Reluctance–Inductance Modeling and Experimental Validation of Inductance-Based Rotor Displacement Self-Sensing in a Dual-Stator Electromagnetic Bearing

Inductance-based self-sensing offers a potential means of reducing reliance on external displacement sensors in electromagnetic bearing systems. This study presents a rotor displacement self-sensing method based on a dual-stator electromagnetic bearing (DSEB) comprising a three-pole auxiliary stator and a six-pole main stator. In the main stator, the control poles generate the electromagnetic force required for rotor levitation, whereas the structurally separated sensing poles are used for displacement estimation without occupying the control-pole winding space. Permanent magnets and the sensing-coil bias current jointly establish the bias flux for partial rotor-weight compensation. To relate rotor displacement to sensing-coil inductance, a linearized mean-reluctance method is developed based on an equivalent magnetic circuit, and analytical models of the sensing-pole air-gap reluctance and sensing-coil inductance are derived. The models are evaluated through three-dimensional finite element analysis and nonrotating quasi-static displacement experiments. The maximum discrepancy between the analytical and finite element inductance results is 5.1%. Under a nominal unilateral air gap of 0.5 mm, a sensing-coil bias current of 0.2 A, and a peak-to-peak displacement range of 0.33 mm, the experimental MAPE values range from 2.09% to 4.42%, and the maximum absolute error relative to an eddy-current displacement sensor is 42.29 μm. These results validate the proposed reluctance–inductance model and demonstrate the feasibility of inductance-based rotor displacement self-sensing under the investigated quasi-static conditions, providing a basis for subsequent dynamic displacement estimation and closed-loop levitation studies.