Radial displacement sensorless control for single-winding bearingless switched reluctance motors

To realize the radial displacement self-sensing of the single winding bearingless switched reluctance motors(SWBSRM), a new method of the rotor radial displacement sensorless control is put forward in this paper, which is based on the adaptive relevance vector machine. The motors’ flux-linkage, current, rotor position and rotor radial displacement obtained from the system with sensors are chosen as the sample data, and then the predict models of rotor radial displacement are built by training the relevance vector machine with these sample data. For purpose of meeting the sample characters in the different operation conditions and improving the fitting precision and generalization ability of the predict models, an adaptive algorithm is proposed to optimize the relevance vector models’ parameters. By simulation and experiment on the test motor, it is proved that the proposed method can estimate the rotor radial displacement accurately in the emulational and experimental conditions in real time, and the estimation precision is satisfactory.