Abstract: The optimization design of conventional wind turbine blades starts from a two-dimensional airfoil,which always takes the maximum lift-drag ratio as the optimization goal. However,the maximum lift-drag ratio of a two-dimensional airfoil is fundamentally different from the high wind energy utilization rate and low aerodynamic load of the three-dimensional blade. The previous blade optimization methods often improves the utilization rate of wind energy and the aerodynamic load. In response to this problem,based on the multi-island genetic algorithm and the blade momentum element theory,this paper proposes that under a given wind condition,the objective function is to take the most weighted wind energy utilization rate and the least aerodynamic load as the objective function,the airfoil profile and the torsion angle as design variables to design and research the method of optimization with multiple objects on the three-dimensional long blade. The optimized design of an actual NREL Phase VI blade shows that compared with the original blade under a given wind condition,the optimized blade has a 3.1% increase in wind energy utilization and a 11.7% reduction in blade root bending moment. Under variable speed and variable wind conditions,the overall aerodynamic efficiency of the optimized blade is improved,and the blade root bending moment is significantly reduced.