Simulation and Analysis of Irregular Multicoil B0 Shimming in C-type Permanent Magnets Using Genetic Algorithm and Simulated Annealing

Abstract

Multicoil (MC) B₀ shimming, a new shimming technology developed in the recent years, exhibits unique advantages. Irregularly shaped coils can further improve the shimming effect of the MC, but they increase the difficulty and complexity of MC design. In this paper, the shimming characteristics of MC in C-type permanent magnets are studied. Aiming at improving the efficiency of finding the optimal shape for irregular MC, this paper presents a method that uses genetic algorithm and simulated annealing (GASA) to optimize irregular MC shapes for shimming. By simulating the shimming of the irregular MC of the flat-plate model and shoulder-joint model in C-type permanent magnets, the effects of parameters such as complexity, size, and number of groups of coils are analyzed. The shimming of MC in superconducting magnets and permanent magnets is compared and the optimization efficiencies of the genetic algorithms and genetic–annealing algorithms are compared. The results show that an irregular MC can achieve a better shimming effect for a locally nonuniform B₀ field in a sample boundary region of a C-type permanent magnet. GASA is found to be a more stable and highly efficient optimization method.