Correlation-Changed-EMD Algorithm for Single Frequency-Sweep Interferometry Signal of High-Speed Rotating Structure Clearance Measurement

Frequency-sweep interferometry (FSI) technology enables high-speed axial clearance measurements but is subject to errors caused by the Doppler effect. Adding auxiliary paths to eliminate the Doppler effect may cause slower measurement speed and low accuracy. In this article, the characteristics of t...

Full description

Saved in:
Bibliographic Details
Published inIEEE sensors journal Vol. 23; no. 23; pp. 29366 - 29375
Main Authors Ren, Yilin, Zhang, Peng, Shao, Bin, Liu, Xianming, Lei, Xiaohua, Liu, Hao, Zou, Wenfei
Format Journal Article
LanguageEnglish
Published New York IEEE 01.12.2023
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects
Online AccessGet full text
ISSN1530-437X
1558-1748
DOI10.1109/JSEN.2023.3326532

Cover

Loading…
More Information
Summary:Frequency-sweep interferometry (FSI) technology enables high-speed axial clearance measurements but is subject to errors caused by the Doppler effect. Adding auxiliary paths to eliminate the Doppler effect may cause slower measurement speed and low accuracy. In this article, the characteristics of the axial clearance variations are analyzed. Subsequently, a correlation-changed empirical mode decomposition (Cor-CEMD) algorithm is developed. The optimal construction value is coarsely located by the correlation algorithm, minimizing the amount of computational data. Then, the optimal construction value is precisely located using changed empirical mode decomposition (EMD) algorithm, thus achieving fast and high-precision clearance reconstruction. This algorithm has a good suppression effect on changes of rotation speed and fluctuation of peak-to-peak clearance. Simulations and experiments have demonstrated that the algorithm has a good effect on high-speed axial clearance reconstruction. For a variable clearance of 4 kHz, the maximum reconstruction error is reduced by 98.4% using the Cor-CEMD algorithm compared with using the EMD algorithm, and the demodulation speed can reach 28 kHz.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1530-437X
1558-1748
DOI:10.1109/JSEN.2023.3326532