Designing jet engines usually involves a trade-off between fast but blurry simulations and slow but crystal-clear ones. This project creates a smart hybrid system that uses a few high-detail snapshots to teach a fast AI model how to be more accurate. It is like using a few high-resolution photos to help a sketch artist draw a perfect map much faster.
Traditional jet engine design relies on simplified models that lack the precision needed for ultra-fuel-efficient engines. This creates a bottleneck that increases development costs and slows down the transition to green aviation.
An automated design workflow combining high-order simulations with machine learning. This includes a multi-fidelity optimization chain tested on compressor cascades.
If you are a jet engine OEM dealing with slow development cycles for fuel-efficient engines — this project developed a hybrid simulation pipeline that reduces the need for expensive physical testing. This allows for the creation of propulsion systems that help meet 2050 climate-neutral goals.
If you are a CAE software provider dealing with the limitations of standard RANS models — this project developed a multi-fidelity optimization chain using machine learning. This enables the delivery of higher aerodynamic accuracy without the massive computational cost of full simulations.
If you are a sustainable propulsion startup dealing with limited budgets for low-TRL testing — this project developed an automated SRS design workflow. This reduces the reliance on physical prototypes to validate new engine concepts for 2035 emission targets.
It reduces costs by shortening development cycles and decreasing the number of expensive physical tests required through higher simulation accuracy. Based on available project data, it replaces low-TRL testing with high-fidelity digital models.
Yes, the project is currently preparing to validate these methods on full-scale 3D fan-stages. The goal is to integrate these tools into standard industrial workflows.
Based on available project data, specific licensing terms are not provided, but the project involves a consortium of 11 partners including 3 industry members.
The design process is specifically established for modern CPU and GPU hardware to ensure it meets industrial turnaround time requirements.
The technology aims to reduce GHG emissions by 2035 and support the EU's target to be climate-neutral by 2050.
The project is led by DLR and features a strong mix of 11 partners across 7 countries. With an industry ratio of 27% (3 companies, including 1 SME), the consortium balances academic research from 5 universities and 3 research centers with practical industrial application, ensuring the tools are compatible with real-world manufacturing needs.
Contact DLR (Deutsches Zentrum für Luft- und Raumfahrt)
Contact us to find licensing opportunities for this hybrid simulation pipeline.
