EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

3698. Certification of MPC-based zonal controller security properties using accuracy-aware machine learning proxies

Contributed abstract in session MC-22: Advancements in energy system optimization and analysis tools, stream Energy Management.

Monday, 12:30-14:00
Room: 81 (building: 116)

Authors (first author is the speaker)

1. Manuel Ruiz
RTE

Abstract

The fast growth of renewable energies increases the power congestion risk. To address this issue, the French Transmission System Operator (RTE) has developed closedloop controllers to handle congestion. To ensure their proper functioning, RTE wishes to estimate the probability that the controllers ensure the equipment’s safety. The naive approach to estimate this probability relies on simulating a large amount of randomly drawn scenarios, and use all the outcomes to build a confidence interval around the probability.

Although theory ensures convergence, the computational cost of power system
simulations makes such a process intractable. The target of the present paper is to propose a faster process using machine-learning-based proxies. The amount of required simulations is greatly reduced thanks to an accuracy-aware proxy built with Multivariate Gaussian Processes. Using a proxy instead of the simulator however adds uncertainty to the outcomes. An
adaptation of the Central Limit Theorem is thus proposed to include the uncertainty of the outcomes predicted with the proxy into the confidence interval. As a case study, we designed a simple simulator that is tested on a small IEEE network. Results show that the proxy learns to accurately approximate the simulator’s answer, allowing a significant time gain for the machine-learning based process.

Keywords

Status: accepted

Back to the list of papers