Decoupling Design Model of Square Uniform-Field Coil System Using Improved PSO-ANN Algorithm

In this article, a decoupling coil design method is proposed to solve the coupling problem of square uniform-field coil design for magnetic field control systems in cell culture devices. First, the particle swarm optimization algorithm is improved by adaptive weights and penalty functions. Second, the artificial neural network is used to construct coil design models with magnetic shielding layers combined with an improved particle swarm optimization algorithm. The simulation result shows that the maximum error of the coil design model is less than 0.8%. Then, the design model was used to design the uniform-field coil system in the magnetic control system of the cell culture system. The simulation results show that the maximum normalized error of the coil system in the target region is 0.966%, which is 2.3 times and 2.4 times less than the standard square Helmholtz coil system (2.24%) and bi-planar coil system (2.32%) under same size. The experimental result of the magnetic field control system shows that the uniform-field coil system can produce $1~\mu $ T with 1.7-nT mean error, which is 3.6 times less than the square Helmholtz coil system (6.1-nT average error).