Electromagnetic Analysis of Flux Characteristics of Double-Sided Switched Reluctance Linear Machine

Abstract

This paper analyzed the flux characteristics of the double-sided switched reluctance linear machine. The magnetic equivalent circuit is given. The magnetization curves at six special mover positions are respectively calculated by the flux tube method, the magnetic flux paths distribution of air gap segments are presented, and the details of the magnetic reluctance component of each path are described. The calculated magnetization curves at six special mover positions is used as the switches for training the magnetic flux linkage model based on the BP neural networks with a GA algorithm. The mathematical mapping relationship between the mover position and the phase current to the magnetic flux linkage is established, and the magnetic flux linkage at certain mover position and certain phase current is calculated by the mathematical mapping relationship. Simulation and the experimental study with the prototype hardware are carried out on the magnetization curves at six special mover positions and three velocities of the mover. It is shown that the magnetization curves at six special mover positions calculated by magnetic equivalent circuit are consistent with the 3-D FEM calculated data and the experimental data, and the simulated phase current waveforms are also consistent with the experimental results. The magnetic equivalent circuit, the analytical calculation of the magnetization curves at six special mover positions, and the BP neural networks magnetic flux linkage model with GA algorithm are validated.