Machine learning for Advanced Gas turbine Injection SysTems to Enhance combustoR performance.

Informations projet

MAGISTER

N° de convention de subvention: 766264

DOI

10.3030/766264

Projet clôturé

Date de signature de la CE 3 Août 2017

Date de début 1 Septembre 2017

Date de fin 30 Novembre 2021

Financé au titre de

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Coût total

€ 3 873 159,00

Contribution de l’UE

€ 3 873 159,00

Coordonné par

UNIVERSITEIT TWENTE
Netherlands

CORDIS fournit des liens vers les livrables publics et les publications des projets HORIZON.

Les liens vers les livrables et les publications des projets du 7e PC, ainsi que les liens vers certains types de résultats spécifiques tels que les jeux de données et les logiciels, sont récupérés dynamiquement sur OpenAIRE .

Livrables

Documents, reports (15)

Control of a thermoacoustic system using machine learning

Control of thermoacoustic system using machine learning

Compressible LES of liquid fuel injection using AVBP.

Compressible LES of liquid fuel injection using AVBP

Modelling of acoustically absorbing liners.
Demo of of combustion instability surrogate model.

Demo of LES of unstable spray flames

Uncertainty handling in engine operation.

Uncertainty handling in engine operation

Droplet measurements data in an atmospheric test rig.

Droplet measurements data in an atmospheric test rig

Development of the Discontinuous Galerkin discretization in SU2: application demo LES of spray flames.

Development of the Discontinuous Galerkin discretization in SU2 application demo LES of spray flames

Compressible LES applied to combustion liners and dilution holes.

Compressible LES applied to combustion liners and dilution holes

Simulation data of the effect of pressure variation on spray combustion.

Simulation data of the effect of pressure variation on spray combustion

Machine learning in thermoacoustic measurements.

Machine learning in thermoacoustic measurements

Comparison of different machine learning algorithms.
Application of machine learning in CFD.
UQ of spray combustion.

Uncertainty Quantification of spray combustion

LES demo thermoacoustic instability in helicopter engine
Measurement data of the acoustic response of kerosene spray flames.

Measurement data of the acoustic response of kerosene spray flames

Other (7)

Summer school: Thermo-acoustics and combustion dynamics in aero gas turbine engines

Thermo-acoustics and combustion dynamics in aero gas turbine engines

Workshop C

Entrepreneurship, ethics, intellectual property rights and management

Workshop B

CFD for spray flame simulations

Workshop D

Measurements of spray flames in aircraft type combustors

Overview of Outreach activities. Final press release

Overview of Outreach activities Final press release

Symposium: Future Aero gas turbine engines Com-bustion Dynamics+Acoustics: Prediction and Remedy

Aero gas turbine engine Combustion Dynamics and Acoustics Prediction and Remedy

Workshop A

Machine Learning, Combustion and Acoustics in aero engine combustors

Open Research Data Pilot (1)

Data Management Plan (DMP)

Mandatory deliverable as consortium decided not to opt out of the pilot on open research.

Websites, patent fillings, videos etc. (1)

MAGISTER web site operational

Publications

Book chapters (1)

Data Assimilation Using Heteroscedastic Bayesian Neural Network Ensembles for Reduced-Order Flame Models

Auteurs: Maximilian L. Croci, Ushnish Sengupta, Matthew P. Juniper
Publié dans: Computational Science – ICCS 2021 - 21st International Conference, Krakow, Poland, June 16–18, 2021, Proceedings, Part V, Numéro 12746, 2021, Page(s) 408-419, ISBN 978-3-030-77976-4
Éditeur: Springer International Publishing
DOI: 10.1007/978-3-030-77977-1_33

Conference proceedings (27)

Reduced order models applied to laminar diffusion flames

Auteurs: Nicole Lopes M. B. Junqueira, Luís Fernando Figueira da Silva, Louise Da Costa Ramos
Publié dans: Procceedings of the 18th Brazilian Congress of Thermal Sciences and Engineering, 2020
Éditeur: ABCM
DOI: 10.26678/abcm.encit2020.cit20-0196

Real-time parameter inference in reduced-order flame models with heteroscedastic Bayesian neural network ensembles

Auteurs: Sengupta, Ushnish; Croci, Maximilian L.; Juniper, Matthew P.
Publié dans: Numéro 1, 2021
Éditeur: Cornell University

Bayesian Machine Learning for the Prognosis of Combustion Instabilities From Noise

Auteurs: Ushnish Sengupta; Carl Edward Rasmussen; Matthew P. Juniper
Publié dans: Numéro 2, 2021
Éditeur: Proceedings of the ASME Turbo Expo
DOI: 10.1115/1.4049762

Confidence in Flame Impulse Response Estimation by LES with Uncertain Thermal Boundary Condition

