In order to obtain a set of combination schemes of ellipsoid surface microstructure parameters with better aerodynamic drag reduction performance, Box-Behnken response surface method was used to optimize the surface microstructure parameters. Four key geometric parameters, namely the major axis, minor axis, depth, and spacing of the ellipsoid, were selected as design variables, and the aerodynamic drag coefficient of the whole vehicle was taken as the response value. A second-order regression model was established to systematically analyze the influence of each parameter and their interactions on the drag reduction performance. The results show that the sensitivity of aerodynamic drag coefficient to the microstructure parameters of ellipsoid surface is first the major axis, then the depth, then the distance, and finally the minor axis. Further parameter optimization research shows that the optimal parameter combination of ellipsoid surface microstructure is 39.9mm long axis, 24.5mm short axis, 6.1mm depth and 131.4mm spacing, which can make the drag reduction rate of the whole vehicle reach up to 6.27%. The percentage difference between the numerical simulation value and that of the parameter optimization prediction value is only 0.03%, which verifies the reliability of the response surface method and the effectiveness of the optimization results. The results of present work are expected to provide an engineering applicable surface microstructure formation parameters for the low resistance design of 600km/h high-speed maglev trains.