基于改进GADF-SeResNet滚动轴承故障诊断方法

Fault diagnosis method of rolling bearing based on improved GADF-SeResnet

针对传统故障诊断方法对滚动轴承进行故障诊断准确率较低且时效性较差的问题,提出了一种基于格拉姆角场(Gramian Angular Field,GAF)编码技术与改进SeResNet50模型对滚动轴承的故障诊断方法。采用格拉姆角场技术将一维振动信号重编码为二维特征图像,将二维特征图像作为模型的输入,结合ResNet算法在图像特征提取和分类识别方面的优势,实现自动特征提取与故障诊断,最终达成对不同故障类型的分类。为验证方法的有效性,选用凯西斯储大学的滚动轴承数据进行验证,并与其他常用智能算法进行对比,结果表明,所提方法较其他智能算法分类准确率更高且时效性较好。

Aiming at the problems of low fault diagnosis accuracy and poor timeliness of traditional fault diagnosis methods for rolling bearings,a fault diagnosis method for rolling bearings based on Gramian Angular Field(GAF)coding technology and improved SeResNet50 model was proposed.The one-dimensional vibration signal was re-encoded into a two-dimensional feature image by using the Gramian angle field technology,and the two-dimensional feature image was used as the input of the model.Combined with the advantages of the ResNet algorithm in image feature extraction and classification and recognition,automatic feature extraction and fault diagnosis were realized and finally the classification of different fault types was reached.In order to verify the effectiveness of the method,the rolling bearing data of Case Western Reserve University was selected for verification and compared with other commonly used intelligent algorithms.The results show that the proposed method has higher classification accuracy and better timeliness than other intelligent algorithms.