Data Clustering Utilization Technologies Using Medians of Current Values for Improving Arc Sensing in Unstructured Environments
- 문서 유형
- 연구출판물 · 학술지 논문
- 등재
- SCIEScopus
- 저널 순위
- JCR Q2Scopus Q1
- 저널
- SENSORS24(13) · 2024
- 피인용
- Web of Science 3Scopus 3
- FWCI (Scopus)
- 0.31
- JCR IF
- JCR 2023 3.4
- ISSN
- 1424-3210
저자
- 김희준제1저자Pusan Natl Univ, Dept Comp Engn, Busan 43241, South Korea
- 김정호Pusan Natl Univ, Dept Comp Engn, Busan 43241, South Korea
- Heo, Shin-NyeongSamsung Heavy Ind Co Ltd, Geoje Si 53261, South Korea
- Jeon, Do-HyungSamsung Heavy Ind Co Ltd, Geoje Si 53261, South Korea
- 김원석교신저자Pusan Natl Univ, Dept Comp Engn, Busan 43241, South Korea
초록
In the shipbuilding industry, welding automation using welding robots often relies on arc-sensing techniques due to spatial limitations. However, the reliability of the feedback current value, core sensing data, is reduced when welding target workpieces have significant curvature or gaps between curved workpieces due to the control of short-circuit transition, leading to seam tracking failure and subsequent damage to the workpieces. To address these problems, this study proposes a new algorithm, MBSC (median-based spatial clustering), based on the DBSCAN (density-based spatial clustering of applications with noise) clustering algorithm. By performing clustering based on the median value of data in each weaving area and considering the characteristics of the feedback current data, the proposed technique utilizes detected outliers to enhance seam tracking accuracy and responsiveness in unstructured and challenging welding environments. The effectiveness of the proposed technique was verified through actual welding experiments in a yard environment.
키워드
arc sensingdata miningDBSCANinverter welderseam trackingFILLET WELD TRACKINGSEAM TRACKINGSENSORSSYSTEMGMAW
식별자
- Scopus EID
- 2-s2.0-85198351589
- WoS UID
- 001269599300001



로그인
연구물 목록으로