Experimental identification of residual unbalances in run-up high-speed rotor system integrated with active magnetic bearing by synchronized virtual trial unbalances

An identification procedure is described to obtain the residual unbalances in high-speed rotor systems integrated with Active Magnetic Bearing (AMB) using Synchronized Virtual Trial Unbalances (SVTU) as magnetic forces. The practical difficulty in identification of residual unbalances utilizing manual trial unbalances, and a large number of rotor runs is reduced with the implementation of Modified Influence Coefficient Method (MICM). The half-spectrum displacement responses of the accelerated system are utilized along with the magnitude and phase of SVTU to estimate the influence coefficients related to operating frequencies of the system. Moreover, the crossing of critical speed with minimal power consumption in the rotor-bearing system is achieved using Switching Mode Control (SMC) of the bias current and PID controller. The bias current is switched between low and high values to incorporate the adequate requirement of the control current from the controller, thereby reducing the overall power consumption of the AMB system. Also, the practical complications encountered in generating the SVTU in the experimental test rig during the run-up are addressed in the present work. The excitation frequencies of SVTU is essential to be matched with the operating frequencies of rotor-AMB system in real-time to imitate the characteristics of the manual added unbalances. To verify the effectiveness of proposed method, the estimation procedure is carried out for both the conventional influence coefficient method using manual trial unbalances and the MICM using SVTU. The magnitude and phases of residual unbalances identified using MICM are consistent with that obtained using the conventional influence coefficient method with AMB control. The balancing of the rotor-AMB system is performed using the estimated unbalances with a 53%–62% reduction in the time-domain vibrational amplitudes and currents generated by controller at high speeds. The present work aims to solve the problems associated with fault identification in the rotor system supported on conventional bearing by the effective application of magnetic excitation through AMB and influence coefficients.