Physics-Informed Neural Networks for Non-linear System Identification for Power System Dynamics

Jochen Stiasny, George S. Misyris, Spyros Chatzivasileiadis

    Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

    322 Downloads (Pure)

    Abstract

    Varying power-infeed from converter-based generation units introduces great uncertainty on system parameters such as inertia and damping. As a consequence, system opera-tors face increasing challenges in performing dynamic security assessment and taking real-time control actions. Exploiting the widespread deployment of phasor measurement units (PMUs) and aiming at developing a fast dynamic state and parameter estimation tool, this paper investigates the performance of Physics-Informed Neural Networks (PINN) for discovering the frequency dynamics of future power systems. PINNs have the potential to address challenges such as the stronger non-linearities of low-inertia systems, increased measurement noise, and limited availability of data. The estimator is demonstrated in several test cases using a 4-bus system, and compared with state of the art algorithms, such as the Unscented Kalman Filter (UKF), to assess its performance.
    Original languageEnglish
    Title of host publicationProceedings of 2021 IEEE Madrid PowerTech
    Number of pages6
    PublisherIEEE
    Publication date2021
    Article number9495063
    ISBN (Print)9781665435970
    DOIs
    Publication statusPublished - 2021
    Event2021 IEEE Madrid PowerTech - Virtual Event - from the Alberto Aguilera Campus of Comillas University, Madrid, Spain
    Duration: 28 Jun 20212 Jul 2021
    Conference number: 14
    https://www.powertech2021.com/

    Conference

    Conference2021 IEEE Madrid PowerTech
    Number14
    LocationVirtual Event - from the Alberto Aguilera Campus of Comillas University
    Country/TerritorySpain
    CityMadrid
    Period28/06/202102/07/2021
    Internet address

    Keywords

    • Physics-informed neural networks
    • State estimation
    • Swing equation
    • System identification

    Fingerprint

    Dive into the research topics of 'Physics-Informed Neural Networks for Non-linear System Identification for Power System Dynamics'. Together they form a unique fingerprint.

    Cite this