Search for Author, Title, Keyword
RESEARCH PAPER
A novel method of health indicator construction and remaining useful life prediction based on deep learning
,
 
,
 
,
 
,
 
,
 
 
 
More details
Hide details
1
Army Engineering University of PLA, China
 
2
Hebei Key Laboratory of Condition Monitoring and Assessment of Mechanical Equipment, China
 
3
North China Institute of Aerospace Engineering, China
 
 
Submission date: 2023-06-14
 
 
Final revision date: 2023-07-18
 
 
Acceptance date: 2023-08-18
 
 
Online publication date: 2023-08-26
 
 
Corresponding author
Xisheng Jia   

Army Engineering University of PLA, Shijiazhuang, China, 050003, Shijiazhuang, China
 
 
Eksploatacja i Niezawodność – Maintenance and Reliability 2023;25(4):171374
 
HIGHLIGHTS
  • Utilize technologies such as SK and VMD to extract multi-domain features as feature sets.
  • Utilize monotonicity, trendiness, and robustness to select features for fusion and construct HI.
  • Combining SSAE with LSTM for condition assessment and residual life prediction.
KEYWORDS
TOPICS
ABSTRACT
The construction of health indicators (HI) for traditional deep learning requires human training labels and poor interpretability. This paper proposes an HI construction method based on Stacked Sparse Autoencoder (SSAE) and combines SSAE with Long short-term memory (LSTM) network to predict the remaining useful life (RUL). Extracting features from a single domain may result in insufficient feature extraction and cannot comprehensively reflect the degradation status information of mechanical equipment. In order to solve the problem, this article extracts features from time domain, frequency domain, and time-frequency domain to construct a comprehensive original feature set. Based on monotonicity, trendiness, and robustness, the most sensitive features from the original feature set are selected and put into the SSAE network to construct HI for state partitioning, and then LSTM is used for RUL prediction. By comparing with the existing methods, it is proved that the prediction effect of the proposed method in this paper is satisfied.
ACKNOWLEDGEMENTS
This work is supported by the Na tional Natural Science Foundation of China (grant no. 71871220 ). The support is gratefully acknowledged. The authors would also like to thank the reviewers for their valuable suggestions and comments.
 
CITATIONS (1):
1.
Ultrasound Brain Tomography: Comparison of Deep Learning and Deterministic Methods
Manuchehr Soleimani, Tomasz Rymarczyk, Grzegorz Kłosowski
IEEE Transactions on Instrumentation and Measurement
 
eISSN:2956-3860
ISSN:1507-2711
Journals System - logo
Scroll to top