A novel design methodology for active shim coil

A novel design approach for active shimming coils for superconducting magnets is proposed to compensate for the previous ten components of the field deviation. The analytic method is first used to obtain the topologies of coils of various order fields and establish a coil model. Then the particle swarm optimization method is adopted to optimize parameters, and the deviation of the magnetic field is taken as the fitness function for minimization of the bias of a magnetic field. The results from the analytic method are taken as a reference to set the initial value ranges of parameters. The results have shown that, compared with the traditional analytic method, the coils with this method can generate a field of better quality. Also the method involves less internal memory and CPU usage than the pure numerical algorithm. In addition, it has fast searching ability and demonstrates high efficiency; and the global solution can be effectively found, which facilitates directly winding. (paper)