CANADA'S LEADING INFORMATION SOURCE FOR THE METALWORKING INDUSTRY

LATEST MAGAZINE

CANADA'S LEADING INFORMATION SOURCE FOR THE METALWORKING INDUSTRY

CANADA'S LEADING INFORMATION SOURCE FOR THE METALWORKING INDUSTRY

Adaptive welding for shipbuilding and offshore structures

Share This Post

Welding in the shipbuilding and offshore platform industries is not for the faint of heart. Welders must contend with harsh environmental conditions that include high humidity, strong winds, extreme temperature variations, saltwater corrosion, and mechanical stresses from waves and currents.

The construction of ships and offshore platforms also often requires welding in tight and confined spaces. Yet somehow these welding operations must meet the stringent quality and safety standards required in these industries and for good reason: One bad weld can cause devastating consequences.

Even if pieces requiring welding can be brought in to be worked on in the relative comfort of the shop, shipbuilding and offshore structures are often made from a variety of materials such as steel, aluminum, and specialized alloys. Welding these materials together requires expertise in selecting the right welding processes, filler materials, and techniques to ensure proper fusion and compatibility and avoid issue such as galvanic corrosion.  Ship hulls and offshore platforms also often have complex geometries with varying thicknesses and joint configurations that require skillful planning and execution and sometimes the use of advanced welding processes such as multi-pass welding.

The obvious challenges presented by shipbuilding and offshore structures is giving rise to adaptive welding, which is the use of advanced welding techniques and technologies capable of adjusting and optimizing welding parameters in real time based on the specific conditions and requirements of the welding process. Adaptive welding in these industries was also part of an informative presentation by Rob Pistor, president of Liburdi Automation, during CWB’s recent Welding Industry Day. The Dundas, Ont.-based company specializes in automation solutions, designing and manufacturing every critical component from software to mechanical and electrical.  

Company president Pistor spoke candidly about the realities of gas metal arc welding (GMAW) and gas tungsten arc welding (GTAW) operations in shipbuilding and offshore structures.

“Anyone who’s had to weld anything knows that there’s what’s on the drawing and then there’s what’s in reality and that could be totally different,” Pistor said. “If you have plates or structures or things that need to be joined, there’s several levels of automation and adaptation that can be used to improve productivity and help you get more out of your labor force.”

Pistor presented a 5-level “hierarchy of automation”, starting with motorized motion and human guided visual sensor technology all the way up to leading-edge technology featuring neural-network based learning and self-driving sensor technology.

Center line movement that would be imperceptible to a human operator is not imperceptible to the machine performing through the arc seem tracking.

LEVEL 1 is motorized motion or “bugs, tracks and brackets” to use Pistor’s more colorful description. It’s very common and it’s decades old. The mechanization is the motor in the track with brackets and levels to adjust the welding torch and a pendant the operator can steer. 

LEVEL 2 is more akin to orbital welding with up to four axes of motion and voltage control as well as height adjustments the welder can control. 

LEVEL 3 is through arc sensing, seam tracking, height control and current control. It’s the first step into automation. Pistor provided an example of welding an elliptical joint on a boat landing for an offshore wind tower. 

“What you need to do this is through the arc seam tracking and tool centerpoint kinematics. It’s a robotic front end on an orbital system. It goes around, adapting to the changing radius. You’ve got a circular on the outside and an ellipse on the inside and the ABC and oscillation axes are adapting to it by sensing through the arc,” he explained. “Very important when you’re GMAW welding is to get the angle right. What this thing does is move in and out at the right angle in compound motion. That’s robotics. It also is very important when you approach the sidewall to get good fusion. You have to be able to hit it at the right angle. Operators want to adjust that but not have to deal with levers and compound motions. This is all automatic, like a robotic arm but it’s on a welding machine on an orbital welder.”

The automation is taking over all the elliptical operations and adjusting the travel speed to be constant so that the bead profile remains consistent. Pistor also showed how center line movement that would be imperceptible to a human operator is not imperceptible to the machine performing through the arc seam tracking.

LEVEL 4 automation applies even more sensors and adaptation. Pistor provided a case study of a spent nuclear waste canister with a lid that couldn’t be properly aligned due to concerns about high radiation. “They just put the lid on top and wherever it falls it falls. We had to adapt to the lid orientation where the gap goes up to 80 thou. In nuclear terms that’s a lot.” Pistor explained, adding that a laser line scanner was used to determine the exact geometry and adapt the fill to provide a constant single pass per layer weld.

LEVEL 5 automation represents the pinnacle of what Pistor’s company is working on: neural network-based machine learning. 

“What we wanted to do is get rid of the laser line scanner. It takes time and adds complexity. We don’t need it because we can get better information from the cameras,” Pistor explains. “We wanted to take that information from the cameras and post and process it in real time and either replace the welder or help the welder better do their job. The neural network is finding where the tungsten is, where the wire is, where the sidewalls are, how thick the weld is and doing all the computations based on what it recognizes in real time. It is doing the exact same weld that was pre-scanned with the laser with now no laser, just video processing using the neural network-based vision.”

Asked if the obvious advantages of adaptive welding technology will lead to the loss of welding jobs, Pistor gave an answer people likely weren’t expecting.

“This will not replace welders. This will make welders do way more than they currently do,” Pistor said. “When you’re ramping up production you have a choice. If you have to do three times the amount of work, you can hire three times the amount of welders or use the same number of welders that you have and help them do three times more with automation. That’s rough numbers but that’s where I see this. I don’t see this as replacing welders. I see it as changing a welder’s job description because these machines don’t install themselves, they don’t crawl out of the box and do it. This is aimed at getting more work out of your labor force than what you’re currently getting because they’re doing it by hand. SMT

Share This Post


Recent Articles



Wordpress Social Share Plugin powered by Ultimatelysocial
error

Enjoy this post? Share with your network