Augmented Neural Ordinary Differential Equations for Power System Identification

TL;DR

This paper introduces an augmented neural ODE framework that learns latent phase angles for power system identification, eliminating the need for direct phase angle measurements and improving modeling accuracy.

Contribution

The paper proposes a novel augmented neural ODE structure that learns latent phase angles using temporal convolutional networks, advancing power system modeling without phase angle data.

Findings

Outperforms simpler augmentation techniques

Effectively learns latent phase angles from historical data

Improves power system identification accuracy

Abstract

Due the complexity of modern power systems, modeling based on first-order principles becomes increasingly difficult. As an alternative, dynamical models for simulation and control design can be obtained by black-box identification techniques. One such technique for the identification of continuous-time systems are neural ordinary differential equations. For training and inference, they require initial values of system states, such as phase angles and frequencies. While frequencies can typically be measured, phase angle measurements are usually not available. To tackle this problem, we propose a novel structure based on augmented neural ordinary differential equations, learning latent phase angle representations on historic observations with temporal convolutional networks. Our approach combines state-of-the art deep learning techniques, avoiding the necessity of phase angle information for the power system identification. Results show, that our approach clearly outperforms simpler augmentation techniques.