Intellectual method for predicting electrical loads based on analysis of big data

A method for predicting electrical loads is proposed, oriented to the big data analysis. The method is based: on decomposition of a time series into a trend and a random component using a fuzzy partition; on a subsequent separate analysis of the trend and the random component using neuro-fuzzy models such as ANFIS (Adaptive Network-based Fuzzy Inference Systems); on combining the forecasting results of the trend and the random component.