The article describes a nonlinear adaptive controller for the attenuation of harmonic disturbances in nonlinear systems. The steepest descent algorithm is used to adapt the controller coefficients and synthesize the optimum periodic waveform at the input of the nonlinear system, giving the best attenuation of the disturbance. A simple frequency-domain model of the system is required to implement the steepest descent approach. The scheme is particularly applicable where the secondary actuator used for active vibration control is a nonlinear device such as a magnetostrictive actuator. Simulation studies show that an adaptive scheme of this type is robust to the choice of frequency-domain model when the nonlinearity is hysteretic, but much more sensitive when the nonlinearity is a saturation function. A real-time adaptive harmonic controller has been built and used to control the motion of a magnetostrictive actuator. With dc and 7 harmonics under control the system was able to overcome the inherent nonlinearity of the actuator at fundamental frequencies in excess of 3 kHz.
[1]
M. Yar,et al.
Parameter estimation for hysteretic systems
,
1987
.
[2]
I. Stothers,et al.
In-flight experiments on the active control of propeller-induced cabin noise
,
1990
.
[3]
Jagannath Mazumdar,et al.
A simplified approach to the large amplitude vibration of plates and membranes
,
1977
.
[4]
Stephen J. Elliott,et al.
Active control of road noise inside vehicles
,
1994
.
[5]
Y. Wen.
Method for Random Vibration of Hysteretic Systems
,
1976
.
[6]
A. Sideris,et al.
A multilayered neural network controller
,
1988,
IEEE Control Systems Magazine.
[7]
Stephen J. Elliott,et al.
The behavior of a multiple channel active control system
,
1992,
IEEE Trans. Signal Process..
[8]
Marilyn Wun-Fogle,et al.
Characterization of Terfenol-D for magnetostrictive transducers
,
1990
.