The invention discloses a random robustness optimization scheduling method of a wind electricity-electric automobile united system, and relates to the technical field of clustering optimization scheduling of electric automobiles. The method comprises steps of based on wind electricity history contribution data, generating a wind electricity contribution probability scene; by taking the minimization of system operation cost as an objective, and combining the wind electricity contribution probability scene, establishing a two-stage day-ahead random scheduling model; based on the robustness theory, establishing a two-stage day-ahead random scheduling model containing an electric automobile, and optimizing contribution of conventional units of the system and charging and discharging of the electric automobile. According to the invention, by combining the wind electricity contribution probability scene, based on the second-stage random optimization method and the robustness theory, by taking the minimization of system operation cost as the objective, the two-stage day-ahead random scheduling model containing an electric automobile is established. The method has certain reference value for guiding the electric automobiles to be connected to the power grid to participate in united scheduling in a clustering manner.