Short-term load forecasting using threshold autoregressive models

The author presents a method of forecasting the hourly load demand on a power system. The forecasting method uses threshold autoregressive models with the stratification rule. With the proposed threshold model algorithm, fewer parameters are required to capture the random component in load dynamics. The techniques employed herein are the determination of an optimum threshold number and the construction of the threshold. The optimum stratification rule attempts not only to remove any judgmental input, but also to render the threshold process entirely mechanistic. Hence, the results demonstrate the proposed method's effectiveness in terms of improving precision and reliability.