Abstract
This article proposes a hierarchical driving framework for a magnetically levitated planar machine (MLPM) based on an field-programmable gate array (FPGA), which can be easily extended to extend the motion stroke by increasing the number of coils. In the commutation strategy, only the coils underneath the magnet array are energized to reduce the computation burden. The physically constrained data-driven wrench model (PCDWM), considering the end-effects of the coil and magnet array, is introduced in the real-time (RT) control system to distribute the coil current at each control cycle. The H bright circuit driving every coil independently maintains the low cost and high efficiency with the inherent parallelism of the FPGA, promising the synchronous driving of each coil. In the experiments, the constructed prototype realizes a micro-meter positioning resolution. A series of comparative tests highlights the lower power consumption of the driving method and the better motion decoupling performance resulting from the data-driven wrench model (DWM).
Funding Information
  • National Key Research and Development Young Scientist Program (2022YFF0605400)
  • National Natural Science Foundation of China (52175543)
  • National Natural Science Foundation of China (52275569)

This publication has 22 references indexed in Scilit: