|
Han Guoming
He Yu
Yun Shaohui
Li Junyue
School of Materials Science
and Engineering,
Tianjin University,
Tianjin 300072, China |
|
|
ACQUIRING AND PATTERN RECOGNITION OF SPECTRUM INFORMATION OF
WELDING METAL TRANSFER*
Abstract: In order to recognize various metal transfer modes, by the creation of a pattern recognition system for metal transfer mode, five kinds of spectrum signal in gas metal arc welding (MIG, MAG
and CO2) are collected and taken as training samples. These
samples are pretreated by computer. Several key characteristic
parameters of the spectrum signal are creatively extracted, and a corresponding recognition function and a minimum-distance-classifier are constructed. The results show that using this method, the pattern recognition of several kinds of metal transfer mode for the metal gas arc welding can be done successfully. It has good accuracy and recognition precision, basis for control-ling the metal gas arc welding metal transfer automatically, and relative important parameters in welding process, such as the frequency of droplet transfer and the approximate diameter of each droplet, can also be obtained.
Key words:
Gas metal arc welding Metal transfer Spectrum Acquiring Pattern recognition |