Hybrid Local General Regression Neural Network and Harmony Search Algorithm for Electricity Price Forecasting

Proposing a new precise price forecasting method is still challenging task as electricity signals generally exhibit various complex features. In this paper, approach for called hybrid local general regression neural network, and harmony search algorithm (LGRNN-HSA) proposed. The proposed LGRNN-HSA developed by combining the coordinate delay (CD) method, paradigm, network (GRNN) (HSA). CD employed in to reconstruct time series dataset. Then paradigm utilized with GRNN predict future based on nearest neighbours only so that shortcomings global methods can be overcome. To enhance accuracy, HSA optimize not parameters of but also smooth parameter which has great effect accuracy GRNN. LGRNN-HSA, different every training point instead utilizing constant value verify real-world demand dataset employed. simulation results prove significantly improved compared other methods.