Seismic assessment of wind turbines: How crucial is rotor-nacelle-assembly numerical modeling?. Issue 141 (February 2021)

• Record Type: Journal Article
• Authors: Ali, Ahmer; De Risi, Raffaele; Sextos, Anastasios
• Is Part Of: Soil dynamics and earthquake engineering. Issue 141(2021)
• Publication Date: 2021-02
• Subjects: Renewable energy -- Offshore wind turbines -- Blade deformability -- Tower failure zones -- Seismic vulnerability; Soil dynamics -- Periodicals; Earthquake engineering -- Periodicals; Sols -- Dynamique -- Périodiques; Génie parasismique -- Périodiques; 624.176205
• DOI: 10.1016/j.soildyn.2020.106483
• Languages: English
• ISSNs: 0267-7261

Abstract

Abstract: The cross-section and the structural twist angle of a typical wind turbine blade vary along its span. This complicates its realistic modeling in nonlinear dynamic analysis of wind turbines when seismic performance estimates are sought. As a result, the lumped mass approach is most commonly used to model the rotor-nacelle-assembly (RNA). The RNA is eccentric to the tower top, and the blades tend to induce rotary inertia on the tower. The exclusion of this rotary inertia and the rotor eccentricity can impact the structural response of the wind turbines as the RNA contributes significantly to the total mass of the system. Moreover, the blades are long, slender structural components that can vibrate and deform independently under seismic excitation. The lumped mass approach intrinsically considers the rigid-body inertia for the RNA, which inevitably acts as a part of the tower top. This can affect the seismic vulnerability estimation of the offshore wind turbines (OWT) at a degree that has not yet been properly quantified. To explore this issue, the present study discusses the effects of the three key RNA parameters, i.e., (i) rotary inertia of the blades, (ii) rotor eccentricity, and (iii) blades' flexibility, on the seismic failure and fragility of OWT under shallow crustal earthquakes. Results show that the rotary inertia affects the higher modes, which in turn influence the height of the tower failure zones. It is also shown that different levels of RNA modeling refinement affect the predicted failure probabilities, particularly under pulse-like ground motions, while the same estimates are overestimated if the conventional rigid body lumped mass rotary inertia is used. Even worse, they can be underestimated (thus less safe) when the rotary inertia is completely ignored, compared with the refined modeling of flexible turbine blades. These results are revealing as they highlight that seismic hazard can indeed pose a significant design issue for OWTs in some regions. Highlights: Numerical modeling of different configurations of the rotor-nacelle-assembly (RNA). Effects of blades' rotary inertia and rotor eccentricity on the nonlinear response of offshore wind turbine (OWT) towers. Changes in tower failure locations and seismic fragilities of OWTs due to blades' deformability.