Wind Power Prediction Based on Three Machine-Learning Algorithms: Decision Tree, K-Nearest Neighbors and Random Forest

Wind power is getting more attention globally as a renewable and cost-efficient power generation resource. However, its characteristics of intermittency and weather-dependent make the accurate prediction a real challenge. This paper tries to achieve an accurate wind power prediction based on analyzing significant meteorological factors included in the predictor. Three advanced algorithms of decision tree, K-nearest neighbors, and random forest are used to construct predictors based on the processed dataset. The models are tuned with hyperparameters varying in a certain range to find the optimal value, and the evaluation indexes indicate that how well the models fit the data. The results show that the average wind speed and wind direction have main effects on wind power output, and the K-nearest neighbors algorithm performs better in the simulation than the other two methods.