TY - GEN
T1 - Multi-channel Parallel Computing in Capsule Network and Its Application in Mechanical Fault Diagnosis
AU - Qiu, Haiwen
AU - Tao, Jie
AU - Xiao, Zhao
AU - Yang, Wenxian
N1 - Funding Information:
This work is supported by the National Key Research and Development Program of People's Republic of China (Grant No. 2022YFF0608700), the Hunan Provincial Natural Science Foundation of China (Grant No. 2023JJ30265, 2022JJ90003 and 2022JJ90012), the National Natural Science Foundation of China (Grant No. 511905165), Hunan Provincial Department of Education Scientific Research Project (Grant No. 22C0262).
Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.
PY - 2024/9/4
Y1 - 2024/9/4
N2 - Due to complex noise environments, it is difficult to extract bearing fault features from vibration signal and it affects the accuracy of fault diagnosis. To solve the problem, this paper proposes a multi-channel parallel computing capsule network (MPCN), which simultaneously uses various scales kernels to extract the features from original signals. In MPCN, the vibration signal is directly input into the model, then various specifications kernels explore multiple aspects characteristic of signals. Finally, MPCN takes advantage of vector neurons to fuse multiple aspects characteristic. In order to verify the effectiveness of MPCN, fault diagnosis experiments were conducted under different noise conditions. The experimental results show that the accuracy of MPCN exceeds 95%, which is significantly better than traditional diagnostic methods.
AB - Due to complex noise environments, it is difficult to extract bearing fault features from vibration signal and it affects the accuracy of fault diagnosis. To solve the problem, this paper proposes a multi-channel parallel computing capsule network (MPCN), which simultaneously uses various scales kernels to extract the features from original signals. In MPCN, the vibration signal is directly input into the model, then various specifications kernels explore multiple aspects characteristic of signals. Finally, MPCN takes advantage of vector neurons to fuse multiple aspects characteristic. In order to verify the effectiveness of MPCN, fault diagnosis experiments were conducted under different noise conditions. The experimental results show that the accuracy of MPCN exceeds 95%, which is significantly better than traditional diagnostic methods.
KW - Capsule Network
KW - Fault Diagnosis
KW - Parallel Wide Convolution
KW - Vector Neurons
UR - http://www.scopus.com/inward/record.url?scp=85204399613&partnerID=8YFLogxK
UR - https://link.springer.com/book/10.1007/978-3-031-69483-7
U2 - 10.1007/978-3-031-69483-7_16
DO - 10.1007/978-3-031-69483-7_16
M3 - Conference contribution
AN - SCOPUS:85204399613
SN - 9783031694820
SN - 9783031694851
VL - 3
T3 - Mechanisms and Machine Science
SP - 180
EP - 189
BT - Proceedings of the TEPEN International Workshop on Fault Diagnostic and Prognostic
A2 - Liu, Tongtong
A2 - Zhang, Fan
A2 - Huang, Shiqing
A2 - Wang, Jingjing
A2 - Gu, Fengshou
PB - Springer, Cham
T2 - TEPEN International Workshop on Fault Diagnostic and Prognostic
Y2 - 8 May 2024 through 11 May 2024
ER -