Auteurs: Kulkarni S, Guo S, Silva CF, Polifke W.
Publié dans: 2021
Éditeur: ASME Turbo Expo 2021
DOI: 10.13140/rg.2.2.25121.12642

Thermoacoustic stabilization of combustors with gradient-augmented Bayesian optimization and adjoint models

Auteurs: Ushnish Sengupta1 and Matthew P. Juniper1
Publié dans: 2021
Éditeur: Symposium on Thermoacoustics in Combustion: Industry meets Academia (SoTiC 2021)

Avoiding High-frequency Thermoacoustic Instabilities in Liquid Propellant Rocket Engines Using Bayesian Deep Learning

Auteurs: Sengupta, Ushnish ; Waxenegger-Wilfing, Guenther ; Martin, Jan ; Hardi, Justin ; Juniper, Matthew
Publié dans: Avoiding High-frequency Thermoacoustic Instabilities in Liquid Propellant Rocket Engines Using Bayesian Deep Learning, 2020
Éditeur: Conference: American Physical Society, Division of Fluid Dynamics Meeting 2020 (APS DFD 2020)
DOI: 10.13140/rg.2.2.15452.00649

Static mesh adaptation for reliable large eddy simulation of turbulent reacting flows

Auteurs: P. W. Agostinelli; B. Rochette; D. Laera; J. Dombard; B. Cuenot; L. Gicquel
Publié dans: Crossref, Numéro 5, 2021, ISSN 1527-2435
Éditeur: Physics of Fluids
DOI: 10.1063/5.0040719

Fusing model ensembles and observations together with Bayesian neural networks

Auteurs: Amos, Matt ; Sengupta, Ushnish ; Hosking, Scott ; Young, Paul
Publié dans: 2021
Éditeur: EGU General Assembly Conference Abstracts

Forecasting Thermoacoustic Instabilities in Liquid Propellant Rocket Engines Using Multimodal Bayesian Deep Learning

Auteurs: Ushnish Senguptaa, G ̈unther Waxenegger-Wilfingb, Jan Martinb,Justin Hardib, Matthew P. Junipera,∗
Publié dans: Fluid Dynamics (physics.flu-dyn); Computational Engineering, Finance, and Science (cs.CE); Machine Learning (cs.LG), 2021
Éditeur: Fluid Dynamics (physics.flu-dyn); Computational Engineering, Finance, and Science (cs.CE); Machine Learning (cs.LG)

Online Detection of Combustion Instabilities Using Supervised Machine Learning

Auteurs: Michael McCartney, Wolfgang Polifke
Publié dans: Proceedings of the ASME Turbo Expo 2020: Turbomachinery Technical Conference and Exposition, 2020
Éditeur: ASME Turbo Expo 2020
DOI: 10.1115/gt2020-14834

A model to study spontaneous oscillations in a lean premixed combustor using non-linear analysis

Auteurs: Sara Navarro Arredondo, Jim Kok
Publié dans: Proceedings of the 26th International Congress on Sound and Vibration, Numéro 26, 2019, ISBN 978-1-9991810-0-0
Éditeur: Canadian Acoustical Association

Numerical Study Of A Swirl Atomized Spray Response To Acoustic Perturbations.

Auteurs: Alireza Ghasemi, J.B.W. Kok
Publié dans: Proceedings of the 26th International Congress on Sound and Vibration, Numéro 26, 2019, ISBN 978-1-9991810-0-0
Éditeur: Canadian Acoustical Association

Bayesian machine learning for the prognosis of combustion instabilities from noise

Auteurs: Ushnish Sengupta Carl Rasmussen Matthew Juniper
Publié dans: Proceedings of the ASME 2020 Turbomachinery Technical Conference Exposition, 2020
Éditeur: Proceedings of the ASME 2020 Turbomachinery Technical Conference Exposition
DOI: 10.31224/osf.io/ysgp4

Ensembling geophysical models with Bayesian Neural Networks

Auteurs: Sengupta, Ushnish; Amos, Matt; Hosking, J. Scott; Rasmussen, Carl Edward; Juniper, Matthew; Young, Paul J.
Publié dans: Numéro 2, 2020
Éditeur: Cornell University
DOI: 10.17863/cam.60032

Real-time parameter inference in reduced-order flame models with heteroscedastic Bayesian neural network ensembles

Auteurs: Ushnish Sengupta, Maximilian L. Croci, Matthew P. Juniper
Publié dans: 2020
Éditeur: Cornell University

Comparison of Machine Learning Algorithms in the Interpolation and Extrapolation of Flame Describing Functions

Auteurs: Michael McCartney, Matthias Haeringer, Wolfgang Polifke
Publié dans: Volume 4B: Combustion, Fuels, and Emissions, 2019, ISBN 978-0-7918-5862-2
Éditeur: American Society of Mechanical Engineers
DOI: 10.1115/gt2019-91319

