Uncertainty quantification of dynamic response in wind turbines with a geometric nonlinear tuned mass-damper-inerter

This paper investigates the dynamic response of a monopile wind turbine equipped with a geometric nonlinear tuned mass-damper-inerter (NTMDI) under parametric uncertainties. Several probability distributions, including normal, uniform, and Rayleigh, are used to describe uncertainties in structural stiffness, damping, and wind intensity. Polynomial chaos expansion is employed for uncertainty quantification, offering a computationally efficient alternative to Monte Carlo simulations while maintaining comparable accuracy. The results show that damping uncertainties have minimal impact on the dynamic response, whereas stiffness uncertainties primarily affect the performance in the low-frequency range and resonance regions, with high-frequency responses being more predictable and less sensitive to variation. Wind load uncertainty exhibits a considerable influence across the entire frequency spectrum. In non-resonance regions with a single uncertain parameter, the system response mirrors the distribution of the uncertain input, while in resonance regions, amplified polynomial-type nonlinearity causes deviations from this pattern. When multiple uncertainties are considered, the wind load distribution plays a dominant role in non-resonance regions, and response variability is more sensitive at the first natural frequency than the second. The NTMDI-equipped wind turbine demonstrates enhanced dynamic performance compared to conventional TMDs, particularly in reducing resonant peak amplitudes. However, the compounded effects of nonlinearity and uncertainties in resonance regions make the system highly sensitive to variations in structural properties and environmental conditions. This suggests a need for refinement of the approach for large-scale applications with high-dimensional uncertainties. These findings highlight the critical importance of uncertainty quantification in the design and optimization of wind turbines. • Polynomial chaos expansion used for uncertainty quantification in wind turbines. • Nonlinear tuned mass-damper-inerter significantly reduces resonant peaks. • Wind load and structural stiffness greatly impact the dynamic response. • Stochastic response follows distribution of uncertain input in non-resonance region. • Deviation observed at natural frequencies due to compounded uncertainties.