Scholarlink

<p>Physics-informed neural networks (PINNs) are adapted and experimentally validated for rotating three-dimensional velocimetry (R3DV). Through the application of plenoptic imaging, R3DV enables the capture of time-resolved three-dimensional/three-component (3D/3C) velocity fields in the rotating frame of reference but suffers from anisotropic uncertainty due to the plenoptic camera’s limited baseline parallax. PINNs offer a means to regularize noisy measurements by enforcing physics-based constraints. Through validation against high resolution stereoscopic particle image velocimetry (stereo-PIV), we demonstrate that integrating PINNs with plenoptic particle tracking velocimetry (PTV) significantly improves estimations of depthwise motion. Across a range of Rossby numbers (2−4.5, <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\infty \)</EquationSource> <EquationSource Format="MATHML"><math> <mi>∞</mi> </math></EquationSource> </InlineEquation>), the out-of-plane velocity error is reduced by an average of 72.6<InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(\%\)</EquationSource> <EquationSource Format="MATHML"><math> <mo>%</mo> </math></EquationSource> </InlineEquation> relative to plenoptic-PTV. In addition, the influence of several design considerations and hyperparameters are explored and we identify critical adjustments to conventional PINN configurations that aid in mitigating errors inherent to plenoptic-PTV, namely: (1) balance of the objective function in proportion to theoretical plenoptic uncertainty ratios, and (2) track preconditioning. This work suggests that PINNs can extend the capabilities of plenoptic-based flow diagnostics, enabling single-camera 3D measurements with accuracies on par with high resolution multi-camera techniques under challenging imaging conditions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Adaptation and validation of physics-informed neural networks for rotating 3D velocimetry

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月3日

查看全文
  • Mahyar Moaven,
  • Abbishek Gururaj,
  • Vrishank Raghav,
  • Brian Thurow

摘要

Physics-informed neural networks (PINNs) are adapted and experimentally validated for rotating three-dimensional velocimetry (R3DV). Through the application of plenoptic imaging, R3DV enables the capture of time-resolved three-dimensional/three-component (3D/3C) velocity fields in the rotating frame of reference but suffers from anisotropic uncertainty due to the plenoptic camera’s limited baseline parallax. PINNs offer a means to regularize noisy measurements by enforcing physics-based constraints. Through validation against high resolution stereoscopic particle image velocimetry (stereo-PIV), we demonstrate that integrating PINNs with plenoptic particle tracking velocimetry (PTV) significantly improves estimations of depthwise motion. Across a range of Rossby numbers (2−4.5, \(\infty \) ∞ ), the out-of-plane velocity error is reduced by an average of 72.6 \(\%\) % relative to plenoptic-PTV. In addition, the influence of several design considerations and hyperparameters are explored and we identify critical adjustments to conventional PINN configurations that aid in mitigating errors inherent to plenoptic-PTV, namely: (1) balance of the objective function in proportion to theoretical plenoptic uncertainty ratios, and (2) track preconditioning. This work suggests that PINNs can extend the capabilities of plenoptic-based flow diagnostics, enabling single-camera 3D measurements with accuracies on par with high resolution multi-camera techniques under challenging imaging conditions.