Short?term nodal load forecasting based on machine learning techniques

This paper introduces an advanced Short-term Nodal Load Forecasting (STNLF) method that forecasts nodal load profiles for the next day in power systems, based on combined use of three machine learning techniques. Least Absolute Shrinkage and Selection Operator (LASSO) is employed to reduce number features a single forecasting. Principal Component Analysis (PCA) used capture historical loads low-dimensional space compared original high-dimensional where are barely possible depict. Bayesian Ridge Regression (BRR) utilized decide parameters prediction model from statistics perspective. Tests modified PJM data demonstrate effectiveness proposed STNLF state-of-the-art General Neural Network (GRNN) method. Moreover, reliability day-ahead Unit Commitment (UC) solution shown have been improved, forecasted using