MS8: Machine Learning-Based Reduced Order Models for Fluid Flow Emulators and Application to Design Optimization

Amirul Khan and He Wang

Abstract

Machine Learning (ML) has emerged as a powerful tool in the development of Reduced Order Models (ROMs) for computational fluid dynamics (CFD) surrogates or emulators, particularly in the context of multidisciplinary design optimisation (MDO). The integration of ML with ROMs offers promising avenues for efficient and accurate predictions, making it well-suited for high-performance computing. In this mini-symposium, we invite contributions that reflect the rapid advancements in ML-based ROMs for the creation of CFD surrogates or emulators and their applications to MDO. We also welcome contributions that explore other ML-based methods and their applications.

Contributions can cover, but are not limited to, the following topics:

• Development and application of non-intrusive ML-based ROMs.
• Uncertainty quantification and robust design using ML-enhanced CFD emulators/surrogates.
• Case studies showcasing the application of ML-based ROMs.
• Challenges and solutions in integrating ML-based ROMs with design optimisation.
• Future directions in the integration of ML and ROMs for design optimization.
• Applications in Multidisciplinary Design Optimization:
  1. Aerospace vehicle design and aerodynamics
  2. Automotive engineering and vehicle performance optimisation
  3. Renewable energy system design and optimization
  4. Biomedical device design and fluid-structure interaction studies

This mini-symposium aims to foster discussions and collaborations among researchers, academicians, and industry professionals interested in the application of ML-based ROMs for flow emulators in design optimization. We look forward to your valuable contributions.

Amirul Khan - University of Leeds, School of Civil Engineering

Dr Amirul Khan is a lecturer in Environmental Fluid Mechanics at the School of Civil Engineering, University of Leeds. His research is primarily focused on the application of the lattice-Boltzmann method (LBM) across various areas. These include turbulent flow simulation in both indoor and outdoor environments, fluid turbulence, and turbulent dispersion. Additionally, his research interests extend to applying neural operator-based deep learning (DL) approaches to solving complex forward and inverse flow problems with applications to subsurface Physics.

He Wang - University College London, Department of Computer Science

Dr. He Wang is an Associate Professor in the Virtual Environment and Computer Graphics (VECG) group at the Department of Computer Science, University College London. He is also a Turing Fellow and an Academic Advisor at the Commonwealth Scholarship Council. He serves as an Associate Editor of Computer Graphics Forum. His current research interest is mainly in computer graphics, vision and machine learning.