Quantitive Assessment of Corrosion Probability—A Bayesian Network Approach

Ayello, Francois; Jain, Swati; Sridhar, Narasi; Koch, Gerhardus H.

doi:10.5006/1226

Cited by 45 publications

(20 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The regression techniques for prediction of corrosion damage must be viewed as complementary to other techniques such as Bayesian belief networks . A pipeline integrity management approach may have inputs from elements of an ECDA approach, physics based corrosion models, structural reliability models such as in , and risk based decision support models that include the impact of consequential failure such as shown in .…”

Section: Discussion and The Way Forwardmentioning

confidence: 99%

Challenges in the application of DCVG‐survey to predict coating defect size on pipelines

Anes-Arteche

Bharadwaj

et al. 2016

Materials & Corrosion

View full text Add to dashboard Cite

Corrosion is often the cause of pipeline failure potentially resulting in disasters causing damage and fatalities. To maintain the integrity of nonpiggable lines, NACE's external corrosion direct assessment (ECDA) methodology is commonly applied to assess external corrosion that can occur at coating defects on underground pipelines. Work presented here is from a validation exercise carried out on the results of ECDA assessment using subsequent excavation data. The ECDA was carried out over 300 km of crude oil pipelines with excavation carried out at 200 locations. This paper models the relationships between pipeline coating defect area (area with coating breakdown), corrosion depth, direct current-voltage gradient (DCVG) measurements (in terms of %IR values) and factors capturing diverse environmental conditions through novel application of regression models. This paper sheds light on the challenges in drawing conclusions in the assessment of corrosion from DCVG inspection data and other types of data that form key inputs to ECDA. We expect that the analyses shown here using innovative regression models will support more reliable predictions of external corrosion in pipelines.

show abstract

Section: Discussion and The Way Forwardmentioning

confidence: 99%

Challenges in the application of DCVG‐survey to predict coating defect size on pipelines

Anes-Arteche

Bharadwaj

et al. 2016

Materials & Corrosion

View full text Add to dashboard Cite

show abstract

“…Bayesian baselining of occurrence frequencies provides an essential component of a consolidated risk statement and risk management plan. By implication, frequencies based on the historical occurrence data could be incorporated into risk assessments and more sophisticated approaches such as Bayesian networks, in which probability density functions from the present analysis could become nodes within a network-scale analysis (Ayello, Jain, Sridhar, & Koch, 2014). Because forward analyses of occurrences from UGS facilities, such as mechanistic or decision-tree models for well integrity (e.g., Bois, Garnier, Galdiolo, & Laudet, 2010), may not be carried far enough out to predict failure occurrences, the historical data can provide a complementary view of postfailure occurrence mechanisms and severity magnitudes.…”

Section: Implications For Storage Operators and Regulatorsmentioning

confidence: 99%

Characterization of Historical Methane Occurrence Frequencies from U.S. Underground Natural Gas Storage Facilities with Implications for Risk Management, Operations, and Regulatory Policy

et al. 2019

View full text Add to dashboard Cite

Defining a baseline for the frequency of occurrences at underground natural gas storage facilities is critical to maintaining safe operation and to the development of appropriate risk management plans and regulatory approaches. Currently used frequency‐estimation methods are reviewed and broadened in this article to include critical factors of cause, severity, and uncertainty that contribute to risk. A Bayesian probabilistic analysis characterizes the aleatoric historical occurrence frequencies given imperfect sampling. Frequencies for the three main storage facility types in the United States (depleted oil‐and‐gas field storage, aquifer storage, solution‐mined salt cavern storage) are generally on the order of 3 to 9 × 10–2 occurrences, of all causes (surface, well integrity, subsurface integrity) and severities (nuisance, serious, catastrophic), per facility‐year. Loss of well integrity is associated with many, but not all, occurrences either within the subsurface or from there up to the surface. The probability of one serious or catastrophic leakage occurrence to the ground surface within the next 10 years, assuming constant number of facilities, is approximately 0.1–0.3% for any facility type. Storage operators and industry regulators can use occurrence frequencies, their associated probabilities and uncertainties, and forecasts of severity magnitudes to better prioritize resources, establish a baseline against which progress toward achieving a reduction target could be measured, and develop more effective mitigation/monitoring/reduction programs in a risk management plan.

show abstract

“…A Dynamic Bayesian Network (DBN) is a BN which relates variables to each other over adjacent time steps. BNs and DBNs are especially suitable to model large-scale and complex failure modes of systems and components, due to their ability to incorporate causal inference relationships and their impact on the probabilities of failure modes, based on information from various sources including physics-based models, field data, expert judgment, and updating failure probability over time steps (Ayello et al, 2014;Chen and Pollino, 2012;Palencia et al, 2019).…”

Section: Dynamic Bn-based Corrosion Predictive Modelingmentioning

confidence: 99%

System-level prognosis and health monitoring modeling framework and software implementation for gas pipeline system integrity management

Chalgham

Mosleh

2020

Journal of Natural Gas Science and Engineering

View full text Add to dashboard Cite

Developed the predictive physics of failure models and system-level pipeline health monitoring methodology. The project involves a multi-disciplinary science, engineering, and operational approach to realize a comprehensive and state-of-the-art solution to pipeline integrity. The PSIM platform integrates real-time and historical data, methods, predictive failure models, and systemlevel inference algorithms to form a total system health management support tool to aid in integrity decision making and planning by the pipeline operators. The approach is innovative and unique in its comprehensive integrative perspective and focuses on providing practical solutions while advancing the critical scientific and engineering foundations.

show abstract

Quantitive Assessment of Corrosion Probability—A Bayesian Network Approach

Cited by 45 publications

References 2 publications

Challenges in the application of DCVG‐survey to predict coating defect size on pipelines

Challenges in the application of DCVG‐survey to predict coating defect size on pipelines

Characterization of Historical Methane Occurrence Frequencies from U.S. Underground Natural Gas Storage Facilities with Implications for Risk Management, Operations, and Regulatory Policy

System-level prognosis and health monitoring modeling framework and software implementation for gas pipeline system integrity management

Contact Info

Product

Resources

About