Periodic Reporting for period 4 - AFIRMATIVE (Acoustic-Flow Interaction Models for Advancing Thermoacoustic Instability prediction in Very low Emission combustors)
Periodo di rendicontazione: 2022-12-01 al 2023-11-30
Gas turbines are an essential ingredient in the long-term energy and aviation mix. However, their combustors are susceptible to damaging thermoacoustic instability, caused by a two-way coupling between unsteady combustion and acoustic waves. Strategies for reducing emissions tend to make instability more likely. Computational methods which can predict and design out thermoacoustic instability are needed: an efficient approach is to couple analytical treatment of the acoustic waves with high-fidelity simulations of the more complex flame. The AFIRMATIVE project has overhauled acoustic models across the entirety of the combustor, from the inlet burner flow, through the main combustor flowfield and attached dampers to the turbine inlet. The new models significantly enhance ability to capture flow-fields of industrial complexity, and their combination with optimisation and machine learning tools has accelerated the pace of exploring design and fuel changes.
An overview of results is as follows:
- Acoustic models for burners and dampers (WP2). We derived the first acoustic models for bluff body burners (Su et al. 2021), new models for angled openings (Su et al. 2022) and new models for heat exchanger tubes (Surendran et al., 2022). Probably our most important achievement was on acoustic-flow interactions for round holes, these being the baseline geometry for burners and dampers. Here we showed that their aeroacoustics was highly sensitive to inlet geometry shaping (Guzman et al, 2019 as well as conference work by Hirschberg et al., 2022), building up to the first geometry optimisation specifically for aeroacoustic damping (Guzman & Morgans 2024). Given that many proposed hydrogen burners/injectors have the form of holes, this is more directly relevant to burners than we could have foreseen at the start of the project and remains also highly relevant to damper optimisation. A new PhD studentship (started 2023) is taking up these exciting findings and further ideas have fed into a Marie Sklodowska Curie ITN proposal. Our work modelling the acoustics of holes and heat exchanger tubes also enabled collaborations with experimentalists at KTH, Sweden and TU Eindhoven, Netherlands.
- Acoustic models for axially varying flows (WP1). We derived a suite of acoustic models, using both WKB-methods as planned, and also Magnus-expansion methods. The new models account for axial variations in both temperature and cross-sectional area (Yeddula & Morgans 2021, Yeddula et al, 2021) as well as heat transfer (Yeddula et al, 2022) and (to be published) wall-friction and in both longitudinal and annular shaped combustors (Yeddula PhD thesis 2022). We also provided new understanding into acoustic damping at area increases (Gaudron et al. 2023), where flow separation causes mean flow losses.
- Acoustic generation at the downstream of the combustor (WP3). We made real progress modelling the combustor termination used in many lab experiments – an orifice plate. We derived a new form of acoustic analogy to study this (Yang et al, 2020) which shed new light on entropy noise generation at both the contraction (Yang et al, 2021) and expansion (Yang et al, 2020). For the acoustic generated at a blade row, we overcame a long-standing challenge in the field, using Rapid Distortion Theory to derive the first validated model for entropy noise generated by interaction with an isolated blade (Guzman et al, 2022). We have extended this to a cascade of blades (published only in conference form so far).
- We applied computational thermoacoustic predictions to complex experimental rigs (WP5), providing insights into measured behaviours (Han et al., 2019a, Han et al, 2019b). We further developed data-driven methods for predicting thermoacoustic stability (Gaudron & Morgans 2022), for optimising the geometries of burners and dampers (Guzman & Morgans 2024) and for predicting the effect of hydrogen enrichment on the flame response (under review).
Dissemination has been through a large number of conference/workshop invited talks, standard conference papers, journal publications and invited seminars, the latter across France, Sweden, Spain, UK, USA, Switzerland etc. In addition to this, many dissemination events specifically with industry have taken place, such that 3 part-industry-funded PhD studentships continue through the project end or start soon afterwards, ensuring industrial exploitation.
