A Novel Three-Module Real-Time Adjustable Accuracy Speed Estimator for Active Magnetic Bearing System

Abstract This research aims to investigate the rotational speed estimation in active magnetic bearing (AMB) system. Rotational speed sensors are commonly employed for speed acquisition. Nevertheless, with the continuous advancement of AMB system, the inherent weaknesses of the speed sensors are becoming increasingly prominent in practical applications. Therefore, achieving online speed estimation can effectively address the engineering challenges arising from deficiencies in speed sensors. However, in terms of robustness, accuracy, and tracking performance, the existing speed estimation algorithms are insufficient to simultaneously meet the requirements of the AMB system. Moreover, present techniques are challenging to address prevalent issues in the AMB system, such as noise, harmonic interference, amplitude variations, and misalignment of the rotor axis trajectory. To tackle these problems, this work designs a novel three-module real-time adjustable accuracy speed estimator for AMB system. The algorithm employs displacement signals to estimate the rotational speed. And, it can flexibly adjust its noise resistance and accuracy through parameter tuning. Additionally, it can significantly diminish the influence of noise, harmonic interference, amplitude variations, and misalignment of the rotor axis trajectory. Finally, the simulation and experimental results demonstrate the effectiveness of this algorithm and validate its superior performance in robustness, accuracy, and tracking performance.