基于RMS趋势一致性的起重机滚动轴承剩余使用寿命预测

Remaining Useful Life Prediction of Crane Rolling Bearings Based on RMS Trend Consistency

滚动轴承作为起重机设备旋转机构关键零部件,其健康状态将直接影响起重机设备的安全。起重机设备工况复杂,导致同型号轴承退化趋势不一致,使得预测模型训练集的训练参数与测试集不适配,降低了模型的预测精度。针对上述问题,提出一种信号重构方法,将振动信号均方根(RMS)通过信号重构以有效提升趋势一致性,并将其输入到轴承剩余使用寿命(RUL)预测模型GRUA中进行预测。采用XJTU-SY数据集对其进行验证,结果表明,重构后的RMS作为GRUA的输入能有效提升模型预测精度。

As a key component of crane equipment rotating mechanism,the health status of rolling bearings will directly affect the safety of crane equipment.Due to the complexity of crane equipment working conditions,the degradation trend of bearings of the same type is inconsistent,which makes the training parameters of the prediction model training set and the test set unsuitable,and reduces the prediction accuracy of the model.Aiming at the above problems,a signal reconstruction method is proposed,in which the root mean square(RMS)of the vibration signal is reconstructed to effectively improve the trend consistency,and is inputted into the remaining useful life(RUL)prediction model GRUA for prediction.The XJTU-SY dataset is used to validate the method,and the results show that the reconstructed RMS is used as the input to GRUA to effectively improve the prediction accuracy of the model.