Surrogate Model-Assisted Robust Multiobjective Optimization of Equivalent Halbach Array Tubular Permanent Magnet Linear Synchronous Motors

Halbach array tubular permanent magnet synchronous motors exhibit uniform magnetic field distribution and high thrust density, making them well-suited for high-speed, high-precision applications. To enhance motor design robustness, it is essential to balance conflicting performance indicators as well as to consider uncertainties in the manufacturing and assembly processes. However, robustness evaluation typically requires extensive simulations, which can reduce optimization efficiency. To address this issue, this article proposes a surrogate model-assisted multiobjective robust optimization method. To substitute time-consuming conventional finite element calculations, a Kriging model is first used, complemented with a novel adaptive point addition strategy. This strategy enhances local accuracy of the Kriging model by selecting new sample points throughout the optimization process. The Design for Six Sigma methodology is implemented for robust design, significantly improving optimization reliability. In addition, a splicing method is used to approximate the equivalent ideal radial magnetization magnetic ring, effectively addressing the challenges associated with obtaining an ideal radial magnetization ring of small radius. An experimental prototype has been manufactured, and the superiority of the proposed approach has been validated through both simulations and real experiments.