Final Report Summary - LIMOUSINE (Limit cycles of thermo-acoustic oscillations in gas turbine combustors)
Publishable Summary LIMOUSINE Marie Curie initial training network (ITN) project grant 214905:
Limousine is a Marie Curie ITN with a multidisciplinary initiative to strengthen the fundamental scientific work in the field of limit cycle thermo-acoustic instabilities in combustion systems. The work requires knowledge of a combination of the sciences of combustion, acoustics, mathematics, structural mechanics, materials and computation. It is motivated by the need for lean gas turbine (GT) combustion technologies and reduced emissions. The network comprised 11 partner institutions across Europe, namely five academic partners, two research institutions and four industrial partners. A total of 18 early stage researched (ESR) and 3 experienced researcher (ER) research positions were active in the network. More information can be found at http://www.utwente.nl/limousine/. GT engines are the most efficient engines for power generation at a large scale, on basis of natural gas, coal gas and sustainable fuel gases. They combine a very compact design with low maintenance and high reliability. Due to their compactness they are very suitable to respond to fast load changes necessary to stabilise the electric grid due to transients of sustainable power generated by for example solar panels and wind turbines. Their robustness, however, is critically dependent on their design and operating conditions. The major risk of failure is due to cracks in certain GT components. These are typically caused by the following chain of events. Feedback between the fuel combustion process and pressure waves leads to violent oscillations in the engine. These are termed 'thermo-acoustic instabilities'. As a consequence, particular structural components are exposed to extreme oscillating pressure loads at high temperature. This results, probably sooner than later, in crack propagation, fatigue failure and ultimately profound damage to the GT. Each time this occurs, the cost for repairs, loss of power generated and grid coordinator penalties is of the order of millions of EUR for a GT engine in a power station.
The project aim in research was to predict the characteristics and occurrence of high amplitude thermo-acoustic limit cycle combustion oscillations, mechanical vibration and the resulting location and time to failure of components by fatigue in a GT engine. Active and passive control processes were developed to allow safe operation of the GT on a variety of fuels and operating conditions. Instability is hard to avoid in the entire operating range of an engine, from idle to nominal power, or between summer and winter operation. The impact on the combustor is assessed by GT operators by the decrease in lifetime. An important research aim of LIMOUSINE was the development of models, based on engine settings, to quantify the lifetime reduction due to elevated vibration amplitudes. The research activities were projected on tasks to be performed in combination, and aligned, with the training program. The training program comprised 5 WPs. The work in these WPs was linked by means of a common object of study. In this case this was a generic combustor designed specifically for this project to explore the phenomenon of limit cycle oscillations in a cylindrical combustor with rectangular cross section and bluff body flame stabilisation. Each WP investigated specific aspects of the combustor with specific analytical, numerical or experimental tools. Different and common aspects of the combustor behaviour were studied with advanced measurement tools in five laboratories that each had an identical copy of the combustor. The studies were performed on a specified set of operating conditions. Apart from studying the combustor behaviour also the measurement tools were further developed specifically towards the target. The fellows were trained by research in the tasks that they were given in the WPs.
In LIMOUSINE three three-day and five two-day workshops were organised for all fellows and others in combination with each project meeting. These workshops provided elements of training in various scientific disciplines: fatigue at high temperature loads, combustion transients in GTs operated on new fuels, analytical methods in thermo-acoustics, experimental acoustics in syngas combustors, Use of ANSYS for combustion modelling, active control, laser diagnostics in transient combustion and Heat transfer and fluid-structure interaction in acoustic flows. In addition to the workshops, the fellows followed in-house courses of various kinds. They all performed a traineeship in industry or at the location of another academic partner, depending on the most suitable learning opportunities. This was all monitored by means of the educational plan they submitted. The planning of their thesis work and role of the supervisor is incorporated in this plan.
