Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Real-time GMAW quality classification using an artificial neural network with airborne acoustic signals as inputs

Conference ·
OSTI ID:33210
; ;  [1]
  1. Naval Surface Warfare Center, Annapolis, MD (United States). Carderock Div.

The acoustic signal produced by the gas metal arc welding (GMAW) arc contains information about the behavior of the arc column, the molten pool and droplet transfer. It is possible to detect some defect producing conditions from the acoustic signal from the GMAW arc. An intelligent sensor, called the Weld Acoustic Monitor (WAM) has been developed to take advantage of this acoustic information in order to provide real-time quality assessment information for process control. The WAM makes use of an Artificial Neural Network (ANN) to classify the characteristic arc acoustic signals of acceptable and unacceptable welds. The ANN used in the Weld Acoustic Monitor developed its own set of rules for this classification problem by learning a data base of known GMAW acoustic signals.

OSTI ID:
33210
Report Number(s):
CONF-930641--; ISBN 0-7918-0785-1
Country of Publication:
United States
Language:
English

Similar Records

Process control of GMAW by detection of discontinuities in the molten weld pool
Conference · Thu Dec 31 23:00:00 EST 1987 · OSTI ID:6772321

Experimental Investigation on Acoustic Control Droplet Transfer in Ultrasonic-Wave-Assisted Gas Metal Arc Welding
Journal Article · Wed Feb 14 23:00:00 EST 2018 · Metallurgical and Materials Transactions B, Process Metallurgy and Materials Processing Science · OSTI ID:22858079

Analysis of acoustic signals on CO{sub 2} arc welding
Conference · Sat Dec 30 23:00:00 EST 1995 · OSTI ID:264135