An intelligent diagnosis method of rolling bearing based on multi-scale residual shrinkage convolutional neural network

doi:10.1088/1361-6501/ac68d1

	An intelligent diagnosis method of rolling bearing based on multi-scale residual shrinkage convolutional neural network
	Zhao, Xiaoqiang 1,2,3; Zhang, Yazhou 1
刊名	Measurement Science and Technology
	2022-08-01
卷号	33 期号:8
关键词	Convolutional neural networks Deep learning Failure (mechanical) Failure analysis Fault detection Roller bearings Shrinkage Bearing fault diagnosis Convolutional neural network Intelligent diagnosis methods Multi-scale residual shrinkage convolutional neural network Multi-scales Noise environments Rolling bearings Separable convolution Variable operating condition Vibration signal
ISSN号	0957-0233
DOI	10.1088/1361-6501/ac68d1
英文摘要	The vibration signals of rolling bearings are affected by changing operating conditions and environmental noise, so they are characterized by multi-scale complexity. Deep residual shrinkage network can achieve bearing fault diagnosis in strong noise environment, but ignore the multi-scale complexity feature. To address this problem, we propose a multi-scale residual shrinkage convolutional neural network for fault diagnosis of rolling bearing. In this method, a multi-scale residual shrinkage layer based on multi-scale learning and a residual shrinkage block is constructed. By stacking multiple multi-scale residual shrinkage layers, the features of vibration signals are automatically learned from the input data. In addition, to establish the connection of different vibration signals and to reduce the number of parameters in the network, we design a separable convolution block using residual connections and separable convolution. By verifying the effectiveness of the proposed method in Case Western Reserve University and Mechanical Failure Prevention Technology datasets, the results show that the proposed method not only has good noise resistance in strong noise environments, but also has high diagnostic accuracy and good generalization performance in different load condition domains. The proposed method is compared with three other deep learning methods under the same experimental conditions, and the results show that it is superior in rolling bearing fault diagnosis. © 2022 IOP Publishing Ltd.
语种	英语
出版者	Institute of Physics
内容类型	期刊论文
源URL	[http://ir.lut.edu.cn/handle/2XXMBERH/158372]
专题	电气工程与信息工程学院
作者单位	1.College of Electrical and Information Engineering, Lanzhou University of Technology, Lanzhou; 730050, China; 2.Gansu Key Laboratory of Advanced Control of Industrial Processes, Lanzhou, 730050, China; 3.National Experimental Teaching Center for Electrical and Control Engineering, Lanzhou University of Technology, Lanzhou; 730050, China
推荐引用方式 GB/T 7714	Zhao, Xiaoqiang,Zhang, Yazhou. An intelligent diagnosis method of rolling bearing based on multi-scale residual shrinkage convolutional neural network[J]. Measurement Science and Technology,2022,33(8).
APA	Zhao, Xiaoqiang,&Zhang, Yazhou.(2022).An intelligent diagnosis method of rolling bearing based on multi-scale residual shrinkage convolutional neural network.Measurement Science and Technology,33(8).
MLA	Zhao, Xiaoqiang,et al."An intelligent diagnosis method of rolling bearing based on multi-scale residual shrinkage convolutional neural network".Measurement Science and Technology 33.8(2022).