Wind energy has been promoted as an environmentally-friendly and sustainable energy. The boost of wind power generation has greatly promoted the developments of modern urbanization. However, the characteristics of complexity, nonlinearity, constraints and multivariability make the controller design challenging. In this work, a Lyapunov-based MPC framework is proposed for nonlinear wind power systems with external disturbances and constraints. Firstly, a typical doubly-fed wind power generation system is established in line with the physical models of key components, i.e., wind turbine, transformation device and generator. Then, considering the exogenous disturbances, a nonlinear Lyapunov-based MPC framework is proposed for the wind power generation system. The optimization objective and constraints are constructed to formulate the control problem. As well, the closed-loop stability of presented MPC approach is derived. In the end, an actual 5MW wind power system is utilized to verify and analyze the effectiveness of designed controller under different wind speeds and weighting parameters.