Enhancement, evaluation and implementation of a load forecasting method

Abstract

The effect of load forecast uncertainty may be well-defined if some of the spinning reserve capacity is needed to supply the load in excess of the amount predicted and, thereby the spinning reserve is reduced. The model was developed for load estimation of Kingdom of Bahrain. The calculation method involves a Monte Carlo technique for the simulation of the load. The model enables the predication of the load against the time during years, where each year is divided into 52 weeks. The forecasting model, computes minimum mean square error (MMSE) forecasts of the conditional mean of reserve power and conditional standard deviation of the innovations in each period over a userspecified forecast possibility. To do this, it views the conditional mean and variance models from a linear filtering perspective, and applies iterated conditional expectations to the recursive equations, one forecast period at a time. The results are obtained and discussed.