Tool wear monitoring in ultrasonic welding using high-order decomposition

Yaser Zerehsaz, Chenhui Shao, Jionghua Jin

Research output: Contribution to journalArticlepeer-review


Ultrasonic welding has been used for joining lithium-ion battery cells in electric vehicle manufacturing. The geometric profile change of tool shape significantly affects the weld quality and should be monitored during production. In this paper, a high-order decomposition method is suggested for tool wear monitoring. In the proposed monitoring scheme, a low dimensional set of monitoring features is extracted from the high dimensional tool profile measurement data for detecting tool wear at an early stage. Furthermore, the proposed method can be effectively used to analyze the data cross-correlation structure in order to help identify the unusual wear pattern and find the associated assignable cause. The effectiveness of the proposed monitoring method was demonstrated using a simulation and a real-world case study.

Original languageEnglish (US)
Pages (from-to)657-669
Number of pages13
JournalJournal of Intelligent Manufacturing
Issue number2
StatePublished - Feb 15 2019


  • High-order representation
  • High-order singular value decomposition (HOSVD)
  • Principal component analysis (PCA)
  • Tool wear monitoring
  • Ultrasonic metal welding

ASJC Scopus subject areas

  • Software
  • Industrial and Manufacturing Engineering
  • Artificial Intelligence


Dive into the research topics of 'Tool wear monitoring in ultrasonic welding using high-order decomposition'. Together they form a unique fingerprint.

Cite this