Introduction
Bimonthly, started in 1957
Administrator
Shanxi Provincial Education Department
Sponsor
Taiyuan University of Technology
Publisher
Ed. Office of Journal of TYUT
Editor-in-Chief
SUN Hongbin
ISSN: 1007-9432
CN: 14-1220/N
Administrator
Shanxi Provincial Education Department
Sponsor
Taiyuan University of Technology
Publisher
Ed. Office of Journal of TYUT
Editor-in-Chief
SUN Hongbin
ISSN: 1007-9432
CN: 14-1220/N
location: home > paper >
PDFdownloadsize：780KBviewed：download：
A Lagged Correlation Oriented Feature Extraction Method for Power Equipment Predictive Maintenance
DOI:
10.16355/j.cnki.issn1007-9432tyut.2018.01.013
Received:
Accepted:
abstract:
In this paper, a feature extraction method was proposed to solve the problem of lagged correlation between multi-sensor sequences based on the curve registration model of maximizing the correlation coefficient.The dimensionality reduction is realized in the premise of avoiding the loss of information as far as possible and then a lightweight predictive model is built.The experiment results indicate that our method can effectively reduce training time and lower the computational cost of predictive model in the premise of ensuring the accuracy of prediction.
Keywords:
predictive maintenance; feature extraction; lagged correlation; lightweight model;