基于混沌分形理论的滚动轴承微小故障诊断

Slight fault diagnosis for rolling bearing based on chaos and fractal theory

为提取轴承微小故障的故障特征,提出一种基于混沌分形理论的滚动轴承故障诊断方法。通过计算滚动轴承振动信号的最大Lyapunov指数,进行轴承运动的混沌识别;然后,对具有混沌特性的振动信号,计算关联维数和盒维数作为故障诊断的状态特征量。当关联维数不能明显区别轴承故障时,利用关联维数与盒维数相结合的方法判别故障;最后,选取滚动轴承滚动体、内圈、外圈存在微小故障和较明显故障以及正常状态7种工况的振动信号进行实验。研究结果表明:该方法能准确提取故障特征并完成滚动轴承的微小故障诊断。该方法为滚动轴承故障诊断提供了新的有效途径。

In order to extract the fault features of bearing slight fault, a new fault diagnosis method for rolling bearing based on chaos and fractal theory was put forward. Firstly, the largest lyapunov exponent of rolling bearing vibration signal was calculated to identify whether the bearing running gets in chaotic or not. Then, the correlation dimension and box dimension of the chaotic signal were calculated as the state character for fault diagnosis. When the correlation dimension can＇t obviously separate bearing faults, the box dimension is used to combine with it for fault diagnosis. Finally, the inner ring, outer ring and rolling element signal of bearing in slight fault, serious fault and the normal state signal in total 7 kinds of conditions were selected to do the experiments. The experimental results show that the method can accurately extract fault feature and complete slight fault diagnosis of rolling bearing. The proposed method provides a new effective approach to the fault diagnosis of rolling bearing.