Two-Stage Stochastic Unit Commitment Considering Distribution Perturbation via Contamination Technique

Abstract

Stochastic unit commitment (SUC) is deemed as an efficient operational strategy to tackle with the uncertainty in power system. However, in order to implement SUC, a predetermined probability distribution function (PDF) P or finite scenarios for uncertain parameters is often required, whose true pattern is hard to obtain in real world due to various misspecification. In this paper, contamination technique, by which continuous perturbations on P can be achieved, is introduced to study the stability and robustness of SUC with respect to P. We show that under certain conditions, analytical upper and lower bounds, i.e., contamination bounds for SUC can be constructed globally. Then, the proposed method is combined with a risk-based two-stage SUC, where wind penetration and demand response are considered. Numerical experiment on a modified IEEE 14-bus system is performed to test the feasibility and efficiency.