This paper proposes a feature extraction method for weather conditions of wind power generation. The proposed method makes use of the regression tree to classify input variables and extract rules. In recent years, power system operations are interested in renewable energy such as wind power generation from a standpoint of environment conservation. In that sense, wind power generation is widely-spread in the world. The operation of wind power generation is affected by the weather conditions. In this paper, the relationship between the wind speed and other variables is clarified by the regression tree. The proposed method is tested for real data.