电子测量与仪器学报

刘文彪,段礼祥,耿帆,张金星,李映初.基于CNN BLSTM网络的轴承性能退化预测[J].电子测量与仪器学报,2021,35(2):80-86

基于CNN BLSTM网络的轴承性能退化预测

Bearing performance degradation prognosis based on CNN BLSTM network

DOI：

英文关键词:bearing condition monitoring performance degradation prognosis long short term memory network

基金项目:国家自然科学基金(51674277)、中石油战略合作科技专项（ZLZX2020-05-02）资助

作者	单位
刘文彪	1中国石油大学（北京）安全与海洋工程学院北京102249
段礼祥	1中国石油大学（北京）安全与海洋工程学院北京102249
耿帆	1中国石油大学（北京）安全与海洋工程学院北京102249
张金星	2中国石油塔里木油田分公司库尔勒841000
李映初	2中国石油塔里木油田分公司库尔勒841000

Author	Institution
Liu Wenbiao	1College of Safety and Ocean Engineering, China University of Petroleum, Beijing 102249, China;
Duan Lixiang	1College of Safety and Ocean Engineering, China University of Petroleum, Beijing 102249, China;
Geng Fan	1College of Safety and Ocean Engineering, China University of Petroleum, Beijing 102249, China;
Zhang Jinxing	2Tarim Oilfield Company, CNPC, Korla 841000, China
Li Yingchu	2Tarim Oilfield Company, CNPC, Korla 841000, China

摘要点击次数: 619

全文下载次数: 5

中文摘要:

轴承作为旋转机械设备的重要部件之一，利用监测数据对其开展性能退化评估及剩余寿命预测，对于提高设备可靠性、降低维修成本至关重要。针对传统数据驱动方法在特征提取中过度依赖先验知识和专家经验，未能有效利用时间序列数据中的中长期依赖关系进行建模等问题，提出了一种基于卷积神经网络（CNN）和双向长短期记忆（BLSTM）网络的端到端深度模型进行轴承性能退化预测。该模型采用3层结构，首先，采用CNN直接从原始数据中提取特征向量；然后，将特征向量以时间序列方式重新构造，引入BLSTM网络捕获数据的时序特征；最后，利用一个全连接层和线性回归层来输出模型的最终预测结果。轴承加速寿命实验结果显示所提方法的RMSE和MAPE相比传统方法分别降低了127%和171%，说明该方法能够有效提高轴承性能退化的预测精度。

英文摘要:

Bearings are one of the important components in the rotating machinery; it is significant to assess its performance degradation and predict the remaining useful life of the bearings by using sensors data to improve the reliability and decrease the maintenance costs. For the traditional data driven approaches that rely on prior knowledge or expert experience in feature extraction and do not fully model middle and long term dependencies hidden in time series data, we propose an end to end deep framework for bearing performance degradation prognosis based on convolutional neural network（CNN）and bidirectional long short term memory（BLSTM）. The model adopts three layer structure, the neural network firstly uses CNN to extract feature vectors directly from raw sensor data, then, the feature vector is constructed in a time series manner, and the BLSTM layer is introduced to capture temporal feature, finally, fully connected layers and the linear regression layer are built on top of BLSTM to predict the target value. The results of bearing accelerated life experiments show that the RMSE and MAPE of the proposed method are 127% and 171% lower than the traditional methods, indicating that the method can effectively improve the prediction accuracy of bearing performance degradation.

查看全文查看/发表评论下载PDF阅读器