Position Sensorless Control For Multiphase Permanent Magnet Bearingless Motor

With the rapid development of the global industry, bearingless motors gradually occupy a space in high speed industry field. And the correlative research on multiphase system is promoted by the improvement of power electronic technology and the development of computer control technology. A novel multiphase bearingless motor with single set of windings, which is based on the theory combination of multiphase system and bearingless motor, has been realized. It could not only operate without the support of mechanical bearing, just like how the traditional bearingless motors work, but also be able to realize the higher power with lower voltage and tolerant control because of its advantages as multiphase system. In this paper, the research on speed sensorless control of the novel motor is of great significance in the research on bearingless motor applications.The main work of this dissertation includes:(a) The mathematical model of a multiphase PM bearingless motor with single set of windings was established. And, the motor’s working principle and control strategies were analyzed. Conventionally, the bearingless motor has two sets of windings embedded in the stator for rotating and suspending respectively. The motor proposed in this paper is different from that traditional one. It has only one set of windings fed with two groups of currents that have the same frequency but different phase difference between adjacent phases. Based on the multiple-freedom-degree control theory, the purposes of motor drive and suspension can be achieved.(b) Searching for the proper position or speed sensorless control method for the multiphase bearingless motor with single set of windings. Based on the success on vector control of the novel motor, the model reference adaptive theory was applied for the sensorless control A model reference adaptive speed estimator was established based on the fourth-order differential current equatioa Both the rotor eccentricity and the current in suspension plane were involved in the speed estimation. A feedback matrix was later introduced to the estimator to get a close-loop observer system which had higher convergence speed. And the resistant identification ensured better robustness. System simulation and experiments have been done. The results have shown the feasibility of the proposed speed sensorless control algorithm,(c) With the TMS320F28335being the digital signal processor, the experiment platform which has special hardware and software structures different from the traditional three-phase system was designed. The proposed control system has good performance in the speed sensorless operation of the novel multiphase bearingless machine.