Abstract: Against the backdrop of high-penetration wind power integration into the power grid, this paper focuses on the impacts of multi-temporal and spatial uncertainties on grid operation and proposes a dynamic optimal scheduling method for power systems based on probability and fuzzy theories. Firstly, by analyzing the nonlinear coupling mechanisms among meteorology, load, and equipment operation, a dynamic coupling model across multiple time scales is constructed. A two-layer optimization objective incorporating conditional risk value is designed, and an improved NSGA-II algorithm combined with a rolling optimization method is employed to achieve rapid solution. Through electro-thermal cooperative control and the hierarchical scheduling of energy storage, the operational flexibility of the system is effectively enhanced. Simulation results demonstrate that the proposed method not only reduces operational costs but also promotes wind power consumption, while simultaneously strengthening the system′s risk control capability. The overall performance is excellent.