Abstract: With the increasing share of new energy sources such as wind power, the climate dependence of the power system is becoming increasingly prominent. Low temperature and cold wave weather can easily lead to unplanned shutdowns of large-scale wind turbines, and existing wind power prediction methods are difficult to predict. Based on this, this paper proposes a short-term combination prediction method for wind power in low temperature and cold wave weather based on extreme scenario division is proposed. Three scenarios were divided based on the different mechanisms of wind power outages affected by low temperatures and cold waves: low-temperature shutdown, strong wind turbine cutting, and icing load reduction. The key meteorological parameters of each scenario were optimized and corrected using HMM. Low-temperature shutdown prediction models, strong wind turbine cutting prediction models, and icing load reduction prediction models were established separately. By combining it with a conventional power prediction model based on normal weather samples, an accurate prediction of wind power output during low temperature and cold wave weather is ultimately achieved. The prediction results of three scenarios under low temperatures and cold wave weather are tested and compared with the conventional power forecast to verify the method's validity in this paper.