Engineering Commons^™

Real-Time Image Pattern Sensor For Weld Pool Penetration Through Reflection In Gtaw, Yu-Ting Chen

Theses and Dissertations--Electrical and Computer Engineering

In gas tungsten arc welding (GTAW), weld pool surface contains crucial information for welding development. In this research, simulate skilled welders to control the welding process and determine the penetration stages based on the weld pool reaction.

This study focuses on solving the uncertainty of the liquid weld pool in joint bases. The weld pool penetration process is highly depending on how the weld pool surface shape. To observe the weld pool, reflect the weld pool surface by the laser and image on the shield glass. The experiments show that the penetration can’t be determine by the reflecting grayness due …

Real-Time Sensing And Control Of Developing Weld Penetration Through Reflection Vibration In Gtaw, Jinsong Chen

Theses and Dissertations--Electrical and Computer Engineering

GTAW (Gas Tungsten Arc Welding) weld pool surface is believed to contain sufficient information to determine the weld penetration, from which skilled welders are able to control the welding process to desired penetration states. However, it is unclear how human welders extract the weld penetration from the observed weld pool surface. In this research, a novel method is studied to determine the weld penetration based on the dynamic change of the weld pool surface.

This study observes/measures/analyzes the development of a weld pool from partial to full penetration and correlates such change to the weld penetration. Similar trends in the …

Double-Electrode Arc Welding Process: Principle, Variants, Control And Developments, Yi Lu, Shujun Chen, Yu Shi, Xiangrong Li, Jinsong Chen, Lee Kvidahl, Yu Ming Zhang

Electrical and Computer Engineering Faculty Publications

Double-electrode gas metal arc welding (DE-GMAW) is a novel welding process in which a second electrode, non-consumable or consumable, is added to bypass part of the wire current. The bypass current reduces the heat input in non-consumable DE-GMAW or increases the deposition rate in consumable DE-GMAW. The fixed correlation of the heat input with the deposition in conventional GMAW and its variants is thus changed and becomes controllable. At the University of Kentucky, DE-GMAW has been tested/developed by adding a plasma arc welding torch, a GTAW (gas tungsten arc welding) torch, a pair of GTAW torches, and a GMAW torch. …

A Tutorial On Learning Human Welder's Behavior: Sensing, Modeling, And Control, Y. K. Liu, W. J. Zhang, Yu Ming Zhang

Electrical and Computer Engineering Faculty Publications

Human welder's experiences and skills are critical for producing quality welds in manual GTAW process. Learning human welder's behavior can help develop next generation intelligent welding machines and train welders faster. In this tutorial paper, various aspects of mechanizing the welder's intelligence are surveyed, including sensing of the weld pool, modeling of the welder's adjustments and this model-based control approach. Specifically, different sensing methods of the weld pool are reviewed and a novel 3D vision-based sensing system developed at University of Kentucky is introduced. Characterization of the weld pool is performed and human intelligent model is constructed, including an extensive …

Virtualized Welding Based Learning Of Human Welder Behaviors For Intelligent Robotic Welding, Yukang Liu

Theses and Dissertations--Electrical and Computer Engineering

Combining human welder (with intelligence and sensing versatility) and automated welding robots (with precision and consistency) can lead to next generation intelligent welding systems. In this dissertation intelligent welding robots are developed by process modeling / control method and learning the human welder behavior.

Weld penetration and 3D weld pool surface are first accurately controlled for an automated Gas Tungsten Arc Welding (GTAW) machine. Closed-form model predictive control (MPC) algorithm is derived for real-time welding applications. Skilled welder response to 3D weld pool surface by adjusting the welding current is then modeled using Adaptive Neuro-Fuzzy Inference System (ANFIS), and compared …