Optimal Power Flow with Multiple Data Uncertainties

Optimal Power Flow (OPF) is an important tool in system planning and operation. Usually, the data required for such a study, is rarely available with complete certainty. The nature of uncertainty in generators’ cost characteristics, network parameters, load model coefficients, and limits on voltages, flows and generations, is generally of non-probabilistic type. Boundary value representation has been useful in such situations. These represent extreme bounds of a variable, in the fuzzy set. This paper, thus attempts to find boundary OPF solution(s) of critical variables and functions, corresponding to multiple input data uncertainties. Such solutions could be of immense value to planners and market players. The proposed approach is based on the Primal-Dual Interior Point method (PDIPM). Results for two IEEE test systems, demonstrate the potential of proposed algorithm. The results have been verified by Monte Carlo Simulations.

► Boundary OPF, a special class of sensitivity based OPF approach is proposed.
► Boundary values of variables and functions are determined for given nonprobabilistic input data uncertainties.
► OPF sensitivities to perturbation in input data are formulated.