Abstract: In order to solve the shortcomings of multi-objective optimization design of dry core reactor,a hierarchical optimization method based on sensitivity analysis is proposed. Firstly,a simulation model is established according to the structural parameters of the core reactor,and the electromagnetic,temperature rise and vibration characteristics of the reactor are obtained. On this basis,the influence rules of structural parameters on the performance targets of the reactor is obtained based on the multi-physics simulation and Latin Hypercube test design method combined with the sensitivity analysis technology. Secondly,the multi-objective optimization model of the core reactor is established,which divides the optimization objectives and structural parameters into two levels to perform dimensionality reduction. Finally,the optimal core and winding structure parameters after stratification are obtained using the RSM response surface method and NSGA-II algorithm respectively,and simulation results are provided to verify the effectiveness of the optimization method. The results show that this method can effectively improve the performance of the dry core reactor while controlling costs and losses.. The above optimization method can provide reference for the transformation of the core reactor to the quality benefit type with the green and low-carbon type.