Artificial intelligence (AI) methods are developing rapidly and are gradually being applied to numerous work steps in order to solve complex problems. The analysis and processing of big data requires high computing power and scalable AI solutions. Therefore, completely new workflows must be developed from current applications, which can be efficiently executed on future high-performance computing (HPC) architectures in the Exascale era. To address these challenges, a new European-funded Center of Excellence "Research on AI- and Simulation-Based Engineering at Exascale" (CoE RAISE) was created.
The project aims to close the gap in full cycles by using forward simulation models and AI-based inverse inference models in conjunction with statistical methods to learn from current and historical data. In this context, new hardware technologies such as modular supercomputing architectures, quantum annealing and HPC prototypes are used to explore unimagined performance in data processing. CoE RAISE's European network will develop and offer best practices, support and training for industry, SMEs, universities, and HPC centers at tier 2 level, thus attracting new user communities. This will be coupled with the development of a company that will offer new services to different user communities.
Within the project, researchers from the Computer Applications in Science and Engineering (CASE) Department of the Barcelona Supercomputing Center (BSC) lead the task of developing AI for wind farm layout optimization.
“We will explore machine learning approaches to develop a model for wind turbines and include them in wind farm simulations,” said Guillaume Houzeaux, CoE RAISE partner and Physical and Numerical Modelling Group Manager at BSC. “Through this use case we will develop all necessary aspects in obtaining Exascale-ready AI solutions for real-world problems.”
About CoE RAISE
CoE RAISE is a three-year project that started on January 1, 2021 with a total budget of around € 5 million. The CoE brings together eleven full partners and two associate partners with expertise in AI and HPC: FZJ (Germany), UOI (Iceland), CYI (Cyprus), RWTH (Germany), BSC (Spain), CERN (Switzerland), CERFACS (France), Bull (France), RTU (Latvia), FM (Belgium), SAFRAN (France), ParTec (Germany) and TU Delft (the Netherlands).
RAISE aims to be an excellent engine for the further development of these technologies in Europe on an industrial and academic level and a driver for novel, interlinked AI / HPC methods. These technologies are advanced based on representative use cases that span a wide range of academic and industrial applications, such as wind power harvesting, hydrodynamics of wetting, manufacturing, physics, turbo-machinery, and aerospace.
The CoE RAISE project receives funding from the European Union’s Horizon 2020 – Research and Innovation Framework Programme H2020-INFRAEDI-2019-1 under grant agreement no. 951733.