Abstract: The high permeability of wind power has brought serious impact on the stable operation of power system, and the time-oriented preventive control method has been difficult to apply. Therefore, a new preventive control strategy for dynamic security of power system containing wind power is proposed. First, to solve the problem that uncertainty set of wind power is difficult to describe objectively, this paper uses a Bayesian multivariate nonlinear model to construct uncertainty set of wind power under the influence of wind direction. Next, the classification of fault set considering expected loss is established, and part of fault set is transferred from emergency control set to preventive control set without greatly increasing the preventive control cost, so as to improve the defense capability of transient stability of the system. Then, in view of the severe change of the operating point of the wind power system, the dynamic security index is introduced to describe the dynamic operating curve corresponding to the expected fault. On this basis, the continuous time dynamic security prevention and control method is proposed. Finally, an adaptive whale algorithm based on Mahalanobis distance is constructed to solve the problem that the whale algorithm is difficult to obtain high quality solution for the prevention and control problem. Simulation examples show that the proposed model and method can effectively improve the dynamic security of the system and reduce the cost of prevention and control.