Identification and cancellation of disturbances having two close sinusoidal components

This paper considers the problem of identifying and cancelling disturbances which contain two sinusoidal components with close frequencies. Previously, Guo and Bodson, 2005, showed that signals composed of two separate but close sinusoids could be represented by a single sinusoid with periodically varying magnitude and phase. Since the magnitude varies periodically, it can be precisely identified and this information can be used to more precisely identify the components of the original signal and more accurately cancel this signal. The base periodic-signal identification algorithm used in this work is the adaptive internal model principle controller rather than the phase locked loop based algorithm used in Guo and Bodson´s work. The results are compared with those from Guo and Bodson´s algorithm. Other approaches based on the internal model are presented and compared in this paper also