Ajit Panesar
Top university
Just Landed
Posted Yesterday
PhD Studentship in Aeronautics: Generative Neural-Operator Surrogates for Real-Time Multiphysics Design (AE0094) Imperial College London in United Kingdom
Degree Level
PhD
Field of study
Computer Science
Funding
Available
Deadline
Jan 4, 2027
Country
United Kingdom
University
Imperial College London
Apply for this scholarship
Keywords
Suggested scholarships
About this position
This PhD project in the Department of Aeronautics at Imperial College London explores how generative design can be combined with neural-operator surrogates to support real-time, closed-loop multiphysics design. The work targets a major engineering bottleneck: geometry generation is fast, but evaluating thermal and electromagnetic performance usually requires slow simulation cycles.
You will develop neural-operator models, including Fourier Neural Operator and DeepONet-style architectures, to learn thermal and electromagnetic performance from training data such as multifunctional metastructures. These surrogate models will then be embedded directly into a conditional generative system, such as a diffusion or transformer-based architecture, so that physics feedback can shape new designs during generation rather than after the fact.
The project focuses on coupling physical prediction with design synthesis. You will build physics constraints into the generative process, create conditioning methods that allow target performance to be specified directly, and benchmark the trade-off between surrogate fidelity and design quality. A further research question is whether one joint multiphysics operator can represent both thermal and electromagnetic behaviour more efficiently than two separate models.
Alongside computational validation, you will also work toward experimental testing where feasible, using 3D-printed prototypes to keep the research grounded in physically realisable structures. The project aims to move toward an interactive design-exploration tool with sub-second feedback, comparing the closed-loop approach against conventional solver-in-the-loop pipelines.
You will be supervised by Dr. Ajit Panesar and Prof. Rob Hewson, whose expertise includes generative design, engineering optimisation, structural metamaterials and scientific machine learning. The student will join a dynamic research environment and work with a group of researchers on this interdisciplinary computational and experimental design problem.
Funding details
Available
What's required
Candidates are expected to achieve or have achieved a First class honours MEng/MSci or higher degree (or international equivalent) in Aeronautics or Mechanical Engineering.
How to apply
Visit the advertised project page and follow the instructions there. Read the application FAQs first if you have questions about the process, and contact the PhD Administrator (Admissions) only if your query is not answered.
Make your next move.
Prepare your application
Review drafts and confirm requirements with the provider.
Professors
Prof. at Imperial College London
United Kingdom
Research Interests
Aerospace Engineering
Mechanical Engineering
Computer Science
Materials Science
Mathematics
Make your next move.
Prepare your application
Review drafts and confirm requirements with the provider.