MS10.1 - Dynamics of Wind Energy Systems

The current trend in offshore wind energy is to design and install systems with larger swept areas that yield unprecedented efficiency standards. Nevertheless, large blades are needed to achieve these objectives. As a result of aerodynamic and structural tailoring, long and slender blades are produced that are also progressively susceptible to various dynamic instability phenomena during operational conditions. One of these phenomena is the bending-torsion flutter that may lead either to structural failures and system or system breakdowns. Over the last years the Author has been involved in the examination of blade flutter under the influence of stochastic disturbances, e.g., turbulence and aeroelastic load errors. The relevance of this research stems from the potential risk of unstable motion at an angular frequency close to the rated angular speed of operational blade motion. A reduced-order Markov model has been proposed and used to examine the effects of the various perturbations. One of the features that the model accounts for, is the coupling between angular speed and rotationally sampled wind turbulence. Mean-square stability has been predominantly considered; results have shown that various perturbations may negatively impact the instability threshold. In this study the model is employed to investigate moment stability beyond mean squares, observing that the instability may involve nonlinear state propagation, parametric perturbation, and vibration amplitude dependency. Third-order instability will be investigated and compared against previous numerical results. The NREL 5MW reference wind turbine blade is used as a benchmark example.

Prediction of aeroelastic stability for large horizontal axis wind turbines is a key step in the design process to ensure structural safety against catastrophic failures. Historically, flutter has not been a major issue to interrupt the safe operation of wind turbines, but the trend of going bigger with blade sizes might render flutter as a stability concern. Very large HAWTs with blade lengths of 143.4 m or 250 m are being considered for reducing LCOE; however, the slender nature of these very long blades might make them more susceptible to flutter. For horizontal axis wind turbines, classical flutter is an aeroelastic instability which occurs in the event of coupling between a flapwise mode and a torsional mode. At the onset of flutter, the vibration of the blade grows rapidly, and the inherent structural damping becomes insufficient to contain the growth of vibration which may lead to catastrophic failure of the structure. The edgewise instability only involves edgewise mode of vibration and the decrease in damping with increased RPM for this instability is slower compared to classical flutter instability. Classical Theodorsen’s unsteady aerodynamics theory has been widely used in obtaining forcing terms to analyze flutter and edgewise instability of horizontal axis wind turbines (HAWTs). In this study, three classical assumptions: 1) thin airfoil, (2) flat wake and (3) small angle of attack (AoA) assumption of Theodorsen’s theory have been examined, while also adding edgewise aerodynamic terms that are typically ignored. These classical assumptions along with addition of the edgewise aerodynamics have been revisited to examine their impact on prediction of both classical coupled-mode flutter and edgewise instability. The relaxation of the assumptions and addition of edgewise aerodynamic terms led to a new aeroelastic model with modified lift and moment equations. The new aeroelastic model is then applied to blades ranging in lengths from 20m to 246m to evaluate the impact of scale. For 3-bladed HAWTs, relaxation of assumptions had more impact on the classical coupled-mode flutter RPM than on edgewise instability RPM. Addition of edgewise aerodynamic terms affected the edgewise instability RPM more than the classic flutter RPM for all 3-bladed cases. For 2-bladed turbines, relaxing the thin airfoil assumption is the most prominent effect. In general, the assumptions have larger impact for the longer blade lengths. The following are the main contributions and findings of this study for flutter and edgewise instability analysis of HAWT blades: • Reformulation of Theodorsen’s unsteady aerodynamic theory by modifying the small angle of attack assumption, flat wake assumption and thin airfoil assumption. • Addition of edgewise aerodynamic terms into the lift and moment equation for flutter and edgewise instability RPM calculation. • Comparative analysis of 2-bladed and 3-bladed HAWT blades under the influence of the reformulated Theodorsen’s lift and moment equations that account for thick airfoils, large angle of attack, arbitrary wake, and edgewise aerodynamics. • Analysis of the effect of the aeroelastic model assumptions on a wide range of blade lengths from 20m to 246m.

Floating Offshore Wind Turbines (FOWTs) are ground-breaking systems in the renewable sector, capable to exploit wind energy in deep-water areas, where the resource is stronger and abundant with respect to onshore and near-cost sites. Their dynamic behavior results from the complex interaction of flexible and rigid structural elements, such as the rotor blades, the turbine tower, the nacelle, and the platform, with wind and waves. In this regard, the development of efficient and reliable numerical tools is fundamental for the design and the optimization of such structures. In this contribution, a site-specific optimization procedure aimed at finding the optimal substructure configuration for a 10MW FOWT is presented. An in-house developed Frequency Domain (FD) model is adopted for the simulation of the coupled system. The tool is based on first order floating platform hydrodynamics. Viscous drag forces are modelled with Morison’s equation and linearized with Borgman linearization. Mooring lines are modelled with a quasi-static approach. A linearized simulation, performed with the well-known code FAST, allows to estimate the rotor blades and moorings contributions to the equation of motion, leading to a fully coupled FD model. The presented formulation is implemented in an optimization procedure, which is proposed to find platform and mooring system configuration which most effectively reduces the manufacturing costs. An installation site near to the Italia coastline is selected. The joint probability distribution of wind speed, significant wave height and peak spectral period is calculated based on a metocean database of wind and wave records lasting 20 years. The optimization, based on a Genetic Algorithm (GA), is constrained on the turbine and moorings structural responses under a 50-year return period extreme event. Moreover, constraints on the admissible platform displacements, cables geometry, and anchor loads are considered. Results show that the optimized solution significantly reduces the cost of the system with a controlled increase of stresses, opening interesting perspectives for the reduction of the Levelized Cost of Energy (LCOE) in sites characterized by mild sea states and low wind resource.

In the last few years, offshore wind energy has been increasing significantly, being the energetic potential of this technology much higher than that of other renewables energies. Most of Offshore Wind Turbines (OWTs) installed in Europe are founded to the sea floor but, due to the increase in offshore wind turbine installations and the visual impact problem, new locations are being considered with greater depths and increasing seismic risk. In this field, jackets substructures are one of the most attractive options. Despite the low natural frequencies that characterise these systems, the effect of the kinematic interaction can be highly relevant in OWTs on monopiles [1]. For this reason, the need arises to analyse the importance of the influence of kinematic interaction of the seismic responses of multi-supported substructures, such as jackets with deep foundations. This paper presents a comparison of the kinematic interaction effects on the seismic response of two types of substructures for OWTs. OpenFAST [2] is used to analyse the dynamic behaviour of OWTs founded either on monopiles or through jacket substructures on deep foundation, taking into account soil-structure interaction. The comparative shows that the influence of kinematic interaction in both cases has a notable difference. As expected, in the case of the monopile, the rotational motion has a strong effect on the accelerations and the internal forces in the structure, although the impact of the translational filtered signal is small. However, in the jacket, the influence of the rotational motion is less pronounced. [1] Kaynia AM. Effect of kinematic interaction on seismic response of offshore wind turbines on monopiles. Earthquake Engineering & Structural Dynamics, 50(12), 2020. [2] OpenFAST Documentation, National Renewable Energy Laboratory, (2022). https://openfast.readthedocs.io/en/main/. Code published at https://github.com/OpenFAST/openfast.