| Title |
Evaluation of Failure Probability and Remaining Gas-Pipe Lifetime using Statistics with Hypothetical Pit Data |
| Authors |
최병학(Byung Hak Choe); 김성준(Seong Jun Kim); 심종헌(Jong Heon Shim); 김우식(Woo Sik Kim); 김철만(Chul Man Kim) |
| DOI |
https://doi.org/10.3365/KJMM.2015.53.6.389 |
| ISSN |
1738-8228(ISSN), 2288-8241(eISSN) |
| Keywords |
gas pipeline; corrosion data; probability of failure; statistical method; Monte Carlo; FOSM; FORM |
| Abstract |
Corrosion data obtained from in-line inspection can be applied to assess present integrity as well as to predict future integrity of gas pipelines. This can be accomplished by using a combination of probabilistic analyses and simulation. With a deterministic method, failure models (e.g., NG 18 or ASME B31G)can only provide “Safe or Fail” criteria for corroded pipelines. However, a probabilistic method can offer “Probability of Failure (POF)” using the same models. This paper is focused on determining the POF for hypothetical corrosion-pit data of a gas pipeline. The study presented a ten-step procedure for obtaining the POF of a gas pipeline using three statistical methods (i.e., Monte Carlo, FOSM: First-Order Second-Moment, and FORM: First-Order Reliability Method). Furthermore, the coefficient of variance, regarding the value of POF sensitivity in relation to data variation, was mostly affected by the corrosion-depth parameter of the pit data. |