A probabilistic methodology based quantitative capacity optimal configuration method of hybrid wind/photovoltaic power supply system is proposed.A discrete probability distribution model is established to represent the random variables in the system，including the uncertainty of power injections and the random failure of components.A multi-objective optimization model is proposed involving three contradictory objectives:minimization of major investment，power inadequacy and voltage deviation.An innovative probabilistic load flow algorithm is introduced for the purpose of fast probability computing while maintaining a relatively high degree of accuracy，which uses moments to calculate and convert cumulants and Gram-Charlier series to obtain probabilistic distribution functions of target variable.The parallel elitist non-dominated sorting genetic algorithm（PNSGA-Ⅱ）is introduced to search the Pareto-optimal solutions.Finally，a numerical example is provided to validate the applications of the proposed method.