Evolution of wall pressure and wall shear stress under adverse pressure gradients from attached to separated flows

Abstract

At high angles of attack (AoA), the boundary layer near the trailing edge of a wind turbine blade can experience flow separation, leading to detrimental performances for the wind park and an increase in noise emission. Too often, many datasets suffer from limitations that prevent acquiring the full range of frequencies associated with flow separation, ranging from the low- frequency breathing up to small scales of backflow. Our objective in this paper is to cover low-, mid-, and high-frequency activities using comprehensive instrumentation to yield a novel database with full spectral resolution that AI algorithms can later use for training. In the present contribution, we present an experimental family of ZPG and APG flow conditions that were investigated in the AWB wind tunnel in DLR Braunschweig. A new database was produced in order to obtain a progressive evolution of wall-pressure and wall-shear stress from attached to separated flows. The presence of a movable NACA-0012 profile over a flat plate allowed us to cover an elongated streamwise direction over the separated region. A comprehensive set of pressure transducers enabled us to resolve both low and high frequency ranges, and synchronously recorded the shear stress mean and fluctuating values. Oil film visualizations and wall quantities have shown a consistent portrait of such APG flows, with an emphasis on the low frequency unsteadiness. A noise pollution filtering technique was applied in the frequency domain and revealed promising results in order to resolve the breathing motion of a turbulent separation bubble in spite of known experimental limitations. In the future, additional work, including machine learning training and monitoring, will be necessary to tackle the challenge of flow separation detection and mitigation in wind energy.