Neural Networks to Approximate Solutions of Ordinary Differential Equations

Georg Engel

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem KonferenzbandForschungBegutachtung

Abstract

We discuss surrogate data models based on machine learning as approximation to the solution of an ordinary differential equation. The surrogate model is designed to work like a simulation unit, i.e. it takes a few recent points of the trajectory and the input variables at the given time and calculates the next point of the trajectory as output. The Dahlquist test equation and the Van der Pol oscillator are considered as case studies. Computational demand and accuracy in terms of local and global error are discussed. Parameter studies are performed to discuss the sensitivity of the method.

Originalspracheenglisch
TitelIntelligent Computing - Proceedings of the 2019 Computing Conference
Redakteure/-innenKohei Arai, Rahul Bhatia, Supriya Kapoor
Herausgeber (Verlag)Springer-Verlag Italia
Seiten776-784
Seitenumfang9
ISBN (Print)9783030228705
DOIs
PublikationsstatusVeröffentlicht - 1 Jan 2019
VeranstaltungComputing Conference, 2019 - London, Großbritannien / Vereinigtes Königreich
Dauer: 16 Jul 201917 Jul 2019

Publikationsreihe

NameAdvances in Intelligent Systems and Computing
Band997
ISSN (Print)2194-5357
ISSN (elektronisch)2194-5365

Konferenz

KonferenzComputing Conference, 2019
LandGroßbritannien / Vereinigtes Königreich
OrtLondon
Zeitraum16/07/1917/07/19

    Fingerprint

Schlagwörter

    ASJC Scopus subject areas

    • !!Control and Systems Engineering
    • !!Computer Science(all)

    Dieses zitieren

    Engel, G. (2019). Neural Networks to Approximate Solutions of Ordinary Differential Equations. in K. Arai, R. Bhatia, & S. Kapoor (Hrsg.), Intelligent Computing - Proceedings of the 2019 Computing Conference (S. 776-784). (Advances in Intelligent Systems and Computing; Band 997). Springer-Verlag Italia. https://doi.org/10.1007/978-3-030-22871-2_54