Unlike traditional magnetic thrust bearings, a new type of permanent magnet biased bearing structure made of soft magnetic composites in the bevel gear coupled rotor system was proposed that requires both lower eddy currents and greater thrust force. This paper presents the optimization of structure parameters, analysis of magnetic field and dynamic stiffness for two kinds of permanent magnet biased bearing. Firstly, the structural parameters for the bearing were optimized based on the adaptive particle swarm optimization. Secondly, the dynamic magnetic flux distribution for the permanent magnet biased bearing made of carbon steel at 1000 Hz was obtained by finite element method. Then, the equivalent reluctance models for the new type of permanent magnet biased bearing considering the effect of eddy currents are derived. Finally, based on the equivalent reluctance models, the dynamic force–current factor and force–displacement factor are given. The results show that the permanent magnet biased bearing made of soft magnetic composites can provide more stable electromagnetic force and larger bandwidth in the frequency range of 1000 Hz, and have greater electromagnetic force at the same frequency comparing with that made of carbon steel.

A new type of three degrees of freedom axial-radial hybrid magnetic bearing (3-DOF ARHMB) with compact structure, shorter axial length and smaller volume is proposed for the flywheel energy storage system. The axial direction adopts the permanent magnet biased thrust bearing (PMB) made of soft magnetic composite materials (SMCs). In the radial direction, the laminated structure is used to reduce the eddy current, and the Halbach array is introduced to strengthen the magnetic density of the radial air gap. Firstly, the dynamic magnetic flux distribution of the 3-DOF ARHMB is analyzed by the finite element method (FEM). Based on the equivalent magnetic circuit method, the equivalent reluctance model with comprehensive consideration of eddy current effect and magnetic leakage effect is established, and then the frequency responses are analyzed. Secondly, a constraint model coupled with structural parameters, equivalent reluctance and magnetic leakage coefficient is established, and an adaptive particle swarm optimization algorithm (APSO) is used to optimize the bearing parameters. Finally, based on the equivalent reluctance model, the axial and radial force-current factor and force-displacement factor are derived, and the dynamic characteristics of bearings with different structures and materials are compared and analyzed. The results show that the new 3-DOF ARHMB made of SMCs can provide much larger and more stable magnetic force and larger bandwidth than that made of carbon steel materials, and has better dynamic characteristics under higher-frequency conditions, which can meet the industrial requirements of flywheel energy storage system.

Based on the working condition of the magnetic thrust bearing (MTB), the single-objective optimization model (SOOM) is built. The optimum design of structure parameters for two types of magnetic bearings including traditional carbon bearing and soft magnetic composite (SMCs) bearing has been carried out using genetic algorithm. The analysis of eddy current field for two types of bearings are obtained by FEM. It is found that the maximum eddy current field of SMCs bearing is less than that of carbon steel one, and the air gap magnetic density is larger than that of carbon steel bearing at the same frequency. SMCs obtained by the compression of insulated soft magnetic powders have low eddy current. Thus, under the same magnetic force conditions, the input current of SMCs bearings is lower than that of carbon steel ones. The analysis shows that SMCs can replace carbon steel and has superiority in bearing application.