Optimizing probabilistic reserve in large power systems

In this paper, a probabilistic methodology for optimizing the spinning reserve requirements is proposed. The spinning reserve amount is determined by minimizing the total cost which includes operating cost, reserve cost and expected interruption cost. The optimization procedure is implemented in two steps: firstly, an approximated schedule can be computed based on simplified probabilities and reduced contingency events; secondly, the optimization is implemented again based on the approximated schedule. A more accurate result can be achieved through the two-step process.