Abstract
Displacement self-sensing active magnetic bearings (AMBs) have garnered significant attention from both academia and industry for their potential to reduce cost, enable system integration, and enhance reliability. While numerous self-sensing methodologies have been researched, the field lacks a unified framework for discussing their theoretical foundation and practical applicability. This paper analyzes and summarizes various displacement self-sensing methods, deriving the underlying principles and essence of these techniques, and clarifying the intrinsic interconnections of different schemes. The process of self-sensing is constructed through two steps: online inductance estimation and electromagnetic inductance modeling. A novel framework is then proposed, categorizing online inductance estimation, with dedicated discussion on modeling and handling critical nonlinearity like magnetic saturation and the eddy current effect. Furthermore, this review conducts a systematic comparative analysis, evaluating prevalent schemes against key performance metrics such as robustness, stability, signal-to-noise ratio (SNR), and system complexity. Finally, persistent challenges and future research trends are discussed. This review provides a valuable reference for both researchers and engineers when selecting and implementing self-sensing technologies for AMB systems.