【摘要】 针对单一退化变量预测滚动轴承性能退化趋势时可靠性和误差精度较低的问题,提出了基于多退化变量灰色预测模型的滚动轴承剩余寿命预测方法。该方法通过提取滚动轴承全寿命周期振动信号的退化趋势特征参数集,结合退化趋势特征参数集及故障早期突变点,实现滚动轴承的早期故障识别;并根据轴承寿命与特征参数之间的映射关系建立多退化变量灰色预测模型对轴承的剩余寿命进行预测。仿真实验结果表明,多退化变量灰色预测模型具有更高的误差精度和可靠性,其预测效果优于BP神经网络、单一退化变量灰色预测以及SVR(支持向量回归)预测模型,能够更好对滚动轴承寿命的变化趋势进行表征。更多还原

【Abstract】 Aiming at the low reliability and error accuracy in predicting the degradation trend in rolling bearing when using single degenerate variable, a novel method based on grey prediction model with multiple degenerate variables was proposed to predict the residual life of rolling bearings. First of all, the characteristic parameter set related to degradation vibration signals of lifetime bearing was extracted. Then, the early fault identification of rolling bearing was completed by integrating the degradation trend characteristic parameter set and early breakdown point. And the grey prediction model with multiple degenerate variables was established through the mapping relationship between bearing life and characteristic parameters to predict the residual life of the rolling bearing. The simulation results showed that the grey prediction model with multiple degradation variables had higher error accuracy and reliability when compared with the grey prediction model with single degenerate variable, the prediction model of BP neural network and the prediction model of SVR(Support Vector Regression). It could characterize the degradation trend of rolling bearing life better.更多还原