Cloud-based Online Access to Computational Fluid Dynamic Simulations
Project Status: running
Start Date: March 2022
End Date: March 2026
Budget (total): 16371 K€
Effort: 121.05 PY
Name: Martin Schifko
Company: Engineering Software Steyr
Engineering Software Steyr, Austria
Citkar GmbH, Germany
AUDI AG, Germany
Scientific Solutions Systems, Poland
MYB Yzilim Muhendislik A.S., Turkey
Not active yet:
KIT Valley, South-Korea
Soda System, South-Korea
Dtonic Corporation, South-Korea
Institute for Advanced Engineering, South-Korea
Dohwa Engineering Co., South-Korea
CFD simulations are used in development of new mechanical parts in the automotive, aerospace and military industries to increase efficiency.
CFD based experiments are cost effective in comparison to conventional methods and can estimate properties which cannot be empirically measured. Engineering Software Steyr GmbH (ESS) is an innovator in the field of CFD simulations and has developed new CFD capabilities (particle-based methods) in the form of four new solvers and hybridizing them to increase usability.
CFD simulations require great expertise often unavailable for small or medium enterprises (SMEs). This restricts their competitiveness as vendors for manufacturing industries. The goal of the Cloud-based Online Access to Computational Fluid Dynamic Simulations (COA-CFD) consortium shall democratize CFD simulations while advancing and promoting state-of-the-art ICT capabilities. This will enable use by nonexperts and open the field for wider audiences and markets.
Democratization will be achieved via hybridizing different solvers and by improving the human interaction interface. This will enable an ondemand cloud solution. Furthermore, a design optimization framework will be integrated into the solution for an even better user experience. Scientific Solutions Systems (SSS) and IONOS, shall design and build the
GAIA-X1 compliant cloud hardware base and configuration mechanisms to allow seamless deployment of software platforms that will withstand high workloads. Machine Learning algorithms will be developed to predict the planned utilization of cloud resources prior to the activation of simulations – pushing forward the ability to plan and to allocate
resources for high demanding tasks in a decentralized cloud environment.
COA-CFD includes demonstration of specific use cases to test the platform in real-life scenarios on Citkar’s eCargo bikes and top coating applications, trickle-bed reactors, wind turbines and water contamination.