Although, transmission pipelines are being hailed as the safest and most cost-efficient way for transporting oil and gas, they are still prone to a variety of metal-loss defects such as aging, corrosion, cracks, dents, etc. These are mainly due to the nature of the environment pipelines operate in (e.g., extreme temperature and pressure inside pipeline, exposure to highly corrosive chemicals, exposure to water and ground which favor corrosion, etc.). The repercussions of not detecting and repairing such defects on time can be very serious: huge financial losses, damage to the environment, health and life hazards, etc., just like what happened in the case of the 2010 methane gas leakage on the Deepwater Horizon oil rig operated by Transocean; a subcontractor of BP Petroleum. This leakage not only killed 11 workers instantly but destroyed and sank the rig, and caused millions of gallons of oil to pour into the Gulf of Mexico, which caused extensive damage to marine and wildlife habitats as well as the Gulf's fishing and tourism industries and its impact still continues. Magnetic Flux Leakage (MFL) scanning is a well established technique for inspecting pipelines made from ferromagnetic material. Experienced pipeline engineers are able to recognize those patterns in MFL scans of pipelines, and use them to characterize defect types (e.g., corrosion, cracks, dents, etc.) and estimate their lengths and depths. This task, however, when done by a human operator, can be a highly cumbersome and error-prone given the amount of data to be analyzed. We propose a solution to automate the pipeline inspection process based on the analysis of MFL scans of oil and gas pipelines. The proposed solution uses a technique based on pattern-adapted wavelets to detect, locate, and estimate the length of metal loss defects along the pipeline. Once a defect is located, we proceed by extracting a number of features from the corresponding MFL signal. Those features are then fed into an artificial neural which returns an estimate of the defect depth. The depth and length are used as the main information needed to assign a severity rating to the detected defect, and decide on the urgency of performing reparations. In practice, pipeline experts use industry standards such as ASME.BG31 to evaluate the severity of a defect given its dimensions and other parameters such as the operating pressure of the pipeline and some other physical properties of the steel from which the pipeline is made. The proposed technique is computationally efficient, achieves high levels of accuracy, and works for a wide range of defect shapes. Besides the ANN-based approach, which works for already laid down pipelines, we are also investigating the usage of higher-order-logic theorem proving to assess the reliability of pipelines structures prior to their installation. The main idea in this complementary approach is to use the reliability block diagram (RBDs) to model the oil and gas pipeline structures in higher-order-logic and reason about their reliability within the a theorem prover.


Article metrics loading...

Loading full text...

Full text loading...

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error