A Novel Framework Based on the Stacking Ensemble Machine Learning (SEML) Method: Application in Wind Speed Modeling

Wind speed (WS) is an important factor in wind power generation. Because of this, drastic changes the WS make it challenging to analyze accurately. Therefore, this study proposed a novel framework based on stacking ensemble machine learning (SEML) method. The application for modeling was developed at sixteen stations Iran. SEML method consists two levels. In particular, eleven (ML) algorithms six categories neuron (artificial neural network (ANN), general regression (GRNN), and radial basis function (RBFNN)), kernel (least squares support vector machine-grid search (LSSVM-GS)), tree (M5 model (M5), gradient boosted (GBR), least boost (LSBoost)), curve (multivariate adaptive splines (MARS)), (multiple linear (MLR) multiple nonlinear (MNLR)), hybrid algorithm (LSSVM-Harris hawks optimization (LSSVM-HHO)) were selected as base level 1 addition, LSBoost used meta-algorithm 2 For purpose, output input LSBoost. A comparison results showed that using greatly affected performance algorithms. highest correlation coefficient (R) 0.89. increased accuracy by >43%.