滚动轴承故障诊断的若干方法研究

发布时间：2018-10-16 20:31

【摘要】：机械设备的安全稳定运行对于提高经济效益、减少人员伤亡而言意义重大,为此需要发展设备状态检测与故障诊断技术。本文的研究对象是滚动轴承,从多普勒矫正和故障特征增强角度出发,开展了基于轨边声学和振动的轴承故障诊断方法研究。首先,简要介绍了滚动轴承的振动机理、主要失效形式和故障特征频率计算方法,对实验装置和实验过程进行了说明。然后,针对轨边声学诊断中单麦克风方法存在的一些问题,提出了两种基于麦克风阵列的多普勒校正算法应用于列车轴承轨边声学诊断,充分利用了麦克风阵列所提供的信息来获得参数,而不是通过提前测量或者蛮力搜索的方式,运算效率高;同时具有更强的抗噪性,传统的基于时频脊线提取的方法对噪声很强、时频脊线辨识度低的情况无能为力,而本方法利用了多个麦克风的信息,仍然可以有效地提取出故障信息。最后,提出了一种改进形态滤波用于轴承故障特征增强,为增强原始信号中的脉冲特性而在膨胀运算中结合正弦结构算子采用内积运算代替原始的加运算,然后以扁平结构算子进行腐蚀降噪,相对于传统的一些方法具有更强的故障特征提取能力。
[Abstract]:The safe and stable operation of mechanical equipment is of great significance for improving economic benefit and reducing casualties. Therefore, it is necessary to develop the technology of equipment condition detection and fault diagnosis. The research object of this paper is rolling bearing. From the angle of Doppler correction and fault feature enhancement, the fault diagnosis method of bearing based on track edge acoustics and vibration is studied. Firstly, the vibration mechanism of rolling bearing, the main failure forms and the calculation method of fault characteristic frequency are briefly introduced, and the experimental device and experimental process are explained. Then, aiming at some problems of single microphone method in rail edge acoustic diagnosis, two Doppler correction algorithms based on microphone array are proposed for train bearing rail edge acoustic diagnosis. It makes full use of the information provided by the microphone array to obtain the parameters, rather than by measuring in advance or by brute force search. The traditional method based on time-frequency ridge extraction is very strong in noise and low in recognition of time-frequency ridge. However, this method can still extract fault information effectively by using the information of multiple microphones. Finally, an improved morphological filter is proposed to enhance the fault feature of bearing. In order to enhance the pulse characteristics of the original signal, the inner product operation is used to replace the original addition operation in the expansion operation. Then the flat structure operator is used to reduce the corrosion noise, which has a stronger ability to extract fault features than some traditional methods.
【学位授予单位】：中国科学技术大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TH133.33

【参考文献】