Thesis
Improving reliability assessment of offshore structures using Bayesian methods
- Creator
- Rights statement
- Awarding institution
- University of Strathclyde
- Date of award
- 2021
- Thesis identifier
- T16244
- Person Identifier (Local)
- 201787997
- Qualification Level
- Qualification Name
- Department, School or Faculty
- Abstract
- Offshore jacket platforms are commonly adopted structures for oil and gas production in shallow water depths. A large number of existing jacket platforms are operating beyond their design life due to the high cost of replacement. The safety of these structures is a major concern of the operators. Jacket structures are constructed as truss frameworks in which tubular members are welded together to create a steel frame. Fatigue damage in jacket structures is most probable to occur at the welded tubular joints due to the geometric discontinuity of the connections which produces high-stress concentrations in these intersections. Fatigue is a complicated phenomenon. As a result of the idealisations and approximations employed in the analysis process, fatigue analysis will be associated with some degree of uncertainty. The overall aim of this research work is to develop an innovative approach to improve the fatigue reliability assessment of jacket structures using Bayesian methods to incorporate new information obtained from the inspection results. Due to the existence of many uncertainties in the fatigue process and other uncertainties in loads and the resistance of the structure, a probabilistic approach for fatigue analysis of jacket structures is a rational and consistent basis for the inclusion of uncertainties. To develop a probabilistic approach for the reliability-based assessment, it is necessary to determine the probability of failure of each joint during the operational life of the structure. However, jacket platforms are redundant structures. Therefore, reliability analysis at a system level is more applicable than at the component level. In this research, the structural reliability analysis at the system level for a jacket platform is performed under both fatigue and extreme loading. At first, the probability of fatigue failure for each component is calculated by using the Monte-Carlo simulation. Then, important failure paths are identified by using a searching process. The system failure criterion is evaluated by comparing the platform strength and loading distributions in terms of base shear. Having calculated the structure strength and loading distributions, the annual probability of failure under an extreme wave is calculated and compared to the tolerable probability of failure. To maintain the safety of jacket platforms in service life concerning fatigue failure, inspection is an important measure. However, the significant costs of inspections, particularly underwater inspections, make it important to properly prioritise inspection locations and inspection frequency. The cost of an inspection is directly proportional to the number of inspections carried out. Therefore, it is required to concentrate only on fatigue sensitive locations in the structures. At the component level, fatigue-sensitive locations are the locations that have low estimated fatigue lives. However, at the system level, critical components are those joints that have a big effect on system reliability. Due to the significant costs of inspections, the identified failure paths can be used as a database for the inspection plan. Inspection activities provide additional information, which includes detection and measurement of crack size. After an inspection of a structure, the perception of structure condition is improved. In general, a Bayesian framework is used to update the probability distributions of the uncertainties such as crack size in a joint. The updated crack size distribution can be used to update the estimation of the probability of failure. Different methods of Bayesian inference to update the probability distribution of the crack size are presented in this research. The credibility of the Bayesian updating process is one of the main concerns for the platforms’ owners. A Bayesian process is a mathematical tool that processes the inputs and generates the outputs based on the provided inputs. Hence, if the inputs are inaccurate, the updating results are worthless and can even lead to wrong decisions for the next inspection activity. Therefore, a novel approach is developed to assess the reliability of the Bayesian methods and assure the platforms’ owners regarding the updating results. This approach is capable to update the probability distributions of all uncertain parameters involved in the fatigue analysis besides the crack size. Three different categories of uncertainties are updated including, Fatigue crack size; POD curve; and uncertainties involved in the predicted model of the fatigue crack size (e.g. initial crack size, crack growth parameter, stress range, etc.). The presented methodology maximises the benefit of the inspection results by updating several uncertain parameters involved in the fracture mechanics approach. Moreover, guidance is provided to help the user to apply the proposed methodology in practice.
- Advisor / supervisor
- Oterkus, Selda
- Barltrop, N. D. P. (Nigel D. P.)
- Resource Type
- DOI
- Funder
Relations
Items
| Thumbnail | Title | Date Uploaded | Visibility | Actions |
|---|---|---|---|---|
|
|
PDF T16244 | 2022-05-30 | Public | Download |