Gaussian process regression‐based load forecasting model

In this paper, Gaussian Process Regression (GPR)-based models which use the Bayesian approach to regression analysis problem such as load forecasting (LF) are proposed. The GPR is a non-parametric kernel-based learning method having ability provide correct predictions with uncertainty in measurements. proposed model provides an hourly and monthly forecast for Australian city four Indian cities Maharashtra state. Twelve trained historical datasets including environmental data. To evaluate model, actual predicted demand curve plotted mean average percentage error (MAPE) calculated corresponding different kernel functions of model. best author's knowledge, prediction using state has been made first time. MAPE LF 0.15% Australia 0.002%, 0.209%, 0.077%, 0.140% viz. Nasik, Bhusawal, Kolhapur, Aurangabad, respectively. test results illustrate that minimum obtained ‘Exponential’ functions. Furthermore, comparative existing approaches confirms dominance