Robust AC Transmission Expansion Planning with Load and Renewable Generation Uncertainties

TL;DR

This paper presents a robust AC transmission expansion planning method that accounts for load and renewable uncertainties, using a bi-level optimization approach with convex relaxation and Benders Decomposition, verified through Monte Carlo simulations.

Contribution

It introduces a novel dual slave model for AC TEP that avoids conic duality issues and enhances computational efficiency, enabling robust planning under uncertainties.

Findings

Achieves 100% robustness in TEP plans across scenarios.

Uses a convex relaxation and Benders Decomposition for efficient solution.

Verifies robustness with nonlinear AC optimal power flow simulations.

Abstract

With the integration of Renewable Energy Sources (RESs), the power network must be robust to handle the system's uncertain scenarios. DC Transmission Expansion Planning (TEP) plans are generally not feasible for the AC network. A first bi-level Robust Optimization (RO) solution approach for AC TEP with injection uncertainties in loads and RESs generations is proposed in this paper. It utilizes a well-known convex relaxation and is solved by Benders Decomposition (BD), where the master determines the robust AC TEP plan. A novel dual slave model is developed for the second level of BD, which circumvents the issues in using the Conic Duality Theory (CDT), and aids in the worst-case realization of uncertainties using a novel set of constraints. The developed dual slave is not a mixed-integer problem and is solved using the Primal-Dual Interior Point Method (PDIPM). The proposed work also uses additional linear constraints to reduce BD's computational burden and direct the master towards optimality. Monte-Carlo Simulation (MCS) based verification, via actual nonlinear and non-convex AC Optimal Power Flow (OPF), proves 100% robustness of the TEP plans for all scenarios.