Reliability Analysis of Hydrogen Bubble Defects in Gas Pipeline Based on Big Data

Reliability Analysis of Hydrogen Bubble Defects in Gas Pipeline Based on Big Data

Hydrogen

Pipelines

Decision making

Big data

Defects

Reliability analysis

Data mining

Normal distribution

Distribution functions

Based on the big data theory and data mining of gas pipeline failure history, the key factors determining the safety of pipeline system are found out, such as probability, function, time and prescribed conditions. Based on the reliability principle, the interference state of probability density distribution of stress and strength is analyzed, and the failure area of pipeline is obtained. Deviation coefficient and reliability coefficient are introduced to establish a reliability evaluation model for a hydrogen blistering defect in gas transmission pipelines, which is distinguished by normal distribution function. After variable replacement, the nonstandard normal distribution function can be transformed into the standard normal distribution function. Referring to the national unified standards for reliability design of building structures, and considering the failure characteristics and consequences of ductile failure of pipeline steel, the criteria for judging reliability of pipeline defects are reasonably determined. Taking the hydrogen bubbling defect of ZQ gas pipeline as an example, the reliability of single bubbling defect and adjacent bubbling defect with different sizes under yield strength and blasting pressure are calculated respectively, and the safety grade of pipeline is evaluated, which provides scientific decision-making basis for safe operation of pipeline. 2019 IEEE.

230-234

Zhang, Qimin

Chen, Ning

18th IEEE International Conference on Cognitive Informatics and Cognitive Computing, ICCI*CC 2019, July 23, 2019 - July 25, 2019

2019

conferencePaper

10.1109/ICCICC46617.2019.9146096

http://dx.doi.org/10.1109/ICCICC46617.2019.9146096