Design of a passive flow control solution for the mitigation of vortex induced vibrations on wind turbines blade sections as a response to extreme weather events

https://doi.org/10.1016/j.seta.2023.103053Get rights and content

Highlights

  • Passive control methods mitigate ViV on wind turbines due to extreme weather events.

  • The interaction between the microcylinders and the airfoil wake reduces ViV.

  • The microcylinders size of 10% of the chord length maximizes the ViV mitigation.

Abstract

The goal of the present work is to perform exploratory assessments on the suitability of microcylinder devices to mitigate wind turbine blade vortex induced vibration in small cross flow regimes, specially occurring under adverse weather events. For this purpose, the deep stall behavior of the NACA0021 has been evaluated for the first time on a wide range of high angles of attack (50°–130°) by means of CFD assessments, and the effects of microcylinders have been studied in terms of the reduction of aerodynamic coefficient magnitudes and fluctuations. In a first step, the passive flow control solution at 90° has been analyzed for a total number of investigated cases equal to 15, keeping constant the diameter of the microcylinders and varying the relative positions of these devices with respect to the blade. Mitigations of the load standard deviations between 63% and 97% across all the tested angles of attack have been found. Additionally, genetic programming (GP) was used to obtain a more general viewpoint of the effect of the microcylinders on the aerodynamics forces. Assessments of the flow fields confirm that properly located microcylinders disrupt the coherent vortical structures in the wake, responsible for periodic loading on the blade section.

Introduction

To face climate change, the interest towards sustainable energy sources has risen dramatically. Sources of environmental friendly power generation range from sunlight [1], water [2], wind [3], biogas [4] up to disruptive technologies such as microbial fuel cells [5], [6]. In this sense, the accurate modeling and forecasting of the energy consumption are challenging tasks that may help to overcome the current problematics [7]. Due to the exceptional situation, some more efforts are still required to guarantee the energy availability, considering the world energy consumption [8].

In terms of wind power, to face the rise in demand of wind energy harvesting, the size of wind turbines has undergone a continuous growth during the last few years, the wind-turbine blades being more susceptible to aeroelastic instabilities (especially in parked or non-operational conditions), such as Vortex-induced Vibrations (ViV), which represent a major risk for the structural integrity of wind turbine components [9], [10], [11].

In non-operational conditions (e.g. under extreme weather conditions), the flow features a small crossflow angle (i.e. the flow primarily parallel to the blade sections, at AoA±90°), and it is massively disrupted compared to the attached regime [9], [10], [11], [12], [13], [14], [15], resulting in a mechanism of alternate coherent vortex shedding similar to that occurring in 2D bluff bodies [11], [12], [13], [14], [15]. As a consequence, lock-in phenomena with the shedding frequency coupled to the oscillation frequency [12], [15], can yield large oscillations of the blades [9], [10], [11] and potentially severe structural damage to the rotors.

For the prediction of the lock-in phenomenon, numerical simulations have become an essential tool. For instance, Skrzypiński et al. [11] perform 2D and 3D RANS, and 3D DES unsteady CFD computations with non-moving, prescribed motion and elastically mounted DU96-W-180 airfoil at AoA=90, highlighting the risk of the lock-in phenomenon for the actual wind turbine blade sections. Large-scale, 2D-URANS sensitivity studies on a NREL S809 section at deep stall AoA in [12], [13], [14] provide maps of the lock-in frequencies at different mean AoA for oscillating motions of the blade sections. Hu et al. [15] observed the dependency of the lock-in features on the oscillation frequency and amplitude, and on the Reynolds number, through URANS on a DU96-W-180 airfoil oscillating in edge direction about AoA=90. On the other hand, Garcia-Ribeiro et al. [16] conduct URANS three-dimensional CFD simulations to analyze the effects of taper ratio and root chord ratio of winglets on the performance of wind turbines concluding that winglets can be optimized for enhancing the performance of the horizontal axis wind turbines.

While some work has been done to assess the origin of ViV on wind turbine blades, (passive) control solutions to mitigate these phenomena have been investigated very limitedly. For instance, Zaki et al. [17] numerically investigated the use of a slat near the leading edge of a S809 airfoil to delay or completely cancel separation, leading to 12% saving of material and 6% of the total cost of the turbine. Acarer [18] performed a parametric optimization of the slot geometry on a DU12W262 airfoil, which enabled the location of the slot both on the suction and on the pressure side, reaching the 16% peak CL/CD improvement and an overall αCL/CD rise.

