Quasi-periodic Disturbance Observer for Wideband Harmonic Suppression
Periodic disturbances composed of harmonics usually appear during periodic operation, impairing performance in mechanical and electrical systems. To improve the performance, control for periodic-disturbance suppression has been studied, such as repetitive control and periodic-disturbance observer. F...
Saved in:
Main Author | |
---|---|
Format | Journal Article |
Language | English |
Published |
01.06.2024
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | Periodic disturbances composed of harmonics usually appear during periodic
operation, impairing performance in mechanical and electrical systems. To
improve the performance, control for periodic-disturbance suppression has been
studied, such as repetitive control and periodic-disturbance observer. For
robustness against perturbations in each cycle, slight changes over cycles,
slight variations in the period, and/or aperiodic disturbances, although
wideband harmonic suppression is expected, the conventional methods have
trade-offs among the wideband harmonic suppression, non-amplification of
aperiodic disturbances, and deviation of harmonic suppression frequencies. This
article proposes a quasi-periodic disturbance observer to estimate and
compensate for a quasi-periodic disturbance. The quasi-periodic disturbance is
defined to consist of harmonics and surrounding signals, based on which the
quasi-periodic disturbance observer is designed using a periodic-pass filter of
a first-order periodic/aperiodic separation filter, time delay integrated with
a zero-phase low-pass filter, and an inverse plant model with a first-order
low-pass filter. For the implementation of the proposed observer, its Q-filter
is discretized by the exact mapping of the s-plane to the z-plane, and the
inverse plant model is discretized by the backward Euler method. The
experiments validated the frequency response and position-control precision of
the quasi-periodic disturbance observer in comparison with conventional
methods. |
---|---|
DOI: | 10.48550/arxiv.2406.00362 |