In order to effectively analyze the scenario characteristics of wind power output, scenario analysis model based on Latin hypercube sampling(LHS) and backward reduction(BR) for the uncertain characteristics of wind speed is constructed. The model provides an important basis for rapid analysis of wind power output at any time. Firstly, wind speed characteristics are analyzed, and Weibull distribution of wind speed is introduced. Secondly, the parameter values of the Weibull distribution at each moment are calculated and a set of scenario generation methods based on LHS are proposed. Thirdly, the model of BR is used to reduce the scenarios, so that the several curves obtained can represent the change characteristics of the original scenario to greater extent. Finally, the example analysis proves that the proposed method is better than the traditional K-means clustering algorithm in compactness(CP), separation(SP) and Davies-Bouldin index(DBI), which means that the reduced scenario can better replace the original scenario.