Robust Optimization and Control for Electricity Generation and Transmission

Chaithanya Bandi, Krishnamurthy Dvijotham, David Morton, Haoxiang Yang

Published 2018-03-19Version 1

In this paper we present the theoretical results of solving a robust optimization problem for the power system under uncertainty. Solving the deterministic alternating current optimal power flow (ACOPF) problem has been considered a hard problem since 1960s because the optimization problem is nonlinear and highly nonconvex. Linear approximation of the AC power flow system (DC approximation) has been deployed in the industry but does not guarantee a physically feasible system configuration. In recently years, different convex relaxation schemes of the ACOPF problem have been researched, and under some assumptions, a physically feasible solution can be recovered. Based on these convex relaxation schemes, we construct a robust convex optimization problem with recourse to solve for optimal controllable injections (fossil fuel, nuclear etc.) in electricity power systems under uncertainty (renewable energy generation, demand fluctuation, etc.). We propose a cutting-plane method to solve this robust optimization problem and prove its convergence property. Extensive experiment results indicate that the robust convex relaxation of the ACOPF problem will provide a tight lower bound, and for the test cases where the nominal relaxation is tight, the convex relaxation solution can also be used for the non-convex robust ACOPF problem.