The invention discloses an electric power system dynamic random economic dispatching method and device based on general distribution. The method comprises the steps that day-ahead system load prediction data, day-ahead wind power prediction data and the like are input, it is assumed that a planned output of a wind power plant is a wind power prediction value, the quadratic programming problem is solved to obtain all thermal power unit outputs based on wind power prediction, the wind power prediction value and all the solved thermal power unit outputs are used as initial iteration points of an interior point method, the interior point method is utilized for iteratively solving a converted convex optimization problem with a constraint condition being linearity until iteration is stopped, and the planned outputs of the thermal power units and the planned output of the wind power plant are output. The method has high popularization value and good application prospects.