- Acoustic models for burners and dampers (WP2). We derived the first acoustic models for bluff body burners (Su et al. 2021), new models for angled openings (Su et al. 2022) and new models for heat exchanger tubes (Surendran et al., 2022). Probably our most important achievement was on acoustic-flow interactions for round holes, these being the baseline geometry for burners and dampers. Here we showed that their aeroacoustics was highly sensitive to inlet geometry shaping (Guzman et al, 2019 as well as conference work by Hirschberg et al., 2022), building up to the first geometry optimisation specifically for aeroacoustic damping (Guzman & Morgans 2024). Given that many proposed hydrogen burners/injectors have the form of holes, this is more directly relevant to burners than we could have foreseen at the start of the project and remains also highly relevant to damper optimisation. A new PhD studentship (started 2023) is taking up these exciting findings and further ideas have fed into a Marie Sklodowska Curie ITN proposal. Our work modelling the acoustics of holes and heat exchanger tubes also enabled collaborations with experimentalists at KTH, Sweden and TU Eindhoven, Netherlands.
- Acoustic models for axially varying flows (WP1). We derived a suite of acoustic models, using both WKB-methods as planned, and also Magnus-expansion methods. The new models account for axial variations in both temperature and cross-sectional area (Yeddula & Morgans 2021, Yeddula et al, 2021) as well as heat transfer (Yeddula et al, 2022) and (to be published) wall-friction and in both longitudinal and annular shaped combustors (Yeddula PhD thesis 2022). We also provided new understanding into acoustic damping at area increases (Gaudron et al. 2023), where flow separation causes mean flow losses.
- Acoustic generation at the downstream of the combustor (WP3). We made real progress modelling the combustor termination used in many lab experiments – an orifice plate. We derived a new form of acoustic analogy to study this (Yang et al, 2020) which shed new light on entropy noise generation at both the contraction (Yang et al, 2021) and expansion (Yang et al, 2020). For the acoustic generated at a blade row, we overcame a long-standing challenge in the field, using Rapid Distortion Theory to derive the first validated model for entropy noise generated by interaction with an isolated blade (Guzman et al, 2022). We have extended this to a cascade of blades (published only in conference form so far).
- We applied computational thermoacoustic predictions to complex experimental rigs (WP5), providing insights into measured behaviours (Han et al., 2019a, Han et al, 2019b). We further developed data-driven methods for predicting thermoacoustic stability (Gaudron & Morgans 2022), for optimising the geometries of burners and dampers (Guzman & Morgans 2024) and for predicting the effect of hydrogen enrichment on the flame response (under review).
Dissemination has been through a large number of conference/workshop invited talks, standard conference papers, journal publications and invited seminars, the latter across France, Sweden, Spain, UK, USA, Switzerland etc. In addition to this, many dissemination events specifically with industry have taken place, such that 3 part-industry-funded PhD studentships continue through the project end or start soon afterwards, ensuring industrial exploitation.
Progress beyond state of the art and expected results can be summarised as follows:
- Acoustic models for burners and dampers (WP2). We derived the first acoustic models for bluff body burners (Su et al. 2021), new models for angled openings (Su et al. 2022) and new models for heat exchanger tubes (Surendran et al., 2022). Probably our most important achievement was on acoustic-flow interactions for round holes, these being the baseline geometry for burners and dampers. Here we showed that their aeroacoustics was highly sensitive to inlet geometry shaping (Guzman et al, 2019 as well as conference work by Hirschberg et al., 2022), building up to the first geometry optimisation specifically for aeroacoustic damping (Guzman & Morgans 2024). Given that many proposed hydrogen burners/injectors have the form of holes, this is more directly relevant to burners than we could have foreseen at the start of the project and remains also highly relevant to damper optimisation. A new PhD studentship (started 2023) is taking up these exciting findings and further ideas have fed into a Marie Sklodowska Curie ITN proposal. Our work modelling the acoustics of holes and heat exchanger tubes also enabled collaborations with experimentalists at KTH, Sweden and TU Eindhoven, Netherlands.
