메뉴 건너뛰기
Library Notice
Institutional Access
If you certify, you can access the articles for free.
Check out your institutions.
ex)Hankuk University, Nuri Motors
Log in Register Help KOR
Subject

Remaining Life Prediction Using Gas Pipeline Segmentation Algorithm Based on Decision Tree
Recommendations
Search
Questions

트리 기반 가스배관 분할 알고리즘을 이용한 잔여수명 예측

논문 기본 정보

Type
Academic journal
Author
Young-Geun Ahn (명지대학교) Seong-Jun Kim (강릉원주대학교) Dohyun Kim (명지대학교) Cheolman Kim (한국가스공사) Woosik Kim (한국가스공사)
Journal
The Korean Reliability Society Journal of Applied Reliability Vol.21 No.2 KCI Accredited Journals
Published
2021.6
Pages
173 - 180 (8page)
DOI
10.33162/JAR.2021.6.21.2.173

Usage

cover
📌
Topic
📖
Background
🔬
Method
🏆
Result
Remaining Life Prediction Using Gas Pipeline Segmentation Algorithm Based on Decision Tree
Ask AI
Recommendations
Search
Questions

Abstract· Keywords

Report Errors
Purpose: This study presents a method of segmenting gas pipelines by considering the locations and characteristics of the defects.
Methods: A pipeline segmentation algorithm based on decision trees is used to segment gas pipelines into optimal areas according to characteristics such as distance and corrosion defects. Then, the growth of the corrosion defects and remaining life of each segment are predicted.
Results: Pipeline segmentation and remaining-life prediction were performed on pigging data, and the proposed method showed that adjacent corrosion defects with similar properties could be grouped. From the remaining-life prediction of each segment, the difference in the results between high-risk and low-risk areas was clear; thus, the segment requiring priority management was identified.
Conclusion: We proposed a framework for optimal segmentation of a gas pipeline. The proposed pipeline segmentation algorithm can be used for effective gas-pipeline management based on the remaining life of each segment.

Contents

1. 서론
2. 트리기반 배관 분할 알고리즘
3. 실험
4. 결론 및 향후 연구과제
References

References (9)

Add References

Recommendations

It is an article recommended by DBpia according to the article similarity. Check out the related articles!

Related Authors

Frequently Viewed Together

Recently viewed articles

Comments(0)

0

Write first comments.

UCI(KEPA) : I410-ECN-0101-2021-323-001725142