AC optimal power flow in the presence of renewable sources and uncertain loads

TL;DR

This paper addresses the challenge of optimizing AC power flow in power grids with high renewable energy penetration and load uncertainty by proposing a probabilistic, scenario-based approach that guarantees constraint satisfaction with high probability.

Contribution

It introduces a novel randomized method called 'scenario with certificates' for AC-OPF that does not rely on prior dependence assumptions and provides probabilistic guarantees.

Findings

The approach effectively manages uncertainty in renewable sources and loads.

It offers probabilistic guarantees on constraint satisfaction.

The method improves dispatch efficiency under uncertainty.

Abstract

The increasing penetration of renewable energy resources, paired with the fact that load can vary significantly, introduce a high degree of uncertainty in the behavior of modern power grids. Given that classical dispatch solutions are "rigid," their performance in such an uncertain environment is in general far from optimal. For this reason, in this paper, we consider AC optimal power flow (AC-OPF) problems in the presence of uncertain loads and (uncertain) renewable energy generators. The goal of AC-OPF design is to guarantee that controllable generation is dispatched at minimum cost, while satisfying constraints on generation and transmission for "almost all" realizations of the uncertainty. We propose an approach based on a randomized technique recently developed, named "scenario with certificates", which allows to tackle the problem without assuming any a-priori dependence of the voltages in the network on the uncertain generators/loads. The proposed solution can exploit the usually available probabilistic description of the uncertainty and variability, and provides solutions with a-priori probabilistic guarantees on the risk of violating the constraints on generation and transmission.