TR2022-126

Eccentricity Severity Estimation of Induction Machines using a Sparsity-Driven Regression Model


    •  Zheng, X., Liu, D., Inoue, H., Kanemaru, M., "Eccentricity Severity Estimation of Induction Machines using a Sparsity-Driven Regression Model", The Fourteenth Annual Energy Conversion Congress and Exposition, October 2022.
      BibTeX TR2022-126 PDF
      • @inproceedings{Zheng2022oct,
      • author = {Zheng, Xiangtian and Liu, Dehong and Inoue, Hiroshi and Kanemaru, Makoto},
      • title = {Eccentricity Severity Estimation of Induction Machines using a Sparsity-Driven Regression Model},
      • booktitle = {The Fourteenth Annual Energy Conversion Congress and Exposition},
      • year = 2022,
      • month = oct,
      • url = {https://www.merl.com/publications/TR2022-126}
      • }
  • MERL Contact:
  • Research Areas:

    Computational Sensing, Electric Systems, Signal Processing

Abstract:

Eccentricity severity level estimation is of great importance in rotary machine fault detection. However, in practice machine operation conditions may influence the magnitude of fault signatures, making eccentricity severity estimation a challenging problem. In this paper, we develop a linear regression model incorporating multiple fault signature features to estimate the eccentricity severity level of induction machines under different operating conditions. In particular, the eccentricity severity level is modeled as a function of operating conditions and fault signature features including rotating speed, load torque, vibration, as well as current harmonics, etc, with corresponding weights to be determined. By imposing sparsity of weights, we learn from training data which dominant features have relatively larger impacts on the estimation. Experimental results show that our trained model exhibits satisfactory accuracy in quantitatively estimating eccentricity under various operation conditions.