Featured Product
This Week in Quality Digest Live
Innovation Features
Anton Ovchinnikov
In competitive environments, operational innovation could well be the answer to inventory risk
Brian A. Weiss
Tech transfer can be more efficient
Zach Winn
Optical milk scanner uses materials-sensing technology to measure the health of cows
Tamara Sheldon
EV subsidies are poorly designed, but simple changes could make them more effective and equitable
Rebecca Beyer
Knowing who creates a robot makes it feel more authentic

More Features

Innovation News
MIT course focuses on the impact of increased longevity on systems and markets
Upgraded with blue laser technology
The Ring Dex 2 filling and capping system is designed to simplify production.
The QM certification is awarded for excellence in curriculum design and quality
Recent research finds organizations unprepared to manage more complex workforce
For scenarios that require a rapid response
With block-based editor, anyone can realize projects using AI-based image processing
Product placement lends depth to sustainability
Industrial Scan&Sand solution wins RBR50 Innovation Award

More News

Emily Newton

Innovation

How AI and Machine Vision Are Changing Welding

Technology equips the welding industry for the future

Published: Tuesday, February 1, 2022 - 13:03

Welding technology has progressed over the years, thanks to innovations that improve accuracy and overall productivity. Some advances have been in welding automation handled by advanced robots. Other breakthroughs rely on artificial intelligence (AI) and machine vision for better defect detection. Here’s a closer look at how those two technologies have helped the industry move forward.

Welding automation reduces human labor needs

One of the reasons for manufacturers’ interest in welding technology is that it could solve or at least ease labor shortages. According to the American Welding Society, more than 50 percent of human-created projects require some type of welding. Additionally, American Welding Society data forecast 400,000 unfilled welding jobs by 2024. Some analysts believe the shortage could surpass that figure.

Training programs make younger generations aware of their opportunities in welding roles. Such programs are good starts, but they won’t bring about an immediate change. AI-powered robots could assist with the deficit in the meantime.

One robot for welding pipes can help a junior employee do as much as three or four highly skilled welders during an eight-hour workday. The product is a collaborative robot with machine-vision capabilities developed by Novarc Technologies. Outfitted with a welding gun, the cobot can move around a fixtured pipe as it welds, while a junior welder oversees the operation and makes adjustments to the cobot as necessary. During the last couple of years, Novarc has worked on a neural network that helps the robot learn and make the appropriate adjustments without human input. The resulting program can examine feathered tack welds and tweak the parameters as needed.

Representatives at Novark say welding this way results in higher consistency. Without the automated system, they noticed variation in the welding inputs depending on who was running the equipment and how much experience they had. Bringing automation into the process reduces that variability.

Defect detection reduces porosity problems

Developers have frequently encountered problems while creating machine-vision applications for welding due to smoke and sparks in the environment. Those issues can complicate decisions about where to place the camera and can limit how well it works.

Tractor manufacturer John Deere recently worked with Intel to combine machine vision with a neural network AI algorithm. For that project, the camera was positioned close to the weld and equipped with technologies that detected what human eyes could not. The solution involves a frame-by-frame examination of a streaming video to look for welding porosity defects, which are cavities in the weld metal that weaken the weld and lead to rework.

The system automatically switches off if it finds something amiss, allowing a technician to take further action. Tests showed this approach achieved up to 97.14-percent accuracy in finding porosity problems. Additionally, because it enabled discovery earlier in the process, there were fewer reworks.

This type of quality control during the welding process could vastly improve how well a finished material performs in post-weld testing, such as tensile tests. When carrying out a tensile test, technicians prepare a sample free from nicks and jagged edges, then place it into grips and pull the specimen until it breaks. The goal is to determine whether the weld is at least as strong as its base material.

The sooner people in the fabrication facility can spot issues by using advanced technologies, the easier it is to remedy problems and lower the defect rate. Improved productivity and fewer quality issues ultimately help the company’s bottom line.

Welding technology helps people expand their skills

Some worry that advancements in welding technology could take jobs away from people. According to the World Economic Forum, by 2035 automation could take on 35 to 50 percent of welding tasks. Industry publication Welding Headquarters notes that automated technology could work for 80 percent of welding jobs, with the remaining 20 percent handled by the most experienced humans. However, the publication also clarifies that such a scenario probably won’t happen anytime soon.

People are still involved in operating and overseeing even the most advanced AI and machine-vision welding solutions. However, a more likely near-term outcome is that welding technology could change how people gain the skills they need to succeed in the profession.

The Augmented Welding project is developing a tool that combines AI, augmented reality, and virtual reality to help people learn to weld. It relies on simulations that give users advice on getting the highest-quality welds while maintaining the most ergonomic positions.

People involved in the initiative believe it will make welding education more efficient and improve the output of those using the tool. They also hope injury rates will decrease.

Welding automation is more accessible

As the robotic welding market continues growing, manufacturers are investigating better ways to set up and train the associated equipment. The authors of a Fortune Business Insights market report anticipate the robotic welding market will reach $9.76 billion by 2028, up from $5.80 billion in 2021. If the makers of welding technology can convince people their products are easy to use, they’ll gain more marketplace traction.

One recently released cobot welding product has an accompanying smartphone app that lets people train the robot without a teaching pendant or programming skills. Mitch Dupont is the director of business development at Hirebotics, which offers the product. “By reducing the time taken to teach new parts by as much [as] 60 percent, Cobot Welder reduces downtime, improves welding quality and productivity, and ensures painless, automated welding deployments,” he says.

Volkswagen is one example of a major brand currently using cobots to help with welding automation tasks. The company reportedly uses the machines for defect detection. The robots perform spot-checks on welding patterns associated with an electric SUV.

Once decision-makers realize it’s getting increasingly easy to deploy AI, machine vision, and associated technologies, they’ll be more open to considering them.

Welding technology is the future

These examples show how AI and machine vision, as well as the robots that often rely on those technologies to function, will help equip the welding industry for the future. The payoffs for implementing these options aren’t always immediately evident, but they usually become apparent after a company customizes the processes and technology to meet its needs.

Discuss

About The Author

Emily Newton’s picture

Emily Newton

Emily Newton is the editor-in-chief of Revolutionized, an online magazine exploring the innovations disrupting the scientific and industrial sectors.