基于开关振荡的变频电机轴承油脂状态监测

Monitoring of Bearing Oil Status in Variable Frequency Motors Based on High-frequency Switching Oscillation

轴承状态监测能保证变频电机系统运行状况良好,预防电机故障,提高设备可靠性和运行效率。轴承油脂是轴承运行重要的润滑剂,油脂的劣化也是轴承故障中的早期故障类型,对油脂状态监测可以及时检测轴承过度使用问题,防止油脂质量下降引发变频电机系统停止运行。此处提出了一种基于变频电机轴电流高频开关振荡的在线监测方法,首先,建立了变频电机系统的高频等效模型并分析了油脂劣化对油膜电容参数的影响。然后,研究了油膜电容变化对du/dt轴电流的影响。最后通过加速劣化实验在3kW感应电机系统进行了验证。实验结果表明,将轴电流高频开关振荡作为一维神经网络的数据集进行状态监测,在单一工况下能够达到90%以上的准确率,验证了该方法的准确性。

Bearing status monitoring can ensure the good operation of the variable frequency motor system,prevent motor failures,and improve equipment reliability and operational efficiency.Bearing grease is an important lubricant for bearing operation,and the deterioration of grease is also an early type of bearing failure.Monitoring the grease status can timely detect the problem of excessive use of bearings and prevent the deterioration of grease quality from causing the frequency conversion motor system to stop running.An online monitoring method is proposed based on highfrequency switching oscillation of shaft current in variable frequency motors.Firstly,a high-frequency equivalent model of the variable frequency motor system is established,and the influence of oil degradation on the parameters of oil film capacitance is analyzed.Then,the influence of changes in oil film capacitance on the du/dt axis current is studied.Finally,accelerated degradation experiments are conducted on a 3 kw induction motor system to verify the accuracy of the method.The experimental results showed that using high-frequency switch oscillations of the shaft current as a one-dimensional neural network dataset for state monitoring can achieve an accuracy of over 90%under a single operating condition,verifying the accuracy of the method.