Time-dependent structural reliability analysis of wind turbine: A comparative study of MCS, PCE and PDEM

Wind turbines are complex structures which generate clean energy, taking advantage of air currents. However, there are several uncertainties in these structures (dimensions and material) and external factors (environment) that can lead to structural failure, such as tower buckling, base bending and global stability. To study these failures, numerical simulations can be performed. This article aims to study the time-dependent structural reliability of a wind tower including both systems and wind uncertainties. Three methods are used for comparison: Monte Carlo Simulation (MCS), Polynomial Chaos Expansion (PCE), and Probability Density Evolution Method (PDEM). MCS is a robust method that can map uncertainties over time, but it requires many simulations, increasing computational costs. On the other hand, PCE uses only a few samples to encompass behavior over time. PDEM also uses fewer samples but manages to provide better information about statistical moments. All simulations are implemented in the Julia programming language, which offers high-performance numerical computation. This article concludes by comparing the trade-off between accuracy and computational cost for reliability over time.