About HyNOISE HyNOISE is a Doctoral Network aimed to train doctoral candidates able to face current and future challenges in the prediction and control of direct and indirect noise generated by hydrogen flames.
Four specific research objectives will be pursued: 1) Unveil the dynamics and sound pressure level generated during unsteady ignition and flashback process 2) Investigate the direct broadband sound generated due to the coupling between unsteady flame dynamics and flame-flame interactions, hydrodynamic and thermo- diffusive instabilities, and turbulence.
3) Understand physical mechanisms leading to the onset of thermoacoustic instabilities coupled with different modes at either with a specific focus on high frequencies 4) Study the contribution of convective waves to combustion noise and thermoacoustic instabilities.
Multiple hydrogen-air archetypes typical of air-breathing hydrogen combustion systems in both single and annular multi-injector configurations will be investigated.
In a collaborative partnership between academic and industry partners who are major European gas turbine manufacturers, HyNOISE bring relevant expertise together and to pursue solutions for hydrogen through integration and combination of unique and advanced theoretical, high-fidelity modelling and experimental expertise.
This will lead to the development of new models and tools able to correctly predict hydrogen flame dynamics and its resulting pressure field, and, on this basis, developing control and mitigation methods to overcome current limitations in hydrogen combustion systems and design innovative, quiet, and stable combustion devices.
The technologies developed will stimulate the green transition in the power generation and transport sectors, thus contributing to the achievement of the EU's challenging decarbonization goals.
About the host organization University of Cambridge (UCAM) is one of the leading scientific research institutions worldwide and is regularly ranked in the top 5 universities.
The Engineering Department is the highest-ranking department of its type in the UK.
It is an integrated department, with several cross-disciplinary collaborations.
The department has 180 members of staff, 700 post-doctoral researchers and a similar number of graduate students.
A sizable fraction of the activity, involving 70 staff in the university, revolves around fluid mechanics, aeroacoustics and combustion, with a correspondingly large number of PhD students.
Individual Research Project(IRP) Problem Definition The overall objective is to optimize the design of hydrogen-fuelled gas turbine injectors, so that the combustors are quiet and not susceptible to thermoacoustic instabilities.
This requires a quantitatively accurate model of the thermoacoustic system, which is the combination of a hydrogen flame with an acoustic network.
This model will be obtained by systematic assimilation of data from laboratory experiments (NTNU), large eddy simulations (TUB), and industrial experiments (RRD).
This data will be assimilated into thermoacoustic models using UCAM’s adjoint-accelerated inference and optimization method (AXIOM).
The first intermediate objective is to extend
AXIOM to typical hydrogen flame geometries using data from a laboratory scale combustor at
NTNU.
Bayesian model selection (a component of
AXIOM) will be used to find the optimal component models for this simple system.
The second intermediate objective is to extend
AXIOM to industrial scale geometries using data from a real gas turbine fuel injector on an industrial test rig at RRD.
Bayesian experimental design (another component of
AXIOM) will be used to identify the experimental configurations that maximize the information that can be learned from a limited number of tests.
The third intermediate objective is to develop and test several candidates mean field flame models with TUB and then to assimilate LES data into these models with
AXIOM.
This is an ambitious objective, but the risk it generates is mitigated by TUB’s use of PINNs to achieve the same objective.
The result will be a Bayesian framework that transfers data from experiments and numerical simulations into low order models.
These models will steadily accumulate data and become more quantitatively accurate over time.
They can therefore be used for optimization and re-optimization, as more data become available in the future and as future designs change.
Research Objectives • Assimilation of experimental data from laboratory scale combustor into thermoacoustic models and flame models.
• Assimilation of experimental data from RRD industrial scale test rig into thermoacoustic models.
• Development and testing of several candidate mean field flame models with TUB.
• Assimilation of LES data into candidate mean field flame models and identification of optimal models.
• Assimilation of experimental data from RRD industrial scale test rig into flame models.
• Optimization of burner design to reduce susceptibility to thermoacoustic instability Expected Results • Detailed description of the acoustic field in premixed hydrogen combustion in canonical configurations.
• Creation of a quantitatively accurate thermoacoustic model of the RRD industrial scale test rig containing an industrial burner, able to match experimental results from the RRD rig within quantified uncertainty bounds and able to extrapolate to different configurations within predicted uncertainty bounds.
• Creation of a quantitatively accurate mean field flame model of a turbulent hydrogen flame, able to match the LES results on which it was trained to within quantified uncertainty bounds, and able to extrapolate to unseen LES data.
• Optimal burner design derived from the quantitatively accurate models obtained from outcomes 1 & 2, and a numerical tool for optimizing the design at different conditions and combustor geometries.
Secondments • Technical University of Berlin (Germany, ca.
2 months): application of developed model to improve mean fields reconstructions of non-measured quantities • ETH Zurich (Switzerland, ca.
2 months): to check capabilities of LOM to predict indirect noise The IRP will have a strong link to other IRPs from different DCs for: • Work on joint modelling and simulations • Provision of input and data