A deep learning approach to wall-shear stress quantification: From numerical training to zero-shot experimental application

TL;DR

This paper introduces a deep learning model trained on numerical simulations to accurately predict wall-shear stress from velocity fields, demonstrating zero-shot application to experimental data and enabling new insights in turbulent flow analysis.

Contribution

A novel deep learning framework that predicts wall-shear stress from velocity fields, bridging numerical training and experimental application without additional training.

Findings

Accurately predicts wall-shear stress from velocity data.

Demonstrates zero-shot applicability to experimental measurements.

Verifies physical accuracy with sensor data up to Re=2000.

Abstract

The accurate quantification of wall-shear stress dynamics is of substantial importance for various applications in fundamental and applied research, spanning areas from human health to aircraft design and optimization. Despite significant progress in experimental measurement techniques and post-processing algorithms, temporally resolved wall-shear stress dynamics with adequate spatial resolution and within a suitable spatial domain remain an elusive goal. To address this gap, we introduce a deep learning architecture that ingests wall-parallel velocity fields from the logarithmic layer of turbulent wall-bounded flows and outputs the corresponding 2D wall-shear stress fields with identical spatial resolution and domain size. From a physical perspective, our framework acts as a surrogate model encapsulating the various mechanisms through which highly energetic outer-layer flow structures influence the governing wall-shear stress dynamics. The network is trained in a supervised fashion on a unified dataset comprising direct numerical simulations of statistically 1D turbulent channel and spatially developing turbulent boundary layer flows at friction Reynolds numbers ranging from 390 to 1,500. We demonstrate a zero-shot applicability to experimental velocity fields obtained from Particle-Image Velocimetry measurements and verify the physical accuracy of the wall-shear stress estimates with synchronized wall-shear stress measurements using the Micro-Pillar Shear-Stress Sensor for Reynolds numbers up to 2,000. In summary, the presented framework lays the groundwork for extracting inaccessible experimental wall-shear stress information from readily available velocity measurements and thus, facilitates advancements in a variety of experimental applications.