The impacts of temperature forecast uncertainty on Bayesian load forecasting

This paper presents a short-term load forecast methodology that is suitable for power system operational planning studies. Bayesian estimation is use to predict multiple step ahead peak forecasts using peak and average temperature forecasts as explanatory variables. Herein, the forecast model is developed and illustrated in a case study with utility-derived power system data. Special attention is given to the practical issue of forecasting the electrical load with imperfect weather information.