Numerical and Experimental Flame Stabilization Analysis in the New SpinningCombustion Technology Framework

Auteurs: Agostinelli, P. W., Kwah, Y. H., Richard, S., Exilard, G., Dawson, J. R., Gicquel, L., & Poinsot, T.
Publié dans: 2020
Éditeur: In Proceedings of the ASME Turbo Expo 2020: Turbomachinery Technical Conference and Exposition
DOI: 10.1115/gt2020-15035

Real-time parameter inference of nonlinear bluff-body-stabilized flame models using Bayesian neural network ensembles

Auteurs: Maximilian L. Croci1 2, Ushnish Sengupta1 and Matthew P. Juniper1
Publié dans: Symposium on Thermoacoustics in Combustion: Industry meets Academia (SoTiC 2021), 2021
Éditeur: SoTiC 2021 - Symposium on Thermoacoustics in Combustion: Industry meets Academia, 2021

Numerical design of Luenberger observers for nonlinear systems

Auteurs: Louise da C. Ramos, Florent Di Meglio, Valery Morgenthaler, Luis F. Figueira da Silva, Pauline Bernard
Publié dans: 2020 59th IEEE Conference on Decision and Control (CDC), 2020, Page(s) 5435-5442, ISBN 978-1-7281-7447-1
Éditeur: IEEE
DOI: 10.1109/cdc42340.2020.9304163

Reduced Order Models Applied to Laminar Diffusion Flames

Auteurs: N. L. M. B. Junqueira, L. F. Figueira da Silva, L. C. Ramos
Publié dans: 2020 Brazilian Congress of Thermal Sciences and Engineering, Online., 2020
Éditeur: 2020 Brazilian Congress of Thermal Sciences and Engineering, Online.

Reduced Order Model of Laminar Premixed Inverted Conical Flames

Auteurs: Louise da Costa Ramos, Florent Di Meglio, Luis Fernando F. Da Silva, Valery Morgenthaler
Publié dans: AIAA Scitech 2020 Forum, 2020, ISBN 978-1-62410-595-1
Éditeur: American Institute of Aeronautics and Astronautics
DOI: 10.2514/6.2020-0416

Estimating Both Reflection Coefficients of 2×2 Linear Hyperbolic Systems with Single Boundary Measurement

Auteurs: Nils Christian A. Wilhelmsen, Florent Di Meglio
Publié dans: 2020 59th IEEE Conference on Decision and Control (CDC), 2020, Page(s) 658-665, ISBN 978-1-7281-7447-1
Éditeur: IEEE
DOI: 10.1109/cdc42340.2020.9304413

The influence of the learning data on the reduced order model of laminar non-premixed flames

Auteurs: Nicole Lopes Junqueira Luis Fernando Figueira da Silva , Louise da Costa Ramos , Igor Braga de Paula
Publié dans: 26 International Congress of Mechanical Engineering, 2021
Éditeur: 26 International Congress of Mechanical Engineering
DOI: 10.26678/abcm.cobem2021.cob2021-0110

Assimilation of Experimental Data to Create a Quantitatively Accurate Reduced-Order Thermoacoustic Model

Auteurs: Francesco Garita; Hans Yu; Matthew P. Juniper
Publié dans: Numéro 4, 2021, ISSN 1528-8919
Éditeur: Journal of Engineering for Gas Turbines and Power
DOI: 10.31224/osf.io/8bmaz

Ongoing Development of Non-reflective Boundary Conditions for Euler and Navier-Stokes Equations via the Discontinuous Galerkin Framework

Auteurs: Edmond Shehadi, Edwin van der Weide
Publié dans: AIAA Scitech 2021 Forum, 2021, ISBN 978-1-62410-609-5
Éditeur: American Institute of Aeronautics and Astronautics
DOI: 10.2514/6.2021-1660

ASSIMILATION OF EXPERIMENTAL DATA TO CREATE A QUANTITATIVELY-ACCURATE REDUCED ORDER THERMOACOUSTIC MODEL

Auteurs: Garita, F., Yu, H., & Juniper, M.
Publié dans: Proceedings of the ASME Turbo Expo 2020: Turbine Technical Conference and Exposition, 2020
Éditeur: ASME Turbo Expo 2020

Influence of Hole-to-Hole Interaction on the Acoustic Behavior of Multi-Orifice Perforated Plates

Auteurs: Alireza Javareshkian, Alexis Dancelme, Hongyu Chen, Thomas Sattelmayer
Publié dans: 2021
Éditeur: Journal of Engineering for Gas Turbines and Power
DOI: 10.1115/gt2021-58535

Peer reviewed articles (11)

