The invention discloses a multi-objective structure optimization method of a magnetic gear brushless direct current motor. The method takes volume minimization and efficiency maximization as optimization objectives, takes current, slip frequency, magnetic flux density, power factors and other related limiting conditions into consideration in the analysis process, and selects the number of stator slots, the number of rotor slots, the outer diameter of a rotor end ring and the inner diameter of a rotor end ring as optimization objects, so that the performance of the magnetic gear brushless direct current motor system is improved. The method mainly comprises the steps that chaotic search is concentrated in a chaotic search reverse learning difference method, and the reverse learning method isfused into a difference evolution method; the search capability of the differential evolution method can be improved, and only two control parameters need to be specified, so that the chaotic searchreverse learning differential method has robustness, and the framework has an online parameter adjustment capability, has a better structural optimization capability for the magnetic gear brushless direct current motor, and can effectively improve the dynamic characteristics of a driving system.