Short-term wind power forecasting model based on a two-stage method

To improve the accuracy of short-term wind power forecasting, this paper proposed a two-stage short-term wind power forecasting method based on wind speed prediction and a wind speed-power conversion model. In the first stage, the Support Vector Machine (SVM) model and the Jaya optimization algorithm were used to predict wind speed, yielding accurate wind speed forecasts. In the second stage, the predicted wind speed, along with environmental factors and system state parameters, was used to construct a dynamic wind speed-to-power conversion model using the Gradient Boosting Decision Tree (GBDT). Compared with conventional forecasting methods and models based on power curves, this approach fully considered the impact of environmental and system factors, resulting in higher reliability and accuracy. Experimental results demonstrated that the performance of the two-stage forecasting model surpassed that of the baseline models, enhancing the effectiveness of wind power forecasting and increasing the credibility of the prediction results.