Online Temperature Estimation of Active Magnetic Bearing System Using Parameter Identification in the Amplifier Model

Abstract

For a magnetically suspended control moment gyroscope (MSCMG), a large current will flow through the coil of active magnetic bearings (AMBs) to suppress the gyroscopic torque when the control torque is output, which will cause the temperature of the AMB to increase. In order to monitor magnetic bearing temperature for fault detection, an online estimation method of bearing temperature is proposed. First, the model of power amplifier is established. Next, the parameter identification algorithm is used to identify the coil resistance online, and then the temperature of the AMB can be estimated indirectly according to the relationship between resistance and temperature. Simulation results show that the proposed method can converge the estimated temperature rapidly in the presence of noise.