Statistical Modeling of Corrosion Failures in Natural Gas Transmission Pipelines
Conference Paper
Overview
Research
Identity
Additional Document Info
Other
View All
Overview
abstract
2016 ASCE. Natural gas pipelines are critical component of U.S. energy infrastructure. The safety of these pipelines plays a key role in the gas industry. Therefore, understanding pipeline failure characteristics and consequences is a very important criterion in decisions making process associated with future design and maintenance of natural gas pipelines. The oil and gas industry spends billions of dollars annually for the corrosion-related issues due to aging and deterioration processes in pipeline networks. Therefore, pipeline operators need to constantly evaluate and update their corrosion prevention strategies. The loss due to corrosion failures is the main motivation for the oil and gas industry to develop accurate maintenance models based on failure frequency. Statistical pipeline failure models and proper maintenance decisions play a significant role in reducing the failure rates of the pipelines and ultimately enhance the cost effectiveness and safety of the pipelines. This paper fulfilled the following two objectives: (1) failure mode characteristics identification, and (2) model development to address the failure rate and reliability. The pipeline failure incident data used for this study was collected from eleven largest natural gas transmission pipeline operators' data from 2001 to 2011. The pipeline networks were divided into two major modes: internal and external corrosion. Internal corrosion refers to the failure inside the pipe due to the liquid characteristics and operating conditions whereas external corrosion refers to the failure due to surrounding environment outside the pipeline. Two major applied stochastic models, homogeneous Poisson process (HPP) and non-homogeneous Poisson process (NHPP), were tested to analyze the reliability trend within categorized installation times. The outcome of the tests on the reliability models shows that the reliability should be modeled with NHPP and the trend is dominantly deteriorating for external and internal corrosion with a few exceptions on stationary and improving. It was also shown that decades of installation and the number of previous failures are the two significant pipeline failure characteristics. The result of this study assists decision-making process in the oil and gas industry to predict the expected number of failures in future transmission pipeline operation more accurately and facilitates the utilization of proper preventive maintenance strategies.
