Design of Neural Network Based Wind Speed Prediction Model Using GWO

The prediction of wind speed is imperative nowadays due to the increased and effective generation power. Wind power clean, free conservative renewable energy. It necessary predict speed, implement generation. This paper proposes a new model, named WT-GWO-BPNN, by integrating Wavelet Transform (WT), Back Propagation Neural Network (BPNN) Grey Wolf Optimization (GWO). wavelet transform adopted decompose original time series data (wind speed) into approximation detailed band. GWO – BPNN applied speed. used optimize parameters back propagation neural network improve convergence state. work uses six months with 25, 086 points test verify performance proposed model. work, predicts using three-step procedure provides better results. Mean Absolute Error (MAE), Squared (MSE), absolute percentage error (MAPE) Root mean squared (RMSE) are calculated validate Experimental results demonstrate that model has when compared other methods in literature.