Spinning Reserve Optimization Based on Unit Commitment with Loss-of-load-probability Constraint

Spinning reserve (SR) is necessary for the security operation of power system since it can avoid involuntary load shedding. SR should be scheduled considering the balance between economy and reliability. To address the computational intractability cursed by the computation of reliability indices such as loss of load probability (LOLP), many probabilistic methods use simplified formulations of reliability indices in which the multi-tradeoff of the model is neglected and only a dominant or primary tradeoff is considered. However, this will lose considerable accuracy under many conditions. In this paper, a new SR optimization method is proposed in which both solution accuracy and computation efficiency are well respected. This method simplifies the LOLP formulation by only considering the marginal units because the marginal units have significant effect on the SR distribution among the units. How to find the marginal units is proposed in the paper. The effectiveness of the proposed method is examined by the IEEE-RTS system.