Corrosion protection and monitoring of off-shore structures

Oil and gas platforms are confronting a problem of ageing as there are many platforms that were constructed over 40 years ago. Every year, the offshore sector incurs considerable losses due to corrosion. On average, oil and gas companies use 6% of their annual income to fight against corrosion. This all in a short-term has a negative impact on the marine environment and in a long-term endeavour toward sustainable energy encountering difficulties. This thesis presents a novel corrosion inspection method by the implementation of the deep neural network and fuzzy logic models. Fuzzy logic is a suitable mathematical tool for the task since it is capable of handling imprecise information from the real world. The benefit of this approach lies in its ability to include personal experiences and acceptable deterministic models in the calculations. This approach can thus help to reduce the dependence upon the precise data, allow modelling even when a phenomenon is incompletely understood, and lessen the difficulties arising due to the complex computation required by more traditional methods. Moreover, image processing based on algorithms can do the automated inspection of external corrosion phenomena. A complete automated system for corrosion detection in pipelines comprises of a drone to flying over these pipelines and capturing photos and/or videos, and an image based on an algorithm to process these visual data and detect corrosion. The proposed deep learning approach effectively wards off the need for manual inspection and other non-vision based non-destructive evaluation techniques for pipeline corrosion which are cost-ineffective and interrupts the functioning of pipelines. Increased production frequently comes with an unknown cost of the increased rate of material degradation and threatening corrosion failures. Therefore, essential topics as corrosion data management and risk assessment are covered.