Alternatively to vortex generators (VGs) Mostafa et al. [19] studied the impact of microcylinders on straight bladed NREL Phase II, horizontal axis wind turbine at four different positions around the leading edge. It is found that the microcylinders placed on the pressure side has less effect on the output power compared to when it is positioned in front of the blade leading edge. Similarly, a small rod in front of the leading edge of a symmetrical airfoil was also used in [20] to control the dynamic stall of the Darrieus vertical-axis wind turbine (DVAWT). The formation of dynamic stall vortices was prevented by the counter-rotating vortex shedding from the rod that continuously transmit kinetic energy into the boundary layer of the airfoils. Wang et al. [21] also investigated the impact of microcylinders in front of the blade leading edge as a flow control method to suppress the flow separation. With this configuration, the blade torque can greatly rise up to a 27.3%.

Even if the use of microcylinders seems very promising to improve the aerodynamic performance of wind turbines, there is a lack in the application of this passive control method to control the aeroelastic response.

The goal of the present work is to assess the suitability of microcylinders to prevent and mitigate ViV that might occur under extreme (e.g. critical wind directions and wind speed) weather environments, when wind turbines are in nonoperating conditions. The deep stall behavior of the NACA0021 has been evaluated on a wide range of angles of attack (50°–130°) and the effects of microcylinders have been evaluated in terms of the reduction of the aerodynamic coefficients and their fluctuations.

A parametric study was performed to assess the effects of the microcylinders diameters and locations on the unsteady aerodynamic response of the blade section. Finally, the optimization of both diameter and layout was carried out. The effectiveness of the microcylinders in wind turbines might be of interest for the forthcoming blade designs or as add-on devices temporarily installed during maintenance tasks under nonoperational conditions.

The manuscript is articulated as follows. Section 2 describes the problem and the numerical settings and discretization. Section 3 discusses the effect of the microcylinders on the mean values and on the standard deviations of the lift and drag forces at different angles of attack. The phase portraits of different force components are described in Section 4, followed by a genetic programming fitting based on the main variables of the problem (say, angle of attack, position and the microcylinder diameter). Some flow field maps are also included in this section for the analysis of the loads. Finally, in Section 5 the conclusions of this study and some future works are presented.

Section snippets

Material and methods

A parametric study on the effectiveness of the location and size of the microcylinders around a NACA0021 foil to mitigate the aerodynamic forces has been conducted numerically. The NACA0021 geometry is chosen for the present numerical assessments as experimental data on this geometry [22] are available to validate with. Consistently with other works [12], [13], [14], two dimensional URANS (Unsteady Reynolds Averaged Navier Stokes) simulations are deemed adequate for the present exploratory

Microcylinders for flow at angle of attack equal to 90°

The different microcylinders configurations have been compared in terms of the variation of the aerodynamic coefficients and their standard deviations (i.e., the magnitude of vibrations) with respect to the base case without passive control (clean airfoil). The weighted averages between the standard deviation reductions of CD and CL have been computed to assess the effectiveness of the microcylinders on the mitigation of the aerodynamic coefficients’ fluctuations.

The diameter of the

Discussion

After comparing the performance of the x1y4D10 and x3y5D10 configurations in terms of mitigation of the vortex-induced vibrations, a summary in terms of the time-averaged loads is provided in Fig. 5, where the mean values of the drag and lift coefficients are shown. The drag (see Fig. 5a ) seems to be rather symmetric in a clean airfoil for the range of AoA in study, taking similar values in the range 70°–110°. Out of this range the drag force decreases. However, although it is found that both

Conclusions

In the present work, an analysis of the vortex induced vibrations on a wind turbine airfoil under deep stall conditions has been performed. Then, a methodology to design and assess a new possible mitigation solution consisting in two microcylinders connected to the airfoil has been developed.

The microcylinders disrupt the coherent vortical structures responsible for oscillations in the aerodynamic forces. This can lead to lock-in phenomena and in turn to vortex induced vibrations. An important

CRediT authorship contribution statement

Antonio Martín-Alcántara: Data curation, Software, Investigation, Writing – review & editing. Valentina Motta: Conceptualization, Methodology, Supervision, Writing – review & editing. Andrea Tarantino: Data curation, Software, Validation, Investigation, Writing – original draft. Maria Grazia De Giorgi: Conceptualization, Methodology, Supervision, Writing – review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The authors are grateful to Dr. Marianne Hartung and Marc Canal for their kind advise in the development of the present work.

References (23)

Cited by (2)

View full text