面向多源传感器信号融合的滚动轴承多层自助最大熵法故障诊断

Multi-layer self-help maximum entropy method for rolling bearing fault diagnosis based on multi-source sensor signal fusion

为了提高滚动轴承故障诊断抗外部因素影响能力,提出了一种面向多源传感器信号融合的滚动轴承多层自助最大熵法故障诊断方法。通过量化分析径向与轴向载荷、转速、温度参数引起的振动状态变化,为判断轴承性能提供了可靠的依据。研究结果表明:振动样本相对径向载荷样本形成了基本一致的概率密度函数,轴承振动性能受到径向载荷的显著影响,轴承载荷与转速引起的轴承振动特性变化情况基本一致。温度引起的轴承振动特性变化情况还不能被精确检测。滚动轴承使用期间,振动状态受到径向载荷、转速、轴向载荷、温度的影响程度指标依次是0.743 5、0.290 2、0.290 8、0.242 9,轴承载荷是对振动状态产生影响最大的因素。

In order to improve the ability of rolling bearing fault diagnosis against external factors,a multi-layer self-service maximum entropy method for rolling bearing fault diagnosis was proposed for multi-source sensor signal fusion. The vibration state changes caused by radial and axial loads,rotational speed and temperature parameters are quantitatively analyzed,which provides a reliable basis for judging the performance of bearings. The results show that the vibration samples form a basically consistent probability density function relative to the radial load data samples,the bearing vibration performance is significantly affected by the radial load,and the bearing vibration characteristics caused by the bearing load and speed are basically consistent. Temperature induced changes in bearing vibration characteristics can’t accurately obtain the influence of temperature on vibration performance. During the use of rolling bearings,the influence degree indexes of vibration state by radial load,rotational speed,axial load and temperature are 0. 743 5,0. 290 2,0. 290 8 and 0. 242 9 respectively. Bearing load is the factor that has the greatest influence on vibration state.