摘要
针对滚动轴承剩余寿命预测问题,利用经验模态分解对滚动轴承全寿命振动信号进行分解,得到有限个IMF分量,对这些IMF分量分别进行模糊熵分析,提取出滚动轴承故障特征信息,得到多频率尺度模糊熵值。然后采用PCA进行降维,建立滚动轴承性能退化评估指标。把经PCA融合后的IMF模糊熵值输入到ELM极限学习机中,训练ELM预测模型,对滚动轴承进行短期退化趋势预测,以及剩余寿命预测,并与经PCA融合后的IMF样本熵值的预测性能进行对比,证明所提指标的预测精度较高且预测结果趋于保守,更适合用于滚动轴承的剩余寿命预测。
The remaining life of rolling bearings is predicted.First of all,empirical mode decomposition(EMD)method is applied to the decomposition of the vibration signals of the total lifespan of the bearing and several IMF components are obtained.The fuzzy entropy analysis of these IMF components is carried out to extract the fault feature information of the rolling bearing,and the fuzzy entropy of multi-frequency scale is obtained.Then,PCA is used to reduce the dimension,and the deterioration evaluation index of the rolling bearing is established.Finally,the index is input into the ELM learning machine,and the ELM prediction model is trained to predict the short term trend of the rolling bearing and the remaining life.Compared with the predicted value of the IMF sample entropy after the fusion of PCA,it is proved that the prediction accuracy of the proposed index is high and the prediction result of the remaining life of the bearing is slightly conservative.
出处
《噪声与振动控制》
CSCD
2018年第1期188-192,共5页
Noise and Vibration Control
基金
国家自然科学基金资助项目(51505002)
安徽省高校自然科学研究重点资助项目(KJ2015A080)
关键词
振动与波
经验模态分解
模糊熵
PCA
ELM
剩余寿命预测
vibration and wave
empirical mode decomposition
fuzzy entropy
PCA
ELM
remaining life prediction
