Improved (non-)parametric identification of dynamic systems excited by periodic signals

• Published in: Mechanical systems and signal processing 2011-10, Vol.25 (7), p.2683-2704
• Main Authors: Pintelon, R., Barbé, K., Vandersteen, G., Schoukens, J.
• Format: Article
• Language: English
• Subjects: Approximation; Dynamical systems; Dynamics; Errors; Errors-in-variables; Exact sciences and technology; Excitation; Frequency response function; Fundamental areas of phenomenology (including applications); Leakage; Measurement and testing methods; Noise; Noise model; Nonparametric; Physics; Solid mechanics; Steady state; Structural and continuum mechanics; System identification; Vibration, mechanical wave, dynamic stability (aeroelasticity, vibration control...)
• ISSN: 0888-3270; 1096-1216
• DOI: 10.1016/j.ymssp.2011.03.003

The steady state response of a system to a periodic input is still subject to noise transients. For lightly damped systems these noise transients can significantly increase the variance of the estimated frequency response function (FRF). This paper presents a method that suppresses the influence of the noise transients (leakage errors) in nonparametric FRF and noise (co-)variance estimates of dynamic systems excited by periodic signals. The method is based on a local polynomial approximation of the noise leakage errors on the FRF. Compared with the classical approaches, the proposed procedure is more robust and needs less measurement time (two signal periods are sufficient). The theory is supported by simulation and real measurement examples.