High-Order Accurate DNS of Flow over an MTU-T161 Low Pressure Turbine Blade

Snapshot, point-probe, and time-average data from a high-fidelity computational simulation of turbulent air flow over a low pressure turbine blade. The simulation was performed using the open-source PyFR flow solver on over 5000 Nvidia K20X GPUs of the Titan supercomputer at Oak Ridge National Laboratory. The data can be used to understand complex three-dimensional unsteady air flow patterns in jet engines.

Use Cases

• Training next-generation Reynolds Averaged Navier-Stokes turbulence models via machine learning based on the high-fidelity simulation data.
• Analyzing complex three-dimensional unsteady air flow patterns over turbine blades based on the snapshot and point-probe data.
• Developing enhanced understanding for the design of greener, more fuel-efficient aircraft based on the time-averaged flow characteristics.

Strengths

• Data produced via a high-fidelity simulation using over 5000 Nvidia K20X GPUs on the Titan supercomputer.
• Simulation undertaken using the open-source PyFR flow solver under a US DOE INCITE award.
• Includes snapshot, point-probe, and time-average data types for a multi-faceted analysis.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and total dataset size are unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

PyFR project (http://www.pyfr.org), simulation run on the Titan supercomputer at Oak Ridge National Laboratory.

Collection Method

High-order accurate direct numerical simulation using the PyFR flow solver.