An optimization design method of short-length actively shielded and open structure superconducting MRI magnets is suggested in the paper. Firstly, the section of the solenoid coil is simplified as a current loop with zero section to solve a linear programming problem. The position coordinates in the radius and axial, and current for the loop can be calculated by the linear programming method. Then, the cross-section of the coil is optimized with a genetic algorithm to get appropriate section size. The method of linear programming, especially combining with genetic algorithm, reduces optimizing variables, which makes the design of a magnet feasible. Based on the method, a full open MRI superconducting magnet is designed with maximum radii of 0.8 m and 1.2 m. In the paper, the detailed optimization technologies are presented.
Read full abstract