Two-stage State-Space Model for Short-term Load Forecasting

The new Energy Act which recently passed the
Norwegian Storting, has increased the value of a good
short-term load forecast for all participants on the
market. This paper presents a method for short-term load
forecasting which takes into account the effects of
temperature, wind speed and sunshine. Our tests show
that all of these weather variables are of significant
importance for the load forecasting capabilities, also
when real forecasts of these variables are applied. The
model structure is simple and each model parameter has
a physical meaning for the users of the model. The
model is employed by several electricity utilities