2021
DOI: 10.1016/j.ress.2021.107646
|View full text |Cite
|
Sign up to set email alerts
|

Fault prediction of bearings based on LSTM and statistical process analysis

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
27
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
5
2

Relationship

0
7

Authors

Journals

citations
Cited by 82 publications
(27 citation statements)
references
References 8 publications
0
27
0
Order By: Relevance
“…NASA and FEMTO-ST Research Institute have verified the effectiveness of this method. It has been demonstrated that the proposed method is feasible and has a higher prediction accuracy than recurrent neural network (RNN) and support vector regression (SVR) [25]. Based on the historical maintenance data and GIS data, Chen et al [26] proposed a merged-LSTM network for the RUL prediction of the automobile.…”
Section: Machine Learning For Predictive Maintenancementioning
confidence: 99%
“…NASA and FEMTO-ST Research Institute have verified the effectiveness of this method. It has been demonstrated that the proposed method is feasible and has a higher prediction accuracy than recurrent neural network (RNN) and support vector regression (SVR) [25]. Based on the historical maintenance data and GIS data, Chen et al [26] proposed a merged-LSTM network for the RUL prediction of the automobile.…”
Section: Machine Learning For Predictive Maintenancementioning
confidence: 99%
“…LSTM has can be considered as the evolution of RNN which has ability to solve the gradient vanishing or explosion problem introduced in 1997, which has the capability to deal with the long and short term dependent problems 49,50 . The LSTM differs the RNN only in terms of the cell unit having memory which has been shown in Figure 16, where the cell structure consists of mainly three gates which are as follows 51 : Forget gate: It is first gate which decide the information that need to be removed or forgotten from the previous cell which can be expressed as: fgtgoodbreak=σ()Wfggoodbreak×[]hlt1,dtgoodbreak+bfg$$ {fg}_t=\sigma \bullet \left({W}_{fg}\times \left[{h}_{l_{t-1}},{d}_t\right]+{b}_{fg}\right) $$ here dt$$ {d}_t $$ denotes the input at the present state. Input gate: It exist between other two gates and decides which new information need to be retained or updated into the current state and is given as: igtgoodbreak=σ()Wiggoodbreak×[]hlt1,dtgoodbreak+big$$ {ig}_t=\sigma \bullet \left({W}_{ig}\times \left[{h}_{l_{t-1}},{d}_t\right]+{b}_{ig}\right) $$ Output gate: It extracts the final updated information stored in the current cell which can be represented as: ogtgoodbreak=σ()Woggoodbreak×[]hlt1,dtgoodbreak+bog$$ {og}_t=\sigma \bullet \left({W}_{og}\times \left[{h}_{l_{t-1}},{d}_t\right]+{b}_{og}\right) $$ where, fgt$$ {fg}_t $$, …”
Section: Forecasting ML Algorithm Usedmentioning
confidence: 99%
“…From an application perspective, fault detection systems have been developed in many areas such as rolling bearing, machines, industrial systems, mechatronics systems, industrial cyber-physical systems, and industrial-scale telescopes, to name a few [15,[23][24][25][26][33][34][35]37,38,41,42]. Some of them describe some advantages and disadvantages over others in the applied methodology to obtain better results.…”
Section: Fault Detectionmentioning
confidence: 99%
“…For example, Yao Li [ 37 ] presented a branched Long-Short Term Memory (LSTM) model with an attention mechanism to discriminate multiple states of a system showing high performance in its prediction based on the F1-score metric. On the other hand, Liu et al [ 38 ] showed a strategy for failure prediction using the LSTM model in a multi-stage regression model to predict the trend; this is then used to classify the level of degradation by similarity with established failure profiles, achieving improvement estimates with better precision.…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation