elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Impressum | Datenschutz | Barrierefreiheit | Kontakt | English
Schriftgröße: [-] Text [+]

Quantum Reservoir-Based Surrogate Modeling for Large-Amplitude Gust Responses of a Two-Dimensional Airfoil

Fischbach, Fabian und Leger, Steffen und Friedewald, Diliana (2026) Quantum Reservoir-Based Surrogate Modeling for Large-Amplitude Gust Responses of a Two-Dimensional Airfoil. 2nd International Conference on Applied Quantum Methods in Computational Science and Engineering, 2026-09-14 - 2026-09-17, Delft, Niederlande.

[img] PDF
142kB
[img] PDF - Nur DLR-intern zugänglich bis 1 April 2027
1MB

Offizielle URL: https://aqmcse2026.dryfta.com/program-schedule/abstract-archive/abstract/public/15/quantum-reservoir-based-surrogate-modeling-for-large-amplitude-gust-responses-of-a-two-dimensional-airfoil

Kurzfassung

The design of a transport aircraft requires a large number of load computations for an optimal structural sizing. Different flight points, maneuver and gust load cases, as well as different mass and failure cases, and control laws need to be taken into account, easily summing up to hundreds of thousands of simulations per design cycle. Current simulations are therefore based on unsteady linearized aerodynamic methods as these are computationally efficient. However, when it comes to modern aircraft design, more accurate methods are necessary in order to exploit the full design space and optimize the aircraft's weight. Especially when it comes to large-amplitude excitations as they need to be computed for gust encounters, the currently used time-linearized methods naturally have their limitations. Time-linearized predictions might result, e.g., in an overprediction of actually occurring loads due to an insufficient modeling of, e.g., unsteady flow separation. Therefore, unsteady nonlinear aerodynamic methods, e.g., based on the Unsteady Reynolds-Averaged Navier-Stokes (URANS) equations, are increasingly applied for gust load computations. One of their biggest drawback, however, is the enormous increase in computational time when compared to a RANS-based time-linearized formulation such as, e.g., a time-linearized frequency-somain solver.

This work addresses this specific bottleneck by exploiting the advantages of a reservoir computing-based (RC) surrogate approach for the time-domain prediction of URANS-based gust responses: In general, RC leverages a fixed nonlinear dynamical system (the reservoir) to project inputs into a high-dimensional space such that complex nonlinear relationships become easier to represent and extract. Unlike deep neural networks, RC requires training only of the output layer, typically via simple linear regression, mapping the high-dimensional space to the training targets. The training is therefore computationally very efficient and stable. The reservoirs intrinsic dynamics naturally retain memory of past inputs and capture temporal dependencies, making RC particularly suited for modeling dynamical systems and time series data.

Using quantum reservoir computing (QRC) might even enhance the strengths of classical RC: First, quantum systems naturally evolve in high-dimensional Hilbert spaces, so a relatively small number of qubits can generate a rich set of nonlinear transformations, potentially boosting the expressive power of the reservoir without increasing system size in the classical sense. Secondly, quantum dynamics inherently include features such as superposition, entanglement, and interference. These properties may provide richer temporal and nonlinear processing capabilities than classical reservoirs, potentially improving performance on complex physical systems such as the one described by the current use case. This leads to the idea of QRC as a data-driven surrogate to complement costly URANS computations, thus reducing the total computation time while balancing the approximation error: Use the high-fidelity model only in a subset of the parameter space, and the cheaper surrogate for the rest. This is explored exemplarily for the problem of computing gust responses of a two-dimensional airfoil in terms of two global coefficients across different gust configurations.

elib-URL des Eintrags:https://elib.dlr.de/228222/
Dokumentart:Konferenzbeitrag (Poster)
Zusätzliche Informationen:This project was made possible by the DLR Quantum Computing Initiative and the Federal Ministry of Research, Technology and Space (BMFTR); qci.dlr.de/projects/nemoqc.
Titel:Quantum Reservoir-Based Surrogate Modeling for Large-Amplitude Gust Responses of a Two-Dimensional Airfoil
Autoren:
AutorenInstitution oder E-Mail-AdresseAutoren-ORCID-iDORCID Put Code
Fischbach, Fabianfabian.fischbach (at) dlr.dehttps://orcid.org/0000-0001-9834-4394NICHT SPEZIFIZIERT
Leger, Steffensteffen.leger (at) dlr.dehttps://orcid.org/0009-0009-3597-1214NICHT SPEZIFIZIERT
Friedewald, DilianaDiliana.Friedewald (at) dlr.dehttps://orcid.org/0000-0002-5591-2459NICHT SPEZIFIZIERT
Datum:2026
Referierte Publikation:Ja
Open Access:Nein
Gold Open Access:Nein
In SCOPUS:Nein
In ISI Web of Science:Nein
Status:veröffentlicht
Stichwörter:Quantum Computing, Quantum Machine Learning, Machine Learning, Surrogate Model, Aerodynamics, Computational Fluid Dynamics, Gust Encounter
Veranstaltungstitel:2nd International Conference on Applied Quantum Methods in Computational Science and Engineering
Veranstaltungsort:Delft, Niederlande
Veranstaltungsart:internationale Konferenz
Veranstaltungsbeginn:14 September 2026
Veranstaltungsende:17 September 2026
Veranstalter :Technische Universiteit Delft
HGF - Forschungsbereich:keine Zuordnung
HGF - Programm:keine Zuordnung
HGF - Programmthema:keine Zuordnung
DLR - Schwerpunkt:Quantencomputing-Initiative
DLR - Forschungsgebiet:QC AW - Anwendungen
DLR - Teilgebiet (Projekt, Vorhaben):QC - NeMoQC
Standort: Rhein-Sieg-Kreis
Institute & Einrichtungen:Institut für KI-Sicherheit
Institut für Aeroelastik > Aeroelastische Simulation
Hinterlegt von: Fischbach, Fabian
Hinterlegt am:13 Nov 2026 14:15
Letzte Änderung:13 Nov 2026 14:15

Nur für Mitarbeiter des Archivs: Kontrollseite des Eintrags

Blättern
Suchen
Hilfe & Kontakt
Informationen
OpenAIRE Validator logo electronic library verwendet EPrints
Gestaltung Webseite und Datenbank: Copyright © Deutsches Zentrum für Luft- und Raumfahrt (DLR). Alle Rechte vorbehalten.