In hard disk drives (HDDs), there exist multiple mechanical resonances whose central frequencies may shift due to the change of environmental conditions such as the temperature. Such slowly varying resonance frequencies, if not handled properly, may degrade the positioning accuracy and even result in the instability of the closed-loop HDD system. Therefore, it is important to identify these resonance frequencies efficiently without interrupting the reading/writing process in HDDs. One main challenge of the frequency identification in a dual-stage HDD lies in the fact that it is a double-input-single-output (DISO) system. The outputs of the voice coil motor (VCM) and the piezoelectric microactuator (PZT) are coupled together. This paper proposes a practical strategy to identify the resonance frequencies in both the VCM and the PZT without disabling the PZT control process. Bandpass filters are utilized to separate the overall position error signal (PES) into several frequency segments based on priorly-known frequency range for each resonance. Two standard parameter adaptation algorithms are studied and discussed. Simulation results validate the effectiveness of proposed identification strategy.
展开▼
