Surrogate Formulation for Chance-Constrained DC Optimal Power Flow with Affine Control Policy

In chance-constrained DC optimal power flow (CC-DCOPF) with affine control policy, the distribution of uncertainty variables is a function of the decision variable. Existing methods fail to transform chance constraints into an analytical form, resulting in computational difficulties. This letter proposes a surrogate formulation for CC-DCOPF considering the affine control policy. The data-driven surrogate model is constructed to identify the mapping relationship between the affine control variables and the tightening boundary of chance constraints. This allows us to analytically reformulate the chance constraints in the presence of non-Gaussian correlated uncertainties. By leveraging the physical power flow features, the surrogate model is scaled into a simple single function. This enables us to use the piecewise linearization technique to approximate the CC-DCOPF to a mixed-integer linear programming (MILP) problem. Case studies show that the proposed method leads to a near-optimal solution with high computational efficiency.