Multi-scale Navier-Stokes analysis of geometrically resolved erosion of wind turbine blade leading edges

Abstract

A multi-scale computational fluid dynamics analysis of wind turbine blade leading edge erosion is presented. The test case is a large set of eroded blade sections. These are obtained by fitting the resolved eroded leading edge geometry of the outboard part of a multi-megawatt offshore wind turbine to the NACA633-618 airfoil. The erosion geometry measured by a blade laser scan is geometrically resolved in the aerodynamic simulations, whereas the aerodynamic effects of unresolved lower-amplitude scales are accounted for by using distributed surface roughness models. The simulations also account for the laminar-to-turbulent transition of the blade boundary layers with and without distributed roughness. An existing semi-empirical model and simulations of the nominal airfoil enable one to estimate the roughness level needed to trip leading edge boundary layer transition at the considered Reynolds number of 9 million. It is found that a) the mean roughness heights of the observed geometry perturbations are well above the critical roughness height, and b) consideration of either large or small erosion scales in isolation results in underestimating the airfoil performance impairment.

Assessing the Progression of Wind Turbine Energy Yield Losses Due to Blade Erosion by Resolving Damage Geometries from Lab Tests and Field Observations

“…As in previous works [17,22], the constant 𝑎 1 in the expression of the SST eddy viscosity is changed from 0.31 to 0.28 in all NACA63 3 -618 airfoil analyses below, to improve the agreement between computed and measured force coefficients of this airfoil [15].…”

“…To date, the actual value of AEP lost to LEE is uncertain, with recently reported estimates for wind farms ranging from 0.5 to 8% [3][4][5][6][7]. The large scatter of the reported AEP losses is partly due to the fact that this quantity depends on the specific geometry of the blade surface perturbation As erosion progresses and the perturbations of the LE geometry approach or exceed the BL thickness, additional aerodynamic losses occur, which may vary significantly with the specific LEE damage [15]. These phenomena contribute, with strength depending on the erosion severity, to premature separation of the BL on the airfoil upper side at AoAs where the nominal airfoil has a clean flow [7,16,17], with consequent reduction of the lift and further increase of the drag.…”

“…So far, most CFD investigations into the aerodynamics of eroded wind turbine blades relied on fairly simple representations of LEE geometry. A recent study by Ortolani et al [15] presented the first assessment of the airfoil performance reduction caused by LEE making use of the eroded LE geometry from the 3D laser scan of a utilityscale offshore turbine blade. Their RANS simulations adopted a scale separation approach to the analysis of multi-scale roughness, resolving the erosion geometry of the scan and modeling the impact of roughness scales unresolved by the scan with a distributed surface roughness model.…”

“…The investigations of the present study are partly based on the eroded LE laser scan of [15], and the main objectives of the work are: (a) to assess the progression of power and AEP losses as LEE severity increases using the scale separation CFD analysis of erosioninduced LE roughness, (b) to investigate the dependence of these losses on the resolved erosion geometry, which, in turn, depends also on the structural properties of the LE material, and (c) to investigate the sensitivity of the blade performance degradation and turbine AEP loss to the level of the equivalent sand grain roughness, a parameter affected by uncertainty due to its dependence on the typically unknown geometry of the unresolved roughness scales.…”

CFD Modeling of Wind Turbine Blades with Eroded Leading Edge

“…Later, Wang et al [41] undertook CFD simulations of wind turbine blades with LEE, showing that the degree of erosion significantly impacts flow separation, aerodynamic coefficients, and wind turbine power output. Finally, Ortolani et al [42] presented RANS analyses of the NACA 63 3 − 618 airfoil with LEE. The erosion pattern studied are extracted from real offshore wind turbine after about six years of operation.…”