Application of fuzzy logic to spatial thermal control in fusion welding

Bingul Z., Cook G., Strauss A.

IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, vol.36, pp.1523-1530, 2000 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 36
  • Publication Date: 2000
  • Doi Number: 10.1109/28.887202
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.1523-1530
  • Keywords: fuzzy logic control, gas metal arc welding process, weld seam tracking control, SYSTEMS
  • Kocaeli University Affiliated: Yes


This paper considers the problem of sensing and controlling torch position in the pulsed gas metal are welding (P-GMAW) process. The attitude and positional control described is essential to the production of quality welds,vith a specified geometry. For constant current are welding processes, as normally employed with P-GMAW, the are voltage signal variations that occur as a function of changes in the contact-tube-to-work distance can be used to automatically control the welding system with respect to bead placement and proper sidewall fusion, However, the are voltage signals are uncertain and noisy because of many inherent disturbances associated with the electrode tip, droplet formation, droplet detachment, and droplet transfer through the are. Additionally, the welding process' complex dynamics, nonlinearity, and time variance render traditional control system techniques difficult to apply. To deal with the nonlinear time-varying process with its inherent stochastic disturbances associated with the metal transfer, the theory of fuzzy sets was used as a general framework to interpret the uncertain are signals and provide logic for control, The fuzzy logic controller weld joint tracking system was implemented and tested with pulsed gas metal are welds under a variety of conditions. The goal was to obtain quick and accurate response to tracking errors in the presence of disturbances. A. series of experiments was conducted to evaluate the performance of the fuzzy logic controller. The experimental results show that the fuzzy logic controller was found to be suitable for these purposes and better than methods based on signal averaging and bipolar decision levels under these criteria.