Multi-Physics Hypersonic Simulation by Adapting Numerical Algorithms from Computer Graphics

Project description

There are currently no rapid and robust purpose-designed analysis algorithms for multi-physics hypersonic simulation that are aligned with 21st century meta-modelling techniques. Current state of the art computer graphics algorithms could offer much faster engineering analysis. These techniques are robust in a variety of situations and computationally-performant in resource constrained environments. These are highly desirable traits for engineering analysis during a rapid turnaround design phase. We propose a PhD research program investigating and re-designing state-of-the-art techniques that include Monte Carlo methods and hex-meshing approaches from computer graphics for rapid solution of coupled multi-physics simulation problems in engineering design. We focus on the use-case of hypersonic aero-thermal-structural design because this pushes the limits of current engineering analysis. The PhD research proposed will assess if state-of-the-art hex meshing could be applied in preliminary-stage design analysis. Whether hex meshing with near-wall anisotropy could be developed for late-stage design analysis (Navier-Stokes equations) for resolution of boundary layers and aerothermal heating in hypersonic flows? Could these meshes be easily deformed in the presence of vehicle structure shape change? Finally it will also investigate if Monte-Carlo techniques can be developed into fit-for-purpose thermal analysis of hot structures over a range of complex geometries. The methodological approach adopted on this project will adopt a comparative evaluation framework that integrates literature review, benchmarking, and case-study validation starting with a known simple test geometry and extended across a range of more complex geometries of vehicle structures relevant to hypersonic vehicle design. Benchmarking will be performed using standard test cases to assess accuracy and scalability of state of the art computer graphics techniques and the results of this will inform the re-imagination of algorithms to meet the accuracy needs of the use cases in preliminary hypersonic vehicle design.

Outcomes

Expected outcomes of this project will be results of benchmarking case studies, new adapted algorithms for rapid preliminary hypersonic vehicle design much like a pre-visualisation, prototype hypersonic design software implementation, 3 research publications and a collaborative research grant proposal.

Essential capabilities

computer programming, aerospace or mechanical engineering, numerical methods

Desireable capabilities

computer graphics, applied mathematics, GPU programming (eg. CUDA)

Expected qualifications (Course/Degrees etc.)

B Engineering (Hons) / B.Tech. in Aerospace or Mechanical or related discipline; or B Science / B.Tech. in Computer Science