Data-driven assisted chance-constrained energy and reserve scheduling with wind curtailment

Chance-constrained optimization (CCO) has been widely used for uncertainty management in power system operation. With the prevalence of wind energy, it becomes necessary to consider curtailment as a dispatch variable CCO. However, will cause an impulse distribution, yielding challenges explicit modeling chance constraint. In this paper, novel data-driven framework is developed handle issue. By cap enforced on output, proposed constructs surrogate describe relationship between and tightening boundary constraints. This allows us reformulate CCO with mixed-integer second-order cone programming (MI-SOCP) problem. An error correction strategy by solving convex linear (LP) improve accuracy. The notable characteristic method that constraints can be explicitly accurately handled without any distribution assumptions, which provides great convenience under high percentage renewables. Case studies performed PJM 5-bus IEEE 118-bus systems demonstrate capable accounting influence It shows outperforms existing methods producing more economic secure solutions chance-constrained energy reserve scheduling problem uncertainties.