摘要

The main magnet in a magnetic resonance imaging (MRI) system creates a static magnetic field that determines the final imaging quality. In a permanent MRI system, shimming rings are commonly used to improve field homogeneity. However, the optimization of the ring structure is challenging owing to the nonlinear properties of the ferromagnetic material. To design a small permanent magnet system, this study explores the application of sensitivity analysis (SA) and particle swarm optimization (PSO) algorithm for the optimization of shimming rings. SA is used to identify the most important parameter of the shimming rings that affects the quality of the magnetic field to simplify the optimization process and improve optimization accuracy and efficiency. PSO is used to solve the complex and nonlinear optimizations of the magnetic field. To illustrate the effectiveness of the proposed method, a specific permanent MRI magnet was modeled. The results show that the inner radius of the shimming ring crucially affects magnetic field quality, with ring height having relatively smaller impact. Compared with the PSO-only optimization procedure, the combined SA-PSO optimization more rapidly converges to a better solution. The optimized shimming rings significantly improve the magnetic field uniformity (similar to 10 fold) compared with that of the initial magnet without shimming rings.