Distributionally Robust Optimal Power Flow with Uncertain Renewable Energy Output

TL;DR

This paper introduces a multi-stage distributionally robust method for solving the optimal power flow problem in power grids with uncertain renewable energy output, enhancing reliability and reducing conservativeness.

Contribution

It develops a novel multi-stage distributionally robust DC-OPF model that handles ambiguous renewable output distributions without assuming specific distribution forms.

Findings

Reduces conservativeness in OPF solutions

Maintains high reliability under renewable uncertainty

Effective in real-world power grid scenarios

Abstract

Optimal power flow (OPF) is an important tool for Independent System Operators (ISOs) to deal with the power generation management. With the increasing penetration of renewable energy into power grids, challenges arise in tackling the OPF problem due to the intermittent nature of renewable energy output. To address these challenges, we develop a multi-stage distributionally robust approach for the direct-current optimal power flow (DC-OPF) problem to minimize total generation cost under renewable energy uncertainty. In our model, we assume the renewable energy output follows an ambiguous distribution that can be characterized by a confidence set. By utilizing the revealed data sequentially, the proposed approach can provide a reliable and robust optimal OPF decision without restricting the renewable energy output distribution to any particular distribution class. The computational results also verify the effectiveness of our approach to reduce the conservativeness and meanwhile maintain the reliability.