Experimental velocity data estimation for imperfect particle images using machine learning

Masaki Morimoto, Kai Fukami, Koji Fukagata

Physics of Fluids 2021 article cited by 96

Source Open-access PDF More in Physics & Astronomy

importance 3/10 partially solved Model Reduction and Neural Networks experimental fluid dynamics stage: research

Problem

Experimental particle image velocimetry often has missing regions in images due to experimental limitations, leaving gaps in measured velocity fields.

Why it matters

Incomplete velocity measurements limit experimental fluid-dynamics studies used in aerodynamics and engineering.

What the AI did

Supervised machine learning models were trained on artificial particle images (with DNS ground truth), including images with missing regions mimicking the wind tunnel setup, to estimate velocity fields.

Results

Estimated velocity fields around a square cylinder (ReD=300) were in 'great agreement' with reference DNS data under various statistical assessments; no numeric error metrics in abstract.

Impact

Demonstrates ML can estimate velocity fields from imperfect particle images, aiding experimental fluid mechanics.

Record

Venue	Physics of Fluids
Published	2021
Work type	article
Authors	Masaki Morimoto, Kai Fukami, Koji Fukagata
Citations	96 (OpenAlex, at import time)
References	194
Model or method	not stated in source
Organization	not stated in source
Primary source	https://doi.org/10.1063/5.0060760
Open access	green
Keywords	particle image velocimetry, machine learning, velocity estimation, missing data, DNS

Importance rating

3 / 10

incremental / niche

Improves a specific experimental measurement technique (PIV) in fluid dynamics; abstract frames it as aiding experimental flow studies, no broad scale claimed.

Basis: inferred. Judged from the source text, which gives no explicit scale. Stated scale: not stated in source. Method and limits.

More in Physics & Astronomy

#	Paper	Year	Citations	Importance
1	Magnetic control of tokamak plasmas through deep reinforcement learning Jonas Degrave, Federico Felici, Jonas Buchli, Michael Neunert, Brendan...	2022	739	7
2	Solving the quantum many-body problem with artificial neural networks Giuseppe Carleo, Matthias Troyer	2017	2,205	7
3	Learning high-accuracy error decoding for quantum processors Johannes Bausch, Andrew Senior, Francisco J. H. Heras, Thomas Edlich,...	2024	90	7
4	Optimizing Quantum Error Correction Codes with Reinforcement Learning Hendrik Poulsen Nautrup, Nicolas Delfosse, Vedran Dunjko, Hans J. Brie...	2019	158	7
5	Machine-learning-assisted correction of correlated qubit errors in a topological code Paul Baireuther, Thomas E. O’Brien, Brian Tarasinski, C. W. J. Beenakk...	2018	141	7