Design of Hybrid Magnetic Bearings Based on Integrated Equivalent Magnetic Network Modeling and Sensitivity Analysis for Optimization

Hybrid magnetic bearings (HMBs) play a critical role in allowing magnetically suspended turbomolecular pumps to attain ultra-high vacuum conditions. However, current modeling and optimization approaches for HMB design do not account for the magnetic field coupling between the HMBs and the motor. This omission leads to deviations in the model, inaccurate optimization directions, and difficulties in designing magnetic suspension systems that simultaneously achieve high load capacity and a more compact integrated structure. To address this issue, this article presents a comprehensive HMB design method integrating modeling, sensitivity analysis, and optimization. First, the integrated equivalent magnetic network model is established to analyze the electromagnetic characteristics of the HMBs, which accounts for magnetic coupling effects. Then, a pointwise sensitivity analysis is proposed to quantify the influence of design variables on multiple objectives across the feasible domain and is incorporated into a multiobjective particle swarm optimization algorithm to efficiently guide the search direction. Compared with the preoptimization design, the proposed method improves the weighted average of the maximum radial forces of the two HMBs by 57.03%, reduces the axial length by 11.52%, and mitigates the magnetic coupling by 71.85%. Finally, both simulation and experimental results confirm the effectiveness of the proposed method.