基于空洞卷积和增强型多尺度特征自适应融合的滚动轴承故障诊断

韩康,战洪飞,余军合,王瑞

Rolling bearing fault diagnosis based on dilated convolution and enhanced multi-scale feature adaptive fusion

Kang HAN,Hongfei ZHAN,Junhe YU,Rui WANG

表 10 噪声和变工况下8种模型准确率

Tab.10 Accuracy of 8 models under noise and variable working conditions

模 型 A B1 B2 B3 MSCNN-LSTM[14] 0.8672 0.8134 0.7568 WDCNN[15] 0.7997 0.7108 0.7392 MSCNN[16] 0.7842 0.6935 0.7317 DC 0.8384 0.7553 0.7046 Resnet[17] 0.8645 0.8287 0.7457 DRSN-CS[18] 0.9006 0.8422 0.7483 MA1DCNN[19] 0.8692 0.8052 0.7705 DC-MAFFM 0.9048 0.8459 0.8071