2020
DOI: 10.1016/j.isatra.2019.11.010
|View full text |Cite
|
Sign up to set email alerts
|

A novel bearing intelligent fault diagnosis framework under time-varying working conditions using recurrent neural network

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
55
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
9

Relationship

1
8

Authors

Journals

citations
Cited by 159 publications
(64 citation statements)
references
References 36 publications
0
55
0
Order By: Relevance
“…• Prone to noise RNN [35] • Parameter sharing is consistent throughout the program • Stacking into deep models is not possible • Long data cannot be interpreted accurately AE [36] • Resultant based on data and not on pre-defined filters • Less complex…”
Section: Architecturementioning
confidence: 99%
“…• Prone to noise RNN [35] • Parameter sharing is consistent throughout the program • Stacking into deep models is not possible • Long data cannot be interpreted accurately AE [36] • Resultant based on data and not on pre-defined filters • Less complex…”
Section: Architecturementioning
confidence: 99%
“…We compared the proposed MLKDCE-PBiLSTM with five advanced methods. They are a DCAE network with five-layer convolutional network [ 15 ], BiLSTM network [ 24 ], LSTM with multiple CNN [ 23 ], MSCNN [ 40 ], LeNet-5 with a new convolutional neural network proposed by Wen [ 41 ]. The six methods adopt the same training strategies in the overall experiments.…”
Section: Performance Verificationmentioning
confidence: 99%
“…The structure of BiLSTM can simultaneously utilize the information of past moments and future moments, which makes the final prediction more accurate than LSTM. An et al [ 23 ] utilized CNN-based LSTM for fault feature extraction of the bearing under time-varying working conditions. Rao et al [ 24 ] utilized convolutional BiLSTM to accurately realize fault diagnosis of rotating machinery.…”
Section: Introductionmentioning
confidence: 99%
“…However, the fly in the ointment is that the signal processing method requires a lot of prior knowledge and manpower [6]. With the rise of deep learning, it is gradually applied to solve the problem of online intelligent fault diagnosis because it can intelligently extract fault features and classify them [7].…”
Section: Introductionmentioning
confidence: 99%