An Improved Informer Model for Short-Term Load Forecasting by Considering Periodic Property of Load Profiles

Short-term load forecasting (STLF) is an important but a difficult task due to the uncertainty and complexity of electric power systems. In recent times, attention-based model, Informer, has been proposed for efficient feature learning lone sequences. To solve quadratic traditional method, this model designs what called ProbSparse self-attention mechanism. However, mechanism may neglect daily-cycle property profiles, affecting its performance STLF. problem, study proposes improved Informer STLF by considering periodic profiles. The concatenates output values input sequences, outputs results through fully connected layer. This makes could not only inherit superior ability long also extract features experimental on three public data sets showed than others