Implicit Solution Strategies on Accelerators


   School of Engineering

This project is no longer listed on FindAPhD.com and may not be available.

Click here to search FindAPhD.com for PhD studentship opportunities
Dr S Timme Prof C A Lin  No more applications being accepted  Funded PhD Project (Students Worldwide)

About the Project

This project will aim to accelerate the current solution time of linear systems of equations, arising from industry-relevant problems in computational aerodynamics using industry-relevant codes with implicit solution algorithms, by at least an order of magnitude. Rather than focussing on the performance of the nonlinear flow solver, as usually done, the project will explore linearised aerodynamics used routinely in aircraft design, which presents algebraic difficulties quite different compared with its nonlinear counterpart. Various aspects will be analysed including, but not limited to, vectorisation strategies on modern CPUs, shared and distributed solution strategies on clusters of CPU nodes as well as accelerators, and advanced communication-avoiding Krylov solvers. For this purpose, the work will make use of freely available software libraries as well as bespoke algorithms to be developed. The engineering application, once viable tools are available, will involve some of the most challenging problems in transonic aircraft aerodynamics.

Dr Timme has expertise in the relevant computational challenges encountered in aircraft aerodynamics and its multidisciplinary applications. He has been striving for several years toward establishing time-linearised methods in industrial design processes related to aerodynamic and aeroelastic analysis of large civil aircraft. In the context of aircraft loads analysis, the routine use of computational fluid dynamics remains a long-term ambition, and any gain in time-to-solution is highly beneficial in this pursuit.

Professor Lin’s research area includes flow physics and turbulence modelling, and development of efficient numerical methods. He has gained vast experience in CPU and GPU-accelerated fluid dynamics in particular applied to turbulence simulation and its physics’ realization.

This ambitious project is aimed at exploiting the next generation of computing architectures, with strong components of heterogeneity, for the state-of-the-art in computational aerodynamics applied to multidisciplinary aircraft design. While much research has been focussed to increase scalability and performance of the nonlinear flow solver, less attention is usually directed at its time-linearised and adjoint incarnations. Whereas fully nonlinear and time-accurate simulations will remain one-off strategies for a foreseeable future, particularly when applied in routine design processes often requiring many thousands of simulations in short timescales, linearised aerodynamics offer exactly what is needed; high-fidelity predictions made accessible in a plug-and-play situation. To that end, implicit solution strategies are explored on graphics processing units.

For academic enquiries please contact Dr Sebastian Timme [Email Address Removed]

For enquiries on the application process or to find out more about the Dual programme please contact Miss Hannah Fosh [Email Address Removed]

Funding Notes

This project is a part of a 4-year dual PhD programme between National Tsing Hua University (NTHU), Taiwan and the University of Liverpool, England. It is planned that students will spend time studying in each institution.

Both the University of Liverpool and NTHU have agreed to waive the tuition fees for the duration of the project. Moreover, a stipend of TWD 10,000/month will be provided to cover part of the living costs.

When applying please Quote the supervisor & project title you wish to apply for and note ‘NTHU-UoL Dual Scholarship’ when asked how you plan to finance your studies.

Where will I study?


Project supervisors

Career overview

Professor Sebastian Timme is a Professor of Aerodynamics at the University of Liverpool. He completed his PhD in 2010 and has since had secondments at DLR Göttingen in 2010 and BAE Systems Filton from 2016 to 2017. Professor Timme has published over 80 journal and conference papers focusing on computational aerodynamics, aeroelasticity, model order reduction, and machine learning. His primary research interest lies in developing fast simulation tools for the routine industrial application of computational fluid dynamics (CFD) within the multidisciplinary contexts of aircraft loads and aeroelastics. He has implemented several methods into the DLR-TAU software tool, a widely used CFD code in the European aerospace industry and academia, and has contributed to the development of the new-generation CFD code CODA.


Research interests

Professor Timme''s research focuses on computational aerodynamics and aeroelasticity, with a particular emphasis on fast simulation tools for the routine industrial application of computational fluid dynamics (CFD) in the context of multidisciplinary aircraft loads and aeroelastics. He has published over 80 journal and conference papers on topics including model order reduction and machine learning. Professor Timme has implemented methods into the DLR-TAU software tool, a widely used CFD code in the European aerospace industry and academia, and has contributed to the development of the new-generation CFD code CODA.

View Professor Sebastian Timme's profile 

Prof C A Lin's profile is coming soon

View other supervisors at University of Liverpool