Binary classification of day-ahead deregulated electricity market prices using neural networks

The electricity price is influenced by many factors and exhibits a very complicated and irregular fluctuation. The accurate forecasting of various approaches is high in forecasting errors. The levenberg-marquardt (LM) algorithm is train the feed forward neural network (FFNN), and cascade-forward neural network (CFNN) in this paper for binary classification of day-ahead electricity market prices of mainland Spain. All market participants expect electricity price classifications than the forecasting prices for making decisions. Price thresholds are used for binary classification of electricity market prices. Eight alternative data representation cum activation function models based on both FFNN and CFNN are proposed in binary classification of day-ahead electricity prices. The proposed CFNN models results shows an accurate and robust for binary classification of prices.