Reliability-based Design of a Monopile Foundation for Offshore Wind Turbines based on CPT Data.

摘要

A reliability-based design optimization (RBDO) of a monopile foundation for offshore wind turbines is conducted to optimize monopile design criteria by explicitly accounting for the effects of uncertainties. The RBDO in this study aims at optimizing cost of construction, installation and failure with respect to the ultimate limit state of a monopile foundation while accounting for the effects of uncertainties in soil parameters and lateral loads. Due to the inherent soil variability and measurement errors, a probabilistic link is constructed to model soil parameter interpretation from CPT data. The probabilistic link is composed of a random field model of CPT data and the interpretation uncertainty associated with existing relations between soil parameters and CPT measurements. Advanced maximum likelihood procedures were employed to account for the effects of spatial correlation in the random field parameters of CPT measurements. Probabilistic models for soil parameters and lateral loads are coupled with the nonlinear p-y finite element model to predict the response of a monopile foundation. The response of a monopile foundation is evaluated with respect to the ultimate limit state, defined by the yield strength of the monopile steel. Based on uncertainties in soil parameters and lateral loads, the probability of exceeding the ultimate limit is evaluated with the Subset Simulation method. The RBDO problem is solved by coupling the Subset Simulation reliability method with the Simulated Annealing stochastic optimization algorithm to minimize the monopile design cost.