Considerations for Application of SOBI to Order Tracking
- 1 Stress Engineering Services,. Houston.
Abstract
Current methods of order tracking, such as synchronous resampling, Gabor filtering, and Vold-Kalman filtering have undesirable traits. Each method has two or more of the following deficiencies: requires measurement or estimate of rotational speed over time, failure to isolate the contribution of crossing orders in the vicinity of the crossing time, large computational expense, end effects. In this work a new approach to the order tracking problem is taken. The Second Order Blind Identification (SOBI) algorithm is applied to synthesized data. The technique is shown to be very successful at isolating crossing orders and circumvents all of the above deficiencies. The method has its own restric-tions: multiple sensors are required and sensors must be mounted on a structure that responds quasi-statically to exci-tation of the rotational system.
- A. Brandt, T. Lago, K. Ahlin, and J. Tuma, “Main principles and limitations of current order tracking methods,” Sound and Vibration Magazine, Vol. 39, No. 3, March 2005, pp. 19-22.
- S. Qian, “Gabor expansion for order tracking,” Sound and Vibration Magazine, Vol. 37, No. 6, June 2003, pp. 18-22.
- H. Vold, J. Leuridan, “High resolution order tracking at extreme slew rates, using Kalman tracking filter”, SAE paper 931288, May 1993.
- A. Belouchrani, K. Abed-Meraim, J.-F. Cardoso, and E. Moulines, “A blind source separation technique using second-order statistics,” IEEE Transctions on Signal Processing, Vol. 45, No. 2, February 1997, pp. 434-444.
- S. I. McNeill, D. C. Zimmerman, “A framework for blind modal identification using joint approximate diagonalization,” Mechanical Systems and Signal Processing Vol. 22, No. 7, October 2008, pp. 1526-1548.