Towards Data-Driven Multi-Stage OPF

TL;DR

This paper proposes a data-driven approach to multi-stage optimal power flow that bypasses traditional network identification, directly constructing the optimization problem from data, and demonstrates its effectiveness on a 118-bus system.

Contribution

It introduces a novel data-driven method for multi-stage OPF that eliminates the need for network topology identification, improving efficiency and robustness.

Findings

The data-driven approach performs comparably to classical methods on a 118-bus system.

It reduces the reliance on detailed grid models and parameters.

The method shows promise for real-time power system optimization.

Abstract

The operation of large-scale power systems is usually scheduled ahead via numerical optimization. However, this requires models of grid topology, line parameters, and bus specifications. Classic approaches first identify the network topology, i.e., the graph of interconnections and the associated impedances. The power generation schedules are then computed by solving a multi-stage optimal power flow (OPF) problem built around the model. In this paper, we explore the prospect of data-driven approaches to multi-stage optimal power flow. Specifically, we leverage recent findings from systems and control to bypass the identification step and to construct the optimization problem directly from data. We illustrate the performance of our method on a 118-bus system and compare it with the classical identification-based approach.