基于同步平均的齿轮故障特征提取及分析研究

发布时间：2018-05-27 03:07

本文选题：包络分析 + 同步平均　；参考：《昆明理工大学》2013年硕士论文

【摘要】：齿轮为旋转机械传递动力、改变转速的重要零件之一,同时也是旋转机械中最容易发生机械故障的典型零件之一。齿轮振动信号由于存在传播路径复杂,且信号源距离传感器安装位置较远等问题,分析通常较为困难,较难准确判断故障发生与否。当故障严重时,不仅影响生产,甚至会对人的生命安全造成威胁。此外,目前对齿轮故障诊断技术的研究大多是针对平稳振动信号,而对升、降速等非平稳过程的故障特征提取研究相对较少,因此,亟需完善非平稳工况下的齿轮故障诊断技术。振动信号特征分析是机械故障诊断中较为常用的技术方法,通过对振动信号特征的提取可以有效判断机械是否存在故障。在基于振动信号特征分析的齿轮故障诊断领域,已不断发展更为有效的信号处理手段；如何从受复杂背景噪声干扰的振动信号中准确提取出齿轮故障特征信息,是目前齿轮故障诊断领域的研究热点,这也是论文探讨和研究的主要内容。本论文就齿轮的典型故障进行了深入的分析研究,在总结前人研究成果的基础上,提出了包络同步平均的齿轮故障诊断技术。首先利用谱峭度算法获得最优的共振解调参数,提取时域复包络信号,突出其中的冲击成分；然后将时域复包络信号进行角域转换,并选取对应的参考轴,对复包络角域信号的实部和虚部分别进行同步平均;最后通过阶比谱图提取其故障特征。同时,论文针对齿轮变速过程中的故障特征提取进行了研究,对于升、降速阶段的齿轮振动信号,利用阶比跟踪技术,把信号转换到角域进行变速过程的齿轮故障特征信息的提取。研究中提出将包络分析与角域同步平均技术相结合,依次选取不同的转轴作为参考轴,可得到不同转轴上齿轮的故障特征,有效解决了多齿轮典型故障特征难以分离的问题。该方法可有效消除原包络信号中的宽带噪声干扰,分离出与故障齿轮所在轴有关的阶比分量,同时可克服转速波动对信号分析产生的频率模糊现象。配合上述理论研究,在论文研究中进行了仿真和实际测试研究,仿真和实测试验结果验证了所提出方法的有效性。
[Abstract]:Gear is one of the most important parts for rotating machinery to transfer power and change rotational speed. It is also one of the typical parts of rotating machinery which is prone to mechanical failure. Due to the complex propagation path and the distance between the signal source and the sensor, the analysis of gear vibration signal is usually difficult, and it is difficult to accurately judge whether the fault occurs or not. When the fault is serious, it will not only affect the production, but also threaten the safety of human life. In addition, most of the researches on gear fault diagnosis are aimed at stationary vibration signals, but the research on fault feature extraction of non-stationary processes such as rising and decelerating is relatively few. It is urgent to improve the gear fault diagnosis technology under non-stationary working conditions. Vibration signal feature analysis is a common technique in mechanical fault diagnosis. The feature extraction of vibration signal can effectively judge whether machinery has fault or not. In the field of gear fault diagnosis based on vibration signal characteristic analysis, more effective signal processing methods have been developed, how to extract gear fault characteristic information from vibration signal interfered by complex background noise accurately, It is a hot spot in the field of gear fault diagnosis, which is also the main content of this paper. In this paper, the typical faults of gears are deeply analyzed and studied. Based on the previous research results, a fault diagnosis technology of gear with envelope synchronous average is proposed. Firstly, the optimal resonance demodulation parameters are obtained by spectral kurtosis algorithm, and the time domain complex envelope signal is extracted to highlight the impact components, and then the time domain complex envelope signal is converted into angular domain, and the corresponding reference axis is selected. The real part and the imaginary part of the complex envelope angle domain signal are synchronously averaged, and the fault characteristics are extracted by order spectrum. At the same time, the paper studies the fault feature extraction in the gear speed changing process. For the gear vibration signal in the rising and falling stages, the order tracking technique is used. The gear fault feature information is extracted by converting the signal to the angle domain. In the study, it is proposed that the fault characteristics of gears on different rotating shafts can be obtained by combining the envelope analysis with the angular domain synchronous averaging technique and selecting different rotating shafts as reference shafts in turn, which effectively solves the problem that the typical fault features of multiple gears are difficult to separate. This method can effectively eliminate the wideband noise interference in the original envelope signal and separate the order component related to the shaft of the fault gear. At the same time, it can overcome the frequency ambiguity caused by the speed fluctuation on the signal analysis. With the above theoretical research, the simulation and practical test are carried out in this paper. The results of simulation and actual test verify the effectiveness of the proposed 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】：2013
【分类号】：TH132.41;TH165.3

【参考文献】