ACTIVE VIBRATION CONTROL OF SHAFTING SYSTEMS WITH ONLINE MODELING AND ADAPTIVE CANCELLATION

Suppressing the vibration of bearings is an effective measure to reduce vibration transmission from the shaft to its supporting structures. This paper investigates an active control method that employs an online identification and filtering algorithm to model the control channel dynamics influenced by the speed-dependent characteristics of the shafting system. The direct FRF measurement method and the subspace filtering algorithm are applied to derive for the control channel an FIR model without noises caused otherwise by the modulated periodic disturbances. Vibration suppression is realized through the Filtered-x LMS algorithm, which is enforced with two additional units to reconstruct disturbance and alleviate output saturation. An experimental system was established to demonstrate the effectiveness of the identification and control methods and the results show that the modulated vibration at the thrust bearing can be suppressed substantially.