A Method for Structural Design of Magnetically Levitated Planar Motor Based on Minimization of Mean Square Error of Spatial Magnetic Flux Density

Abstract

The pursuit of extreme acceleration in next-generation mechatronic systems necessitates a design paradigm shift for actuators like magnetically levitated planar motors (MLPMs). Conventional design methods, which optimize the electromagnetic plant in isolation, often result in motors whose theoretical performance cannot be achieved due to limitations of the drive electronics. This article introduces an MLPM-oriented mechatronic co-design framework, termed square error minimization-multiple parameter normalization (SSM-MPN), that fundamentally links the motor's physical structure to its final system-level performance. The framework synergistically optimizes the relationships among the Halbach magnet array, coil geometry, and critical drive-system metrics, such as thrust density and thrust-to-weight ratio. By establishing a unified analytical model in which all vital parameters are functions of the magnet array's pole pitch, the SSM-MPN method facilitates holistic, multiobjective optimization. To validate this framework, a complete prototype system was built, featuring the SSM-MPN-optimized motor plant, custom high-current-density flat-wire coils, and high-performance linear amplifiers. The prototype successfully achieved a peak acceleration of 50 m/s$^{2}$ with a 5 kg payload, rigorously validating that our co-design approach creates a superior physical plant capable of fully leveraging the power of advanced drive systems.