Pipeline Risk Assessment Using Dynamic Bayesian Network (DBN) for Internal Corrosion
| dc.contributor.advisor | Mannan, Sam | |
| dc.contributor.committeeMember | Banerjee, Debjyoti | |
| dc.contributor.committeeMember | Schubert, Jerome | |
| dc.creator | Palaniappan, Visalatchi | |
| dc.date.accessioned | 2019-01-18T16:43:38Z | |
| dc.date.available | 2019-01-18T16:43:38Z | |
| dc.date.created | 2018-08 | |
| dc.date.issued | 2018-08-03 | |
| dc.date.submitted | August 2018 | |
| dc.date.updated | 2019-01-18T16:43:38Z | |
| dc.description.abstract | Pipelines are the most efficient mode of transportation for various chemicals and are considered as safe, yet pipeline incidents remain occurring. Corrosion is one of the main reasons for incidents especially in subsea pipelines due to the harsh corrosive environment that prevails. Corrosion can be attributed to 36% amongst all the causes of subsea pipeline failure. Internal corrosion being an incoherent process, one can never forecast exact occurrences inside a pipeline resulting in highly unpredictable risk. Therefore, this paper focuses on risk assessment of internal corrosion in subsea pipelines. Corrosion is time-dependent phenomena, and conventional risk assessment tools have limited capabilities of quantifying risk in terms of time dependency. Hence, this paper presents a Dynamic Bayesian Network (DBN) model to assess and manage the risk of internal corrosion in subsea. DBN possesses certain advantages such as representation of temporal dependence between variable, ability to handle missing data, ability to deal with continuous data, time- based risk update, observation of the change of variables with time and better representation of cause and effect relationship. This model aims to find the cause of internal corrosion and predict the consequence in case of pipeline failure given the reliability of safety barrier in place at each time step. It also demonstrates the variation of corrosion promoting agents, corrosion rate and safety barriers with time. | en |
| dc.format.mimetype | application/pdf | |
| dc.identifier.uri | https://hdl.handle.net/1969.1/174151 | |
| dc.language.iso | en | |
| dc.subject | Dynamic Bayesian Network | en |
| dc.subject | risk assessment | en |
| dc.subject | subsea pipeline | en |
| dc.subject | internal corrosion | en |
| dc.subject | probability distribution | en |
| dc.subject | probability of failure on demand | en |
| dc.title | Pipeline Risk Assessment Using Dynamic Bayesian Network (DBN) for Internal Corrosion | en |
| dc.type | Thesis | en |
| dc.type.material | text | en |
| local.etdauthor.orcid | 0000-0002-8896-196X | |
| thesis.degree.department | Chemical Engineering | en |
| thesis.degree.discipline | Safety Engineering | en |
| thesis.degree.grantor | Texas A & M University | en |
| thesis.degree.level | Masters | en |
| thesis.degree.name | Master of Science | en |