A Fast Optimization Scheme of Coaxial Magnetic Gears Based on Exact Analytical Model Considering Magnetic Saturation

This article presents a harmonic modeling method (HMM) for coaxial magnetic gears (CMGs) by considering the nonlinear B-H curve for soft-magnetic material. CMGs are divided into several annular layers, where Maxwell equation is solved. A convolution matrix of permeability is introduced to obtain the relationship between the magnetic flux density and magnetic field strength, and the continuous boundary conditions can be further satisfied. Subsequently, the magnetic field distribution and electromagnetic performance of CMGs can be calculated by solving the system equations. The computation time and accuracy of HMM are compared with those of finite element analysis (FEA) and subdomain method. The results show that the HMM is relatively fast and precise, and its computation time is about 80% of that of FEA. Then, the parameters of CMGs are optimized based on the HMM and non-dominated sorting genetic algorithm II (NSGA-II) combined optimization method. Particularly, the modulator is divided into multilayers, so its saturation can be modeled more precisely, and its shape can be further optimized. Finally, two optimal cases are evaluated and compared, and the result proves the effectiveness of the integrated optimization method.