In the work package (WP) analytical acoustics analytical models were developed for onset and growth of a limit cycle in the project's generic combustor. In the WP on numerical acoustics, aerodynamic coupling and combustion transients numerical simulations were performed with various techniques like large eddy simulation (LES), Reynolds-averaged Navier-Stokes (RaNS) and structural analysis. Predicted were coupled combustion-structural vibration instabilities in the generic combustor. Models were developed to predict liquid fuel combustion. Experimental acoustics and combustion dynamics were studied on the LIMOUSINE generic combustor. This combustor can run in high amplitude limit cycle pressure oscillations or in stable operation, depending on operating conditions. The generic combustor was built in 6 copies and used in five laboratories, each equipped with specific high end instruments. Laser diagnostic tools, advanced heat flux sensors and multichannel microphone techniques were applied on the copies of the generic combustor in laboratories of 5 project partners, depending on local expertise. Acoustic measurement equipment and predictive models were developed with a view to application on a high pressure combustor. Active control and non-linear studies were performed analytically, numerically and on experiments with the generic combustor. Fatigue and heat transfer in limit cycle combustion oscillation was studied by means of numerical models and experimentally in the generic combustor. Numerical simulation tools were developed to predict transient heat transfer in an oscillating combusting flow and applied on the generic combustor. An experimental test was developed to explore the effect on heat transfer to the structure by a limit cycle oscillation in the project's generic combustor. A numerical model was developed to predict crack propagation, and was validated on the generic combustor. Finally a full size GT engine geometry was numerically simulated and its transient performance analysed by means of computational fluid dynamics (CFD) simulation.
The project has raised a team of 18 talented ESR fellows, supported by 3 ERs, who have been trained thoroughly in diverse aspects of physics, mathematics, design and operation of GT engine combustors. Together with their host institutions they form an expert network on this topic in Europe. A unique experimental database was created for a generic combustor in limit cycle oscillation, for model validation. The project was successful in producing good output in scientific conference and journal papers. The computational models and design tools developed in the project have attracted the attention of European GT industry already for application in their design processes. Design tools were developed to numerically predict and characteristic model parameters in combustion dynamics. Network models and analytical models are developed to predict the onset and growth of possibly fatal instabilities. Models were developed for rate of material crack growth in limit cycle vibration. These models were demonstrated on laboratory scale but also on full commercial GT engine scale. That there is enormous need for scientists with this expertise in industry follows from the fact that the pull of industry to recruit the fellows was of such strength, that two fellows decided to accept a job at Alstom, three months before finishing their thesis work. All fellows easily find jobs in the European (GT) industry after finishing their training. As a result the European power and GT industry will have 20 extremely valuable young scientists to improve the reliability of GT power generation and to reduce the emission of carbon dioxide (CO2), nitrogen oxides (NOx) and to increase the use of sustainable gaseous and liquid fuels.
Limousine is a Marie Curie ITN with a multidisciplinary initiative to strengthen the fundamental scientific work in the field of limit cycle thermo-acoustic instabilities in combustion systems. The work requires knowledge of a combination of the sciences of combustion, acoustics, mathematics, structural mechanics, materials and computation. It is motivated by the need for lean gas turbine (GT) combustion technologies and reduced emissions. The network comprised 11 partner institutions across Europe, namely five academic partners, two research institutions and four industrial partners. A total of 18 early stage researched (ESR) and 3 experienced researcher (ER) research positions were active in the network. More information can be found at http://www.utwente.nl/limousine/. GT engines are the most efficient engines for power generation at a large scale, on basis of natural gas, coal gas and sustainable fuel gases. They combine a very compact design with low maintenance and high reliability. Due to their compactness they are very suitable to respond to fast load changes necessary to stabilise the electric grid due to transients of sustainable power generated by for example solar panels and wind turbines. Their robustness, however, is critically dependent on their design and operating conditions. The major risk of failure is due to cracks in certain GT components. These are typically caused by the following chain of events. Feedback between the fuel combustion process and pressure waves leads to violent oscillations in the engine. These are termed 'thermo-acoustic instabilities'. As a consequence, particular structural components are exposed to extreme oscillating pressure loads at high temperature. This results, probably sooner than later, in crack propagation, fatigue failure and ultimately profound damage to the GT. Each time this occurs, the cost for repairs, loss of power generated and grid coordinator penalties is of the order of millions of EUR for a GT engine in a power station.
