Robust Dispatch with Demand Response under Decision-dependent Uncertainty

Demand response is recognized as a promising solution to reducing operation costs and alleviating system risks by encouraging end-users to participate in system regulation. In practice, demand response will inevitably introduce uncertainty to decision-making in power system operation due to volatile weather and complicated social behaviors. The decisions, conversely, may affect the uncertainty itself on parameters or information structures. This kind of uncertainty is the so-called decision-dependent uncertainty (DDU), which has not been explored in demand response management. In this paper, we propose a novel two-stage robust economic dispatch model with demand response under DDU, where the uncertainty set of demand response varies with respect to its set-point determined by the decision. To solve this problem, we propose an improved column and constraint generation (C&CG) algorithm with a scenario mapping embedded. The numerical experiments carried on the IEEE 5-bus system illustrate the influence of decision dependence of uncertainty on demand response management and validates the effectiveness of the proposed algorithm.