- Acoustic models for axially varying flows (WP1). We derived a suite of acoustic models, using both WKB-methods as planned, and also Magnus-expansion methods. The new models account for axial variations in both temperature and cross-sectional area (Yeddula & Morgans 2021, Yeddula et al, 2021) as well as heat transfer (Yeddula et al, 2022) and (to be published) wall-friction and in both longitudinal and annular shaped combustors (Yeddula PhD thesis 2022). We also provided new understanding into acoustic damping at area increases (Gaudron et al. 2023), where flow separation causes mean flow losses.
- Acoustic generation at the downstream of the combustor (WP3). We made real progress modelling the combustor termination used in many lab experiments – an orifice plate. We derived a new form of acoustic analogy to study this (Yang et al, 2020) which shed new light on entropy noise generation at both the contraction (Yang et al, 2021) and expansion (Yang et al, 2020). For the acoustic generated at a blade row, we overcame a long-standing challenge in the field, using Rapid Distortion Theory to derive the first validated model for entropy noise generated by interaction with an isolated blade (Guzman et al, 2022). We have extended this to a cascade of blades (published only in conference form so far).
- We applied computational thermoacoustic predictions to complex experimental rigs (WP5), providing insights into measured behaviours (Han et al., 2019a, Han et al, 2019b). We further developed data-driven methods for predicting thermoacoustic stability (Gaudron & Morgans 2022), for optimising the geometries of burners and dampers (Guzman & Morgans 2024) and for predicting the effect of hydrogen enrichment on the flame response (under review).
- Acoustic models for burners and dampers (WP2). We derived the first acoustic models for bluff body burners (Su et al. 2021), new models for angled openings (Su et al. 2022) and new models for heat exchanger tubes (Surendran et al., 2022). Probably our most important achievement was on acoustic-flow interactions for round holes, these being the baseline geometry for burners and dampers. Here we showed that their aeroacoustics was highly sensitive to inlet geometry shaping (Guzman et al, 2019 as well as conference work by Hirschberg et al., 2022), building up to the first geometry optimisation specifically for aeroacoustic damping (Guzman & Morgans 2024). Given that many proposed hydrogen burners/injectors have the form of holes, this is more directly relevant to burners than we could have foreseen at the start of the project and remains also highly relevant to damper optimisation. A new PhD studentship (started 2023) is taking up these exciting findings and further ideas have fed into a Marie Sklodowska Curie ITN proposal. Our work modelling the acoustics of holes and heat exchanger tubes also enabled collaborations with experimentalists at KTH, Sweden and TU Eindhoven, Netherlands.
- Acoustic models for axially varying flows (WP1). We derived a suite of acoustic models, using both WKB-methods as planned, and also Magnus-expansion methods. The new models account for axial variations in both temperature and cross-sectional area (Yeddula & Morgans 2021, Yeddula et al, 2021) as well as heat transfer (Yeddula et al, 2022) and (to be published) wall-friction and in both longitudinal and annular shaped combustors (Yeddula PhD thesis 2022). We also provided new understanding into acoustic damping at area increases (Gaudron et al. 2023), where flow separation causes mean flow losses.
- Acoustic generation at the downstream of the combustor (WP3). We made real progress modelling the combustor termination used in many lab experiments – an orifice plate. We derived a new form of acoustic analogy to study this (Yang et al, 2020) which shed new light on entropy noise generation at both the contraction (Yang et al, 2021) and expansion (Yang et al, 2020). For the acoustic generated at a blade row, we overcame a long-standing challenge in the field, using Rapid Distortion Theory to derive the first validated model for entropy noise generated by interaction with an isolated blade (Guzman et al, 2022). We have extended this to a cascade of blades (published only in conference form so far).
- We applied computational thermoacoustic predictions to complex experimental rigs (WP5), providing insights into measured behaviours (Han et al., 2019a, Han et al, 2019b). We further developed data-driven methods for predicting thermoacoustic stability (Gaudron & Morgans 2022), for optimising the geometries of burners and dampers (Guzman & Morgans 2024) and for predicting the effect of hydrogen enrichment on the flame response (under review).