滚动轴承故障非接触多传感器声信号融合及诊断技术研究

发布时间：2018-01-05 20:07

本文关键词：滚动轴承故障非接触多传感器声信号融合及诊断技术研究　出处：《东北石油大学》2012年硕士论文　论文类型：学位论文

【摘要】：滚动轴承是各工业领域的基本元件，在各类机械设备中扮演着重要的角色。滚动轴承的故障诊断方法有很多种，如红外轴温探测法、振动信号分析法、润滑油液分析法等，这些方法各具特色，都能够对常见的滚动轴承故障类型做出有效判断。但这些方法由于自身特点及使用环境的限制，都不能有效地诊断滚动轴承早期故障。另外，在役滚动轴承往往处于移动状态，移动中滚动轴承运行状态的监测是各工业领域的技术难题。为了保证机械设备安全、平稳的运行，降低事故的发生率，本文研究了基于周期性非接触声发射信号的滚动轴承故障诊断方法，目的在于完善滚动轴承故障诊断技术，为滚动轴承状态监测奠定基础。搭建了滚动轴承故障非接触多传感器声发射检测试验台，对移动中带有滚动体、内圈、外圈故障的滚动轴承，分别进行了不同转速、不同移动速度的故障声发射信号采集。研究了形态学滤波技术，根据滚动轴承各类故障声发射信号的特点，找到了合适的结构元素并对各组试验数据进行了形态学滤波处理，剔除了低频、高频噪声等干扰信息，为后续数据分析扫清了障碍。针对多传感器采集时信号重叠和不完整问题，根据声发射撞击信号时差和传感器阵列几何关系建立同声源信号判别公式，以及多传感器同声源信号融合算法，对各组试验信号进行了辨识融合处理，处理结果表明融合信号与故障源信号的相似程度高于各同声源信号。提出了基于周期性声发射撞击计数的滚动轴承故障诊断方法和基于周期性声信号特征参数的滚动轴承故障诊断方法。前者利用滚动轴承故障融合信号的数量与声发射累计撞击计数的对应关系诊断滚动轴承故障；后者通过计算滚动轴承故障周期性声信号的波形特征参数诊断滚动轴承故障。诊断结果表明前者在小周期测试时精确度较高，多周期测试时存在一定的误差，但在可承受范围以内；后者由于是对周期性声信号进行处理，排除了个别非故障源信号的干扰，因而具有较高的准确率。两种方法均可用于滚动轴承早期故障判断，并有效区分故障类型。
[Abstract]:The rolling bearing is the basic element of various industries, plays an important role in all kinds of mechanical equipment. There are many kinds of fault diagnosis method of rolling bearing, such as infrared axle temperature detection method, vibration signal analysis method, the lubricating oil analysis method etc. these methods have their own characteristics, are able to make effective judgment on rolling bearing fault common types. But these methods because of its own characteristics and the use of environmental constraints, can effectively diagnose the incipient fault of rolling bearing. In addition, in the service of rolling bearings are often in a mobile state, monitoring the running state of rolling bearing movement is a technical problem in various industrial fields. In order to ensure the safety of machinery and equipment running smoothly. Reduce the incidence of accidents, this paper studied the periodic non contact fault diagnosis method of rolling bearing based on acoustic emission signal, in order to improve the fault diagnosis of rolling bearing off technology for The foundation of rolling bearing state monitoring is laid.
Build a rolling bearing fault of non contacting sensor acoustic emission testing bench, to move with the rolling body, the inner ring, the outer ring of the rolling bearing fault, respectively, different speed, different moving speed of fault acoustic emission signal acquisition. Research on the morphological filtering technology, according to the characteristics of rolling bearing faults of AE signal, find the appropriate structural elements and each group of test data were studied by morphological filtering, eliminating low frequency, high frequency noise and other interference information for subsequent data analysis to clear the obstacle.
According to the multi sensor signal overlap and incomplete problems, according to the acoustic emission signal and time impact sensor array geometry simultaneous source signal discrimination formula is set up, and the multi-sensor fusion algorithm for simultaneous signals, each test signal identification of fusion processing results show that the degree of similarity fusion signal and fault source signal is higher than the simultaneous the source of the signal.
The cumulative impact count of acoustic emission of the rolling bearing fault diagnosis method and fault diagnosis method of rolling bearing cyclic acoustic signals based on characteristic parameters. Based on the former by the number of fusion signals and the acoustic emissioncumulative impact rolling bearing fault count the corresponding relationship between the fault diagnosis of rolling bearing; the latter through the calculation of rolling bearing fault diagnosis waveform characteristic parameters of bearing periodic fault acoustic signal of rolling. The diagnosis results show that the former higher accuracy in a few period test, there are some errors in multi period test, but within the acceptable range; because the latter is the processing of periodic acoustic signal, eliminate the interference of individual non fault source signals, so it has high accuracy. The two methods can be used for rolling bearing early faults judgment, and distinguish fault types.

【学位授予单位】：东北石油大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH133.33;TH165.3

【参考文献】