EPSRC DLA Studentship - Machine-Learning-Driven Inverse Design for 3D Turbomachinery

Project Overview

Iterative RANS-based CFD design is approaching its practical limits.

While high-fidelity simulation remains essential, the repeated geometry-CFD-evaluation loop that dominates turbomachinery design is increasingly the primary bottleneck. Within the next decade, this workflow will be largely replaced by data-driven inverse design models, enabled by the growing availability of high-quality simulation data.

This PhD aims to fundamentally change the turbomachinery design process by replacing iterative RANS-based CFD loops with regression-driven inverse design. Instead of repeatedly modifying geometry and re-running CFD, the research will develop models that map aerodynamic requirements directly to 3D blade geometries and associated flow fields.

The project builds on a proven 2D inverse design framework, already adopted by major industrial partners including Rolls-Royce, and extends it to full 3D turbomachinery configurations. By leveraging inverse design, the approach deliberately stays within physically sensible design spaces avoiding the need to learn every pathological flow scenario and making machine learning both efficient and reliable.

The ultimate goal is to retain the fidelity and robustness of RANS-based design while reducing design iteration times by over 90%.

Expected Outcomes

• A validated 3D inverse design tool capable of producing accurate geometries and RANS-consistent flow fields up to 100x faster than conventional CFD-based workflows.
• Demonstration of 90% reduction in design iteration time for real turbomachinery components.
• A scalable framework for data-driven design acceleration applicable to compressors and turbines

Ideal Candidate Profile

This project is aimed at candidates who want to work at the intersection of aerodynamics, CFD, and machine learning, with a strong emphasis on engineering realism.

The ideal candidate will have:

• A strong background in aerodynamics, fluid mechanics, or turbomachinery
• Experience with RANS CFD (commercial or in-house solvers)
• Strong programming and software skills (Python essential; C++/Fortran beneficial)
• Interest in machine-learning-based regression and surrogate modelling
• Motivation to challenge established CFD-driven design workflows using data-driven methods

You do not need to be an expert in machine learning at the start of the PhD, but you should be keen to develop rigorous ML and software engineering skills alongside deep aerodynamic understanding. 

Why This PhD?

• Work on a problem with clear industrial relevance and near-term impact
• Combine physics-based CFD with modern data-driven modelling
• Develop skills directly applicable to aerospace, energy, and advanced engineering industries

EPSRC DLA studentships are available for eligible home students and a limited number of international students.

Applicants should have (or expect to obtain by the start date) a high 2.1 degree, in a relevant engineering or science discipline.

Applications should be submitted via the University of Cambridge Applicant Portal (via the above 'Apply' button), with Dr Chris Clark identified as the potential supervisor. Applications may close early if the position is filled before the advertised date. Please note there is a £20 application fee attached to using the Cambridge Applicant Portal.

The University actively supports equality, diversity and inclusion and encourages applications from all sections of society.

Advert information