Abstract: A novel method based on optimal nonuniform subarray configuration and optimal subarray's excitation is proposed to study low sidelobe beamforming with Millimeter wave conical phased array.This method employs multi-objective evolutionary algorithms following two procedures.Firstly,sampling is made in scanning space and then conformal array patterns are formed.Secondly,two multi-objective functions which realize low sidelobe patterns inside the scanning space anywhere are introduced for constraining these patterns at sampling points.By optimizing subarrays' configuration and their excitations with the multi-objective evolutionary algorithms,the sidelobe level can be suppressed effectively.Simulation results verify the efficiency of the proposed method.