A Long-Term Wind Power Prediction using Support Vector Regression and Ensemble Boosted Tree Algorithm (SVR-EBTA)

Wind Power Forecasting corresponds to an estimate of the expected power production of one or more wind turbines for future reference. First, this paper uses the algorithm Support Vector Regression (SVR) to forecast the possible two-year wind power of Angeles City, Pampanga, Philippines. Support vector Regression is a good for forecasting, classifying and regression which undergo training and testing processes of mean hourly wind speed and errors. To further strengthen the results of SVR, the researchers compared the results through another algorithm, which is Ensembled Boosted Trees (EBT). Using the statistical tool, Paired T-test, the researchers found out that there is no significant difference between the sample means of the results of the two algorithms. Through analysis, having h=0 means that the null hypothesis was accepted at 5% confidence level.