The inclusion of `spinning reserves' in investmetn and simulation models for electricity generation

Abstract The demand for electricity varies substantially from hour to hour, from day to day, and from season to season. Because electricity is a highly perishable commodity which can be stored only at great expense, and because the penalty for failing to meet demand is severe, the fluctuations of demand over time are crucial in determining an optimal investment strategy for generating equipment. Traditionally, demand is represented by a load-duration curve which neglects the time of day when peaks and troughs occur. By incorporating a load-distribution curve into the model in a computationally tractable manner, this paper demonstrates that the results of the traditional models may be sensitive to changes in the assumptions implied by the load-duration curve.