Health Index Development for Planetary Gearboxes

• Author / Creator

  Tang,Weixuan

• Planetary gearboxes are widely used in machineries such as wind turbines and helicopters. To maximize their effectiveness over their lifecycle, condition monitoring is often used, and proper health indexes can be developed utilizing condition monitoring data. Health index (HI) development for planetary gearboxes contains two important parts: input feature selection and HI smoothing procedure. Input feature selection is to select the best combination of features as the HI modeling input that provides the highest HI prediction accuracy. HI smoothing procedure is to further improve the modeled HI to get an even higher HI prediction accuracy.
  A reported method uses a feedforward neural network (FFNN) to develop an HI for a type of electric motor. The FFNN is to find the relationship between condition monitoring data and the HI. The reported method uses a fixed stepsize following sequential ordering to select the optimal input features. In addition, the reported method reports an HI dynamic smoothing procedure to further smooth the modeled HI in order to get a higher accuracy of HI prediction. This thesis investigates in-depth the reported method and finds that there are two aspects that are unclear and deficient. These two aspects are thus investigated and the suggestions are provided to address the shortcomings. The findings of this thesis are listed as follows:
  1) The impact of the combinations of the input features in the FFNN-based HI model is investigated. A feature selection method is proposed to find the optimal subset of features.
  2) The impact of both the window size parameter and the maxdrop parameter in the reported HI dynamic smoothing procedure is investigated. An improved HI dynamic smoothing procedure using the optimized window size parameter and the optimized maxdrop parameter is proposed.

• Subjects / Keywords

  • Feature Selection
  • Health Index
  • Planetary Gearboxes
  • Vibration Signal Analysis

• Graduation date

  Spring 2020

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/r3-djpm-k438

• License

  Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.