Remodel

Context: The REMODEL project seeks to revolutionise high-performance computing (HPC) applications for engineering modelling and simulation by advancing parallel mesh generation and geometry representation. As HPC continues to evolve, particularly with the advent of Exascale supercomputers, the need for precise and efficient simulations of complex engineering systems, such as jet engines and fusion reactors, has never been greater. Despite significant advances in simulating physical phenomena, current methods for handling geometries and generating high-quality meshes in parallel have not kept pace, posing a major barrier to the full utilisation of these powerful computational resources. Challenge: The project addresses the critical challenges impeding the adoption of high-fidelity modelling in industrial applications. These challenges include the difficulty of creating accurate geometric representations that can be efficiently processed in parallel, the lack of standardised parallel CAD formats, and the need for intelligent meshing tools that can predict optimal meshes for new geometries and conditions. Existing methods are often slow, require significant user intervention, and struggle with the dynamic and multi-scale nature of real-world problems. Aims and Objectives: REMODEL aims to develop cutting-edge algorithms and tools, new knowledge, enhancing the accuracy and efficiency of simulations, to drive the next generation of HPC systems. Key objectives include: Parallel Geometry Handling: Developing innovative methods for managing complex geometric and topological structures in parallel, ensuring accurate feature recognition and preservation. Hybrid and Adaptive Meshing: Creating adaptive meshing strategies that can handle evolving geometries and enable continuous adaptation for time-dependent simulations. Machine Learning Integration: Utilizing AI to enhance mesh generation and adaptation, reducing the need for iterative processes and ensuring optimal mesh resolution from the start. Multi-Physics Integration: Facilitating seamless, high-fidelity simulations that can capture the complexities of multi-physics problems on parallel platforms. Applications and Impact: The REMODEL project will have far-reaching implications for various industries, including aerospace, automotive, energy, and space exploration. By advancing the state of parallel mesh generation, the project will enable: Enhanced Simulation Accuracy: More precise modelling of complex systems, leading to better product designs and improved performance. Reduced Computational Overhead: Efficient use of HPC resources, minimising energy consumption and computational costs. Broader Industrial Adoption: The development of standardised tools and techniques that can be easily integrated into existing workflows, making high-fidelity simulations more accessible to a wider range of industries. This will stimulate the development of new products and services that contribute to national economic growth and sustainability goals. UK Innovation Leadership: By aligning with the UK's commitment to investing in Exascale computing, REMODEL will strengthen the nation's position at the forefront of scientific and technological innovation. Furthermore, the project will orchestrate meaningful collaboration and knowledge transfer between academia and industry. The consortia amalgamates internationally leading scholars, establishing a robust partnership that will serve as a platform for knowledge exchange, socio-economic impact, and the co-creation of future strategic projects. As part of dedicated engagement and outreach commitments, REMODEL’s outputs will be available as open-source libraries to support global reach, the team will actively promote visibility of the findings, engaging with stakeholders and launching a series of workshops, conferences, and a dedicated project website