University of Kentucky, Center for Manufacturing-graphic header including drop down menu

Welding Research and Development Lab
Applied Welding Technology

Weld joint penetration control based on geometrical appearance of weld pool

Topic: Precise monitoring and control of weld joint penetration status
Difficulty: Invisibility of weld joint penetration
Approach: Emulation of human operator inference using neurofuzzy model and geometry of weld pool
Tools: Neurofuzzy modeling, high speed image processing, predictive and robust control algorithms

Sponsors: NSF, Navy Joining Center, Army Aviation Systems Command, Allison Engine Company

Control System

Publications:
  • Y. M. Zhang and L. Li, "Interval model based robust control of weld joint penetration," ASME Journal of Manufacturing Science and Engineering, 121(3): 425-433, 1999 .
  • Y. M. Zhang, R. Kovacevic, "Neurofuzzy model based control of weld fusion zone geometry," Accepted for publication, October 28, 1997, IEEE Transactions on Fuzzy Systems, 6(3):389-401, 1998.
  • Y. M. Zhang, L. Li, and R. Kovacevic, "Dynamic estimation of full penetration using geometry of adjacent weld pools," ASME Journal of Manufacturing Science and Engineering, 119(4): pp. 631-643, 1997.
  • R. Kovacevic, Y. M. Zhang, "Neurofuzzy model-based weld fusion state estimation," IEEE Control Systems, 17(2): 30-42, 1997.
  • R. Kovacevic, Y. M. Zhang, and L. Li, "Monitoring of weld penetration based on weld pool geometrical appearance," Welding Journal, 75(10): 317s-329s, 1996.
  • Y. M. Zhang, R. Kovacevic, and L. Li, "Characterization and real-time measurement of geometrical appearance of weld pool," International Journal of Machine Tool and Manufacturing, 36(7): 799-816, 1996.
  • R. Kovacevic, Y. M. Zhang, and S. Ruan, "Sensing and control of weld pool geometry for automated GTA welding," ASME Journal of Engineering for Industry, 117(2): 210-222, 1995.
  • R. Kovacevic and Y. M. Zhang, "Machine vision recognition of weld pool in GTAW," Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, 208(B2): 141-152, 1995.

 

Weld sag geometry monitoring and control

Topic: Robust sensing and control of depression and geometry of weld sag
Difficulty: Strong arc intensity and sensing delay
Approach: Structured-light sensing of weld sag behind the pool
Tools: Robust image processing, adaptive long-range predictive control

Targeting Sponsors: Aerospace and space industries

Control System

Publications:

  • Y. M. Zhang and R. Kovacevic, "Real-time sensing of sag geometry during GTA welding," ASME Journal of Manufacturing Science and Engineering, 119(2): 151-160, 1997.
  • Y. M. Zhang, R. Kovacevic, and L. Li, "Adaptive control of full penetration GTA welding," IEEE Transactions on Control Systems Technology, 4(4): 394-403, 1996.
  • Y. M. Zhang, R. Kovacevic, and L. Wu, "Dynamic analysis and identification of gas tungsten arc welding process for full penetration control," ASME Journal of Engineering for Industry, 118(1): 123-136, 1996.
  • Y. M. Zhang, L. Wu, B. L. Walcott, and D. H. Chen, "Determining joint penetration in GTAW with vision sensing of weld-face geometry," Welding Journal, 72(10): 463s-469s, 1993.

Double-sided arc welding for deep penetration

Topic: Deep penetration for improving productivity, narrowing heat-affected zone, and reducing cost
Difficulty: Limited penetration capability associated with conventional arc welding processes
Approach: Concentrating welding arc by directing the current through the work and the second electrode

Sponsors: NSF, ONR SBIR, National Shipbuilding Research Program, Lincoln Electric, NASA Marshall Flight Center, Thermal Arc.

Publications:

 

Metal transfer control in gas metal arc welding

Topic: Precise control of melted metal drop size and transfer frequency
Difficulty: Dependence of detaching and retaining forces on process parameters
Approach: Lowering peak current to prevent undesired droplet detachment, taking advantage of the downward momentum of oscillating droplet to guarantee the detachment of the droplet under lowered peak current, oscillating the droplet as an active control action
Features: Electrical (voltage) signal sensing, guaranteed precise control of drop size and transfer, lowered peak current for thin-sectioned material welding

Supporting Companies: Thermal Arc
Targeting Sponsors: NSF, automotive industry

System

Publications:

Structured-light based adaptive submerged arc
welding: seam tracking and process control

Process: A cylinder is lapped on a cast head and joined by submerged arc welding
Problems: Varied groove geometry and weld seam
Requirement: Specified root pass penetration and reinforcement, X-ray test (ASME)
Approach: Structured-light based 3D vision for sensing groove geometry and weld profile, first pass for root penetration control, second pass for reinforcement control
Adjusted parameters: wire speed, arc voltage, torch position
Tools: Robust high-speed image processing

Applications: Tank manufacturing and other metal fabricators

SAW Process Control For Propane Tank

Machine vision based high-speed inspection systems

Problems: Human inspectors can not precisely measure the sizes of welds, and often fail to detect undercuts, craters, and other weld defects Requirement: Short inspection cycle, manufacturing environment Approach: High-speed machine vision based automated inspection Tools: High frame rate camera (up to 3,000 frames of images per second), high speed image processing, and pattern recognition

Sponsors: Central Manufacturing Company, Thompson Steel Pipe, Ford Motors, Motor Wheels
Applications: Replacement of human inspectors

Laser machining and welding

Topics: Quality monitoring/control in laser drilling and welding, productivity improvement of drilling and assembly process
Applications: Engine blades, cylindrical structures
Approach: Sensing of reflected plasma arc intensity during drilling, sensing of geometry of drilled holes, sensing of temperature fields
Tools: CCD camera, IR sensors, illumination laser, high speed image processing

Facilities: 2kW CO2 laser, computer-controlled X-Y-Z table, IR sensors, high frame rate camera.

Sponsors: NSF, Pratt&Whitney
Targeting Sponsors: Automotive industries

Monitoring of 3D surface of weld pool

Topic: Monitoring of 3D topography of liquid metal pool deformed by arc pressure and fluid flow
Requirement: Moving torch
Difficulty: Strong arc light, high temperature, mirror-like pool surface
Approach: Optical grid with frosted glass, high-shutter speed camera
Application: Welding processes observation

Sponsor: NSF

Image of Weld Pool Surface Processed Image
Image of Weld Pool Surface Processed Image

 

Publications:

  • R. Kovacevic, Y. M. Zhang, "Real-time image processing for monitoring of free weld pool surface," ASME Journal of Manufacturing Science and Engineering, 119(2): 161-169, 1997.
  • R. Kovacevic, and Y. M. Zhang, "Sensing free surface of arc weld pool using specular reflection: principle and analysis," Proceedings of the Institution of Mechanical Engineers, Part B, Journal of Engineering Manufacturing, 210(6): 553-564, 1996.

Technical Specialties

Welding Processes; Monitoring of Weld Joint Penetration From Front; Real-Time Image Processing; Dynamic Modeling; Neural Networks Modeling and Control; Neurofuzzy Modeling and Control; Self-Tuning Control; Predictive Control; Real-Time Control Software; Systems Integration.

 

Return to Top
©Copyright 1997-2008, UK Center for Manufacturing, College of Engineering, University of Kentucky
text only

An Equal Opportunity University
Last Updated: April 29, 2008