Noise Reduction Disturbance Observer for Disturbance Attenuation and Noise Suppression

In most research works on the disturbance observer (DOB)-based controller, stability conditions are only sufficient and not effective for handling large uncertainties. To deal with large uncertainties, a necessary and sufficient condition for robust stability has been recently provided by utilizing a novel design technique for the Q-filter. Although this design method can deal with large uncertainties, it is not capable of suppressing the effect of sensor noise. To achieve noise suppression as well as disturbance attenuation, this paper proposes a robust controller by utilizing a modified DOB structure. In contrast to the conventional DOB structure, noise suppression ability of the proposed controller can be adjusted by an appropriate choice of the Q-filter. Moreover, a necessary and sufficient condition for the modified DOB system to be robustly stable is newly presented. To show the effectiveness of the proposed controller, the performance is compared with other disturbance attenuation methods through simulation studies. Finally, in order to verify the validity of the proposed method, laboratory experiments are also conducted with brushless dc motor drives, which confirms that the proposed controller is quite useful for attenuating both input disturbance and sensor noise, while DOB controller is not capable of suppressing sensor noise.