Interest in evaluating the performance and safety of offshore oil and gas platforms has been expanding due to the growing world energy supply and recent offshore catastrophes. In order to accurately assess the reliability of an offshore platform, all relevant uncertainties must be properly accounted for. This necessitates the development of a probabilistic demand model that accounts for the relevant uncertainties and model errors.
In this study, a probabilistic demand model is developed to assess the deformation demand on asymmetric offshore jacket platforms subject to wave and current loadings. The probabilistic model is constructed by adding correction terms and a model error to an existing deterministic deformation demand model. The correction terms are developed to capture the bias inherent in the deterministic model. The model error is developed to capture the accuracy of the model. The correction terms and model errors are estimated through a Bayesian approach using simulation data obtained from detailed dynamic analyses of a set of representative asymmetric offshore platform configurations. The proposed demand model provides accurate and unbiased estimates of the deformation demand on offshore jacket platforms.
The developed probabilistic demand model is then used to assess the reliability of a typical offshore platform considering serviceability and ultimate performance levels. In addition, a sensitivity analysis is conducted to assess the effect of key parameters on the results of the analyses. The proposed demand model can be used to assess the reliability of different design options and for the reliability-based optimal design of offshore jacket platforms.
Identifer | oai:union.ndltd.org:tamu.edu/oai:repository.tamu.edu:1969.1/148215 |
Date | 14 March 2013 |
Creators | Fallon, Michael Brooks |
Contributors | Barroso, Luciana, Gardoni, Paolo, Murphy, Colleen |
Source Sets | Texas A and M University |
Detected Language | English |
Type | Thesis, text |
Format | application/pdf |
Page generated in 0.0013 seconds