Adaptive Clustering Long Short-Term Memory Network for Short-Term Power Load Forecasting

Short-term load forecasting (STLF) plays an important role in facilitating efficient and reliable operations of power systems optimizing energy planning the electricity market. To improve accuracy prediction, adaptive clustering long short-term memory network is proposed to effectively combine process prediction process. More specifically, adopts maximum deviation similarity criterion algorithm (MDSC) as framework. A bee-foraging learning particle swarm optimization further applied realize its hyperparameters. The consists three parts: (i) a 9-dimensional feature vector classification SVM obtain cluster predicted days; (ii) same kind data are used training network; (iii) trained predict curve day. Finally, experimental results presented show that scheme achieves advantage accuracy, where mean absolute percentage error between value real only 8.05% for first