POSCO recently partnered with POSCO Technical Research Laboratories, the Department of Systems Management Engineering at Sungkyunkwan University and other steel mills to develop a smart solution to more efficiently manufacture automotive steel. It was tested in January and represents just one example of a successful partnership between POSCO and academic institutions.
Automotive coated steel, one of POSCO’s World Premium Products (WPP), is a high-grade product that requires a high level of sophisticated technology. Only 20 out of 800 steel companies in the world can produce it at this time. Last year, POSCO sold approximately 9 million tons of automotive steel sheets, accounting for 10% of the global automotive steel sheet market.
The ‘Smart Solution for Coating Weight Control Based on AI‘ is a technology that drastically reduces the deviation of coating weight by using artificial intelligence to precisely control the Continuous Galvanizing Line (CGL), the primary technology used in automotive steel sheet production.
In particular, this technology uses an automated control technology that predicts the coating weight in real time and accurately meets the target coating weight by combining the coating weight production model of the artificial intelligence technique with the control model of the optimization technique.
A developer and worker are seen in the operating room monitoring the optimal coating weight predicted through artificial intelligence. POSCO succeeded in developing the ‘Smart Solution for Coating Weight Control Based on AI‘ through a joint industry-university effort, and applied it to POSCO 3CGL last January. With artificial intelligence, it is now possible to collect hundreds of different types of data in real time, as well as accurately predict and control target coating weights.
Coating weight control is a highly-sophisticated technology that keeps the thickness of the coating layer consistent even when operating conditions change at the request of automakers. When coating weight is controlled manually, quality deviates depending on the skill level of the worker, inevitably resulting in significant amounts of wasted zinc. However, the plating process is now automatically controlled by artificial intelligence, increasing the quality of POSCO’s automotive coated steel while decreasing production costs. These new automated processes have also helped to increase work efficiency and productivity with workers.
Since the inauguration of CEO Ohjoon Kwon in 2014, POSCO has been preparing to take the lead in smart solutions by integrating AI into its smart factories. Last June, POSCO Technical Research Laboratories identified the need for an automated coating weight control system and collected the relevant data after discussing with several departments. POSCO worked with Prof. Jon-seok Lee of the Department of Systems Management Engineering at Sungkyunkwan University in Seoul, Korea, to develop the artificial intelligence coating weight prediction model algorithm. Prof. Lee is an expert in statistics, data mining, machine learning and optimization methodologies. He collaborated with POSCO researchers to develop the coating weight prediction program after analysis of the plating process.
Starting in July 2016, a beta program was introduced for two months. After successful completion of the beta test, POSCO Technical Research Laboratories added an additional program to ensure quality control even under changing operating conditions.
The core AI technology that was applied to the coating weight control automation is a self-learning software that utilizes big data deep learning techniques. This method operates in real time allowing the AI program to stay up-to-date by analyzing hundreds of different types of data generated during the plating process. It can accurately predict and control coating weight through real-time, even if equipment has been replaced or operating conditions have changed.
The completed coating weight control automation solution was applied to POSCO’s 3CGL line in Gwangyang on a trial basis for approximately 2 months, enhancing accuracy and stability. Normally, when operating manually, deviations in the coating weight can extend up to 7g per m², but with POSCO’s AI-based technology, this number is reduced to 0.5g per m². After a technical verification process, the technology was incorporated into the 3CGL line on January 5.
To maintain its position as a global leader in automotive coated steel technology, POSCO is planning to apply this coating weight control automation solution to other CGLs at home and abroad, like the recently opened Thailand CGL. POSCO is also taking steps to introduce artificial intelligence technology to the manufacturing processes of other steel products while also continuing to build smart factories that can help POSCO continue to be a leader in the steel industry.
Don’t miss any of the exciting stories from The Steel Wire – subscribe via email today.