Application of Fuzzy-RBF-CNN Ensemble Model for Short-Term Load Forecasting

Accurate load forecasting (LF) plays an important role in the operation and decision-making process of power grid. Although stochastic nonlinear behavior loads is highly dependent on consumer energy requirements, that demands a high level accuracy LF. In spite several research studies being performed this field, accurate remains consideration. article, design hybrid short-term model (STLF) proposed. This work combines features artificial neural network (ANN), ensemble forecasting, deep learning network. RBFNNs CNNs are trained two phases using functional link (FLANN) optimization method with structure. The predictions made from have been computed produced as forecast each activated cluster. framework known fuzzy-RBFNN. proposed outlined to anticipate one-week ahead demand hourly basis, its determined case studies, i.e., Hellenic Cretan systems. Its results validated while comparing four benchmark models like multiple linear regression (MLR), support vector machine (SVM), ML-SVM, fuzzy-RBFNN terms accuracy. To demonstrate performance RBF-CNN, SVMs replace RBF-CNN regressor, identified ML-SVM having 3 layers.