PhD studentship in sensitivity analysis of thermoacoustic oscillations in gas turbine and rocket engines
Starting in October 2016 for up to 4 years.
When Yuri Gagarin was launched into orbit in 1961, the probability of a rocket blowing up on take-off was around 50%. In those days, one of the most persistent causes of failure was a violent oscillation caused by acoustics coupling with heat release in the combustion chamber. If more heat release than average occurs during moments of high pressure and less heat release than average occurs during moments of low pressure then, over a cycle, more work is done during the expansion phase than is absorbed during the compression phase, causing oscillations to grow. These thermoacoustic oscillations have caused countless rocket engine and gas turbine failures since the 1930s and have been studied extensively. Nevertheless, they are still one of the major problems facing rocket and gas turbine manufacturers, such as Rolls-Royce, today.
The ultimate goal of rocket and gas turbine manufacturers is to eliminate thermoacoustic oscillations through passive control - i.e. either by adding a passive device to an existing system or by designing a system that is naturally stable. Experiments and numerical simulations show that even small changes to a combustion chamber can significantly alter its stability. The challenge is to predict those small changes and to use those predictions to optimize the combustion chamber's design.
The aim of this project is to develop techniques that have been proven to work on small idealised systems into tools that can analyse actual industrial gas turbine and rocket engines. These tools will predict the thermoacoustic behaviour of a given system and the changes that need to be made to the shape of the gas turbine's combustion chamber in order to reduce or eliminate thermoacoustic oscillations.
The project will use adjoint-based sensitivity analysis, a technique that has been pioneered by a handful of research groups around the world, including Matthew Juniper's. This is a branch of Flow Instability (course 4A10 in CUED) in which one examines the behaviour of small perturbations to a system. Matthew Juniper's group has successfully applied adjoint-based sensitivity analysis to simple thermo-acoustic systems, laying the ground work for this project. In particular, the group has developed a 1D Finite Element model in Matlab as a test bed for these concepts in a FE framework.
This project will implement adjoint-based sensitivity analysis on a thermoacoustic system modelled with an open source FE code, Fenics (http://fenicsproject.org). The first model will be 1D in order to compare directly with the Matlab model. Subsequent models will be 2D and 3D. This will give the sensitivity of the thermoacoustic oscillations to changes in the geometry or the addition of a passive feedback device. Predictions from these models will be compared with results obtained from experiments done by other members of the research group.
The final stage of the project will be to wrap an optimization loop around the model in order to iterate towards a optimal design. This will be extremely challenging and will involve incorporating optimization techniques from other research groups. These techniques can be tested on the 1D Matlab model before being wrapped around the 2D and 3D models.
Applicants should have an excellent academic track record in a scientific, mathematical, or engineering discipline.
Some experience of object-oriented programming (e.g. Python) would be an advantage but is not essential.
Full funding is available to UK nationals only
(click here for nationality requirements and
here for qualification requirements).
To apply, please complete form CHRIS/6 (cover sheet for C.V.s) available at
http://www.admin.cam.ac.uk/offices/hr/forms/ and send it with your C.V. and a covering letter to Matthew Juniper (firstname.lastname@example.org) to arrive no later than 31 January 2016.
Engineering Department Advertisement