Model Predictive Control of GTAW Weld Pool Penetration

Jinsong Chen, Jian Chen, Zhili Feng, Yuming Zhang

Research output: Contribution to journalArticlepeer-review

6 Scopus citations


Weld pool penetration is a major indicator of weld integrity but it is not directly measurable. The key to solving this issue is to find or create observable phenomena that can be used to derive the penetration. An innovative method has recently been proposed/developed to derive the penetration from the reflection of projected laser from an oscillating weld pool surface. However, in addition to the penetration, the oscillation, thus the reflection determined by it, also depends on the pulse amplitude of the current, which needs to be adjusted as the control variable to feedback control the penetration as the output. This dependence imposes a constraint on the control algorithm such that the adjustment of the control variable must be relatively slow to minimize additional sources that affect the oscillation/sensor. This restriction on the control variable adjustment also brings challenges to the identification of the dynamic model of the weld process to be controlled, which has to apply a varying control signal to stimulate the dynamics. To solve these issues, this letter proposes using a low order model to catch the dominant dynamics such that a high speed variation in the control signal can be avoided when choosing the model predictive control algorithm. This allows use of relatively poor models to determine the adjustment on the pulsing amplitude, and still achieves relatively accurate control. Experimental results verified the effectiveness of the resultant penetration control system despite the restrictions imposed by the penetration sensor, which affect the identification of high order dynamics and free adjustments of control variables.

Original languageEnglish
Article number8721135
Pages (from-to)2762-2768
Number of pages7
JournalIEEE Robotics and Automation Letters
Issue number3
StatePublished - Jul 2019

Bibliographical note

Publisher Copyright:
© 2016 IEEE.


  • control
  • feedback
  • GTAW
  • image sensor
  • model
  • MPC
  • simulation
  • weld pool

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Biomedical Engineering
  • Human-Computer Interaction
  • Mechanical Engineering
  • Computer Vision and Pattern Recognition
  • Computer Science Applications
  • Control and Optimization
  • Artificial Intelligence


Dive into the research topics of 'Model Predictive Control of GTAW Weld Pool Penetration'. Together they form a unique fingerprint.

Cite this