Identification of flow field regions by Machine Learning

Saetta, Ettore; Tognaccini, Renato

doi:10.2514/6.2022-0457

Cited by 5 publications

(1 citation statement)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is performed by the definition of boundary layer and shockwave sensors discussed in [24,25]. Recently, Saetta and Tognaccini proposed a region identification based on a machine-learning algorithm independent of numerical input or thresholds [26]. Finally, the lift-induced drag is computed by the classical Maskell formula [27] or by indirectly subtracting viscous and spurious drag to the near-field value.…”

Section: Far-field Force Methodsmentioning

confidence: 99%

Drag Reduction by Wingtip-Mounted Propellers in Distributed Propulsion Configurations

Minervino

Andreutti

Russo

et al. 2022

Fluids

Self Cite

View full text Add to dashboard Cite

Tip-mounted propellers can increase wing aerodynamic efficiency, and the concept is gaining appeal in the context of hybrid electrical propulsion for greener aviation, as smaller and lighter electrical motors can help with mitigating structural drawbacks of a tip engine installation. A numerical study of tip propeller effects on wing aerodynamics is herein illustrated, considering different power configurations of a Regional Aircraft wing. A drag breakdown analysis using far-field methods is presented for one of the most promising configurations, and a comparison between drag reductions obtained with a tip propeller or a standard winglet installation is also provided. Numerical flow simulations using Finite Volume Methods with actuator disk models are compared with results of a Vortex-Lattice Method, and far-field aerodynamic force calculation is performed for different mesh sizes. A wing drag reduction up to 6% (10%) is predicted under typical cruise (climb) flight conditions when wingtip-mounted propellers take over half of the total thrust usually provided by turbo-prop engines installed at inboard wing position. Drag breakdown analysis confirmed that the observed benefits mainly come from a reduction in the reversible drag component, increasing the effective wing span efficiency.

show abstract

Section: Far-field Force Methodsmentioning

confidence: 99%

Drag Reduction by Wingtip-Mounted Propellers in Distributed Propulsion Configurations

Minervino

Andreutti

Russo

et al. 2022

Fluids

Self Cite

View full text Add to dashboard Cite

show abstract

An Invariant Feature Space for Flow Region Identification Using Machine Learning

Otmani,

Ferrer,

Ntoukas

2023

Sustainable Aviation

View full text Add to dashboard Cite

Machine learning mesh-adaptation for laminar and turbulent flows: applications to high-order discontinuous Galerkin solvers

Tlales,

Otmani,

Ntoukas

et al. 2024

Engineering with Computers

View full text Add to dashboard Cite

We present a machine learning-based mesh refinement technique for steady and unsteady incompressible flows. The clustering technique proposed by Otmani et al. (Phys Fluids 35(2):027112, 2023) is used to mark the viscous and turbulent regions for the flow past a cylinder at $$Re=40$$ R e = 40 (steady laminar flow), at $$Re=100$$ R e = 100 (unsteady laminar flow), and at $$Re=3900$$ R e = 3900 (unsteady turbulent flow). Within this clustered region, we use high mesh resolution, while downgrading the resolution outside, to show that it is possible to obtain levels of accuracy similar to those obtained when using a uniformly refined mesh. The mesh adaptation is effective, as the clustering successfully identifies the two flow regions, a viscous/turbulent dominated region (including the boundary layer and wake) that requires high resolution and an inviscid/irrotational region, which only requires low resolution. The new clustering sensor is compared with traditional feature-based sensors (Q-criterion and vorticity based) commonly used for mesh adaptation. Unlike traditional sensors that rely on problem-dependent thresholds, our novel approach eliminates the need for such thresholds and locates the regions that require adaptation. After the initial validation using flows past cylinders, the clustering technique is applied in an engineering context to study the flow around a horizontal axis wind turbine configuration which has been tested experimentally at the Norwegian University of Science and Technology. The data used within this framework are generated using a high-order discontinuous Galerkin solver, allowing to locally refine the polynomial order (p-refinement) in each element of the clustered region. For the laminar test cases, we can reduce the computational cost by 32% (steady $$Re=40$$ R e = 40 case) and 20% (unsteady $$Re=100$$ R e = 100 case), while we get a reduction of 33% for the $$Re=3900$$ R e = 3900 turbulent case. In the context of the wind turbine, a reduction of 43% in computational cost is observed, while maintaining the accuracy.

show abstract

Identification of flow field regions by Machine Learning

Cited by 5 publications

References 18 publications

Drag Reduction by Wingtip-Mounted Propellers in Distributed Propulsion Configurations

Drag Reduction by Wingtip-Mounted Propellers in Distributed Propulsion Configurations

An Invariant Feature Space for Flow Region Identification Using Machine Learning

Machine learning mesh-adaptation for laminar and turbulent flows: applications to high-order discontinuous Galerkin solvers

Contact Info

Product

Resources

About