearbox to bear the dynamic heavy load force, which will lead to the serious wear of the parts of the planetary gearbox, the failure of the internal gear and bearing, and so on. The efficiency of planetary gearbox is greatly reduced. Once the planetary gearbox is damaged, it will lead to incalculable personal and economic losses. Therefore, it is of great significance to carry out real-time and effective state monitoring and fault diagnosis for the planetary gearbox transmission system in the time-varying working environment. Because the extremely complex vibration signal of planetary gearbox is the bottleneck problem that restricts the development of fault diagnosis technology of planetary gearbox, how to reduce the complexity of vibration signal effectively becomes an important research direction. This is also the research topic of this paper. There are two methods to reduce the complexity of the vibration signal of the planetary gearbox. One is to simplify the whole vibration signal by using the order analysis technique. The calculation order analysis method and the Vold-Kalman order analysis method are selected. The other is based on the unique structure of the vibration signal of the planetary gearbox, only the phase information of the vibration signal is extracted to reduce the complexity of the vibration signal. To sum up, two simplified methods of unsteady vibration signal of planetary gearbox are applied in this paper, and the state monitoring and fault diagnosis technology of planetary gearbox under time-varying working conditions are studied deeply. The main work of this paper includes: (1) in order to solve the bandwidth bottleneck problem of Vold-Kalman order analysis method, this paper selects the bandwidth of Vold-Kalman filter by using the calculation order map and statistical index as auxiliary means. A bandwidth optimization method for Vold-Kalman filtering based on order spectrum is proposed. (2) the Vold-Kalman order analysis method is combined with the computational order analysis method. A simplified strategy for planetary gearbox signals based on order analysis is proposed. The vibration signal of the planetary gearbox collected by the sensor is simplified and the corresponding features are extracted. (3) the phase information of the vibration signal of the planetary gearbox is studied deeply. Combined with Hilbert transform technology, the phase information of vibration signal is extracted, and a fault diagnosis method of planetary gearbox based on cosine phase angle signal is proposed to eliminate the amplitude modulation effect caused by time-varying working conditions. Thus the vibration signal can be simplified and the condition monitoring and fault diagnosis of planetary gearbox can be carried out more effectively. Through the analysis and comparison of simulation model and experimental data, the feasibility and validity of the proposed method are explained and verified, which makes an original contribution to the application and development of fault diagnosis theory and method.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TH132.425

【参考文献】