The project aim in research was to predict the characteristics and occurrence of high amplitude thermo-acoustic limit cycle combustion oscillations, mechanical vibration and the resulting location and time to failure of components by fatigue in a GT engine. Active and passive control processes were developed to allow safe operation of the GT on a variety of fuels and operating conditions. Instability is hard to avoid in the entire operating range of an engine, from idle to nominal power, or between summer and winter operation. The impact on the combustor is assessed by GT operators by the decrease in lifetime. An important research aim of LIMOUSINE was the development of models, based on engine settings, to quantify the lifetime reduction due to elevated vibration amplitudes. The research activities were projected on tasks to be performed in combination, and aligned, with the training program. The training program comprised 5 WPs. The work in these WPs was linked by means of a common object of study. In this case this was a generic combustor designed specifically for this project to explore the phenomenon of limit cycle oscillations in a cylindrical combustor with rectangular cross section and bluff body flame stabilisation. Each WP investigated specific aspects of the combustor with specific analytical, numerical or experimental tools. Different and common aspects of the combustor behaviour were studied with advanced measurement tools in five laboratories that each had an identical copy of the combustor. The studies were performed on a specified set of operating conditions. Apart from studying the combustor behaviour also the measurement tools were further developed specifically towards the target. The fellows were trained by research in the tasks that they were given in the WPs.
In LIMOUSINE three three-day and five two-day workshops were organised for all fellows and others in combination with each project meeting. These workshops provided elements of training in various scientific disciplines: fatigue at high temperature loads, combustion transients in GTs operated on new fuels, analytical methods in thermo-acoustics, experimental acoustics in syngas combustors, Use of ANSYS for combustion modelling, active control, laser diagnostics in transient combustion and Heat transfer and fluid-structure interaction in acoustic flows. In addition to the workshops, the fellows followed in-house courses of various kinds. They all performed a traineeship in industry or at the location of another academic partner, depending on the most suitable learning opportunities. This was all monitored by means of the educational plan they submitted. The planning of their thesis work and role of the supervisor is incorporated in this plan.
In the work package (WP) analytical acoustics analytical models were developed for onset and growth of a limit cycle in the project's generic combustor. In the WP on numerical acoustics, aerodynamic coupling and combustion transients numerical simulations were performed with various techniques like large eddy simulation (LES), Reynolds-averaged Navier-Stokes (RaNS) and structural analysis. Predicted were coupled combustion-structural vibration instabilities in the generic combustor. Models were developed to predict liquid fuel combustion. Experimental acoustics and combustion dynamics were studied on the LIMOUSINE generic combustor. This combustor can run in high amplitude limit cycle pressure oscillations or in stable operation, depending on operating conditions. The generic combustor was built in 6 copies and used in five laboratories, each equipped with specific high end instruments. Laser diagnostic tools, advanced heat flux sensors and multichannel microphone techniques were applied on the copies of the generic combustor in laboratories of 5 project partners, depending on local expertise. Acoustic measurement equipment and predictive models were developed with a view to application on a high pressure combustor. Active control and non-linear studies were performed analytically, numerically and on experiments with the generic combustor. Fatigue and heat transfer in limit cycle combustion oscillation was studied by means of numerical models and experimentally in the generic combustor. Numerical simulation tools were developed to predict transient heat transfer in an oscillating combusting flow and applied on the generic combustor. An experimental test was developed to explore the effect on heat transfer to the structure by a limit cycle oscillation in the project's generic combustor. A numerical model was developed to predict crack propagation, and was validated on the generic combustor. Finally a full size GT engine geometry was numerically simulated and its transient performance analysed by means of computational fluid dynamics (CFD) simulation.
The project has raised a team of 18 talented ESR fellows, supported by 3 ERs, who have been trained thoroughly in diverse aspects of physics, mathematics, design and operation of GT engine combustors. Together with their host institutions they form an expert network on this topic in Europe. A unique experimental database was created for a generic combustor in limit cycle oscillation, for model validation. The project was successful in producing good output in scientific conference and journal papers. The computational models and design tools developed in the project have attracted the attention of European GT industry already for application in their design processes. Design tools were developed to numerically predict and characteristic model parameters in combustion dynamics. Network models and analytical models are developed to predict the onset and growth of possibly fatal instabilities. Models were developed for rate of material crack growth in limit cycle vibration. These models were demonstrated on laboratory scale but also on full commercial GT engine scale. That there is enormous need for scientists with this expertise in industry follows from the fact that the pull of industry to recruit the fellows was of such strength, that two fellows decided to accept a job at Alstom, three months before finishing their thesis work. All fellows easily find jobs in the European (GT) industry after finishing their training. As a result the European power and GT industry will have 20 extremely valuable young scientists to improve the reliability of GT power generation and to reduce the emission of carbon dioxide (CO2), nitrogen oxides (NOx) and to increase the use of sustainable gaseous and liquid fuels.