Scalable uncertainty and reliability analysis by integration of advanced Monte Carlo simulation and generic finite element solvers

This contribution describes how the uncertainty associated with structures can be modeled and analyzed, in context with state-of-the-art FE software and modern computing infrastructure. Uncertainty modeling with high-dimensional random variables and random fields motivates the adoption of advanced Monte Carlo methods for reliability analysis. On the implementation side, object-orientation and parallelization have been embraced to ensure flexibility and performance. A novel, Matlab-based toolkit, COSSAN-X, embodying these characteristics, is presented. The application to a satellite under harmonic excitation and a turbine blade under centrifugal loading indicates the importance of considering spatial fluctuations and the scalability with respect to realistic FE models.