Project Grant 2229409

This three-year, $110,155 project grant from the National Science Foundation's Division of Mathematical Sciences aims to develop new mathematical methods, computer models, and algorithms for electric grid operational planning under the Mathematical and Physical Sciences program (CFDA 47.049). Specifically, the University of Chicago researchers will contribute a general methodology, including novel mathematical models, theory, and algorithms, to systematically account for non-Gaussian error distributions of renewable energy forecasts in optimal power flow problems. This represents the first general treatment of non-Gaussian uncertainties in load and renewable forecasting for grid planning. The researchers will also design new non-Gaussian ambiguity sets to rigorously model misspecification of distribution parameters, with the goal of discovering more robust and reliable network operating points. Validation will utilize real utility data, and practical recommendations will be provided to facilitate adoption. The outcomes are expected to advance chance-constrained optimization approaches for nonlinear, nonconvex problems with mixture uncertainties.