• DocumentCode
    572846
  • Title

    Feature extraction from the engine measurement signal of mass characteristic based on wavelet analysis

  • Author

    Sun Baihong ; Yuan Junshe

  • Author_Institution
    Basic Theor. Res. Center, Xi´an Aerosp. Propulsion Inst., Xi´an, China
  • fYear
    2012
  • fDate
    24-26 Aug. 2012
  • Firstpage
    128
  • Lastpage
    130
  • Abstract
    It was observed that the measurement signals contained some undesired parts in engine mass characteristic measurement test by single pendulum method, which brought some difficulties for extracting parameters of engine mass characteristic. The useable signals were separated from the undesired signals through the wavelet multi-resolution-analysis. Some parameters of the engine system such as the natural frequencies, logarithmic decrement, and the damping ratio were obtained using the useable signals. The impact of the damping on the natural frequencies of the system was also studied. The sources of the undesired signals and the defective workmanship were found by the features analysis of the undesired signals. Results show that the wavelet analysis is a useful tool with practical meanings that may be used to extract the structural dynamic parameters, improve the measurement accuracy and locate the damage in the engine system.
  • Keywords
    damping; engines; feature extraction; mass measurement; pendulums; signal resolution; wavelet transforms; damping ratio; engine mass characteristic measurement test; engine system; feature extraction; logarithmic decrement; signal sources; single pendulum method; structural dynamic parameter extraction; wavelet analysis; wavelet multiresolution-analysis; Acceleration; Accuracy; Chemicals; Damping; Signal resolution; Sun; Engine; Feature extraction; Wavelet analysis;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Science and Information Processing (CSIP), 2012 International Conference on
  • Conference_Location
    Xi´an, Shaanxi
  • Print_ISBN
    978-1-4673-1410-7
  • Type

    conf

  • DOI
    10.1109/CSIP.2012.6308811
  • Filename
    6308811