Improved color-gradient method for lattice Boltzmann modeling of two-phase flows

Auteurs: T. Lafarge; P. Boivin; N. Odier; B. Cuenot
Publié dans: EISSN: 1089-7666, Numéro 1, 2021, ISSN 1527-2435
Éditeur: Physics of Fluids
DOI: 10.1063/5.0061638

Modeling of Pulsating Inverted Conical Flames: a Numerical Instability Analysis

Auteurs: L. C. Ramos, L. F. Figueira da Silva, F. Di Meglio, V. Morgenthaler
Publié dans: Combustion Theory and Modeling, 2021, ISSN 1364-7830
Éditeur: Institute of Physics Publishing

Influence of an Oscillating Airflow on the PrefilmingAirblast Atomization Process

Auteurs: Thomas Christou Björn Stelzner Nikolaos Zarzalis
Publié dans: Atomization and Sprays, 2021, ISSN 1936-2684
Éditeur: Atomization and Sprays
DOI: 10.5445/ir/1000132623

Modeling of the nonlinear flame response of a Bunsen-type flame via multi-layer perceptron

Auteurs: Nilam Tathawadekar, Nguyen Anh Khoa Doan, Camilo F. Silva, Nils Thuerey
Publié dans: Proceedings of the Combustion Institute, 2020, ISSN 1540-7489
Éditeur: Combustion Institute
DOI: 10.1016/j.proci.2020.07.115

Numerical study of multicomponent spray flame propagation

Auteurs: Varun Shastry Quentin Cazeres Bastien Rochette Eleonore Riber Bénédicte Cuenot
Publié dans: Proceedings of the Combustion Institute, 2019, ISSN 1540-7489
Éditeur: Combustion Institute
DOI: 10.1016/j.proci.2020.07.090

Stabilization mechanisms of CH4 premixed swirled flame enriched with a non-premixed hydrogen injection

Auteurs: Laera, D., Agostinelli, P. W., Selle, L., Caz res, Q., Oztarlik, G., Schuller, T., Gicquel, L., & Poinsot
Publié dans: Proceedings of the Combustion Institute, 2020, ISSN 1540-7489
Éditeur: Combustion Institute

Impact of wall heat transfer in Large Eddy Simulation of flame dynamics in a swirled combustion chamber

Auteurs: P.W.Agostinelli D.Laera I.Boxx L.Gicquel T.Poinsotd
Publié dans: Combustion and Flame, 2021, ISSN 0010-2180
Éditeur: Elsevier BV
DOI: 10.1016/j.combustflame.2021.111728

Reducing Uncertainty in the Onset of Combustion Instabilities Using Dynamic Pressure Information and Bayesian Neural Networks

Auteurs: Michael McCartney, Ushnish Sengupta, Matthew Juniper
Publié dans: Journal of Engineering for Gas Turbines and Power, 2021, ISSN 0742-4795
Éditeur: American Society of Mechanical Engineers
DOI: 10.1115/1.4052145

Comparison of Machine Learning Algorithms in the Interpolation and Extrapolation of Flame Describing Functions

Auteurs: Michael McCartney, Matthias Haeringer, Wolfgang Polifke
Publié dans: Journal of Engineering for Gas Turbines and Power, Numéro 142/6, 2020, ISSN 0742-4795
Éditeur: American Society of Mechanical Engineers
DOI: 10.1115/1.4045516

An Observer for the Electrically Heated Vertical Rijke Tube with Nonlinear Heat Release

Auteurs: Nils Christian A. Wilhelmsen, Florent Di Meglio
Publié dans: IFAC-PapersOnLine, Numéro 53/2, 2020, Page(s) 4181-4188, ISSN 2405-8963
Éditeur: IFAC-PapersOnLine
DOI: 10.1016/j.ifacol.2020.12.2461

Early detection of thermoacoustic instabilities in a cryogenic rocket thrust chamber using combustion noise features and machine learning

Auteurs: Günther Waxenegger-Wilfing, Ushnish Sengupta, Jan Martin, Wolfgang Armbruster, Justin Hardi, Matthew Juniper, Michael Oschwald
Publié dans: Chaos: An Interdisciplinary Journal of Nonlinear Science, Numéro 31/6, 2021, Page(s) 063128, ISSN 1054-1500
Éditeur: American Institute of Physics
DOI: 10.1063/5.0038817

Thesis and dissertations (1)

Ensembling geophysical models with Bayesian Neural Networks

Auteurs: Ushnish Sengupta, Matt Amos, J. Scott Hosking, Carl Edward Rasmussen, Matthew Juniper, Paul J. Young
Publié dans: Advances in Neural Information Processing Systems (NeurIPS) 2020, 2020
Éditeur: Advances in Neural Information Processing Systems (NeurIPS) 2020