With more and more complex artificial intelligence (AI) systems emerging, the steel industry is in urgent need of improving its own production processes while scaling up production. In terms of “smart production”, the steel industry is leading the way and is actively pursuing the opportunities brought by the fourth industrial revolution. From the perspective of the steel industry, Industry 4.0 means almost completely intelligent production of steel plants, operators, products, and production equipment will be fully connected through the Internet of Things (IoT). Sensors capture vast amounts of data, which is then translated by artificial intelligence, thereby optimizing the production line and generating considerable synergies between different devices.
1. Intelligent operation
Today, smart factories have embraced human beings, making the production process full of wisdom. For example, South Korea’s Pohang Iron and Steel Company has two large steel plants, located in Gwangyang and Pohang respectively. The company’s POSFrame software platform collects big data through a huge sensor group, accumulates it, and then sends it to the artificial intelligence process.
There is still a lot of room for improvement in the application of the POSFrame platform to intelligent operations. POSCO has summoned the most experienced employees to teach it artificial intelligence.
The person in charge of Pohang’s smart factory project said that much of the know-how that Posco has is still hidden among operating staff and engineers, and the company is trying to capture knowledge from their brains and program it into the required intelligence.
2. Safe and efficient
The head of the R&D and Innovation Department of the Austrian Voestalpine Group emphasized that as long as all production steps are fully recorded, they will play an extremely important role in defining the concept of Industry 4.0. In terms of controlling equipment instruments and ensuring product quality, fully automatic recording of electronic measurement and monitoring systems, network means, and the use of various device and product parameters are required, which makes it easier to implement shutdown plans, improve production efficiency, and ensure product quality.
The core of an intelligent steel plant is technical know-how, and the core equipment of a steel plant is a blast furnace. The integrated plant at the Voestalpine plant in Linz has already produced exciting changes. The plant’s blast furnaces required continuous charging with alternating packing layers of coke and sinter. These alternating packing layers ensured uniform and efficient gas flow, but morphology and temperature anomalies were difficult to identify. Just recently, with the help of 3D radar technology, the voestalpine has developed a comprehensive model of the charging process, which can measure blast furnace conditions in real-time, thereby increasing production and reducing harmful gas emissions. The fusion of practical experience from the factory floor with new technologies can also make factories safer workplaces.
In order to create zero-accident factories, Japan’s Nippon Steel and Sumitomo Metal Corporation try to predict dangerous situations in time before they occur. Engineers wear smart hard hats equipped with cameras, hazardous gas sensors, and accelerometers. Geo-fencing technology provides an indoor positioning system within the plant. If a worker enters a dangerous area, the hard hat will sound an alarm, and if a worker slips or loses contact in an unusual area, the hard hat will also alert the system.
3. The data revolution
Previously, various data on the steelmaking process could be set, but the fourth industrial revolution has opened up new possibilities, allowing steel producers to collect more data in different ways, namely through different types of smart sensors Communicate with the intelligent system in a local area network. For example, when a slab is headed to a furnace, laser sensors check the intake air, and those results are fed back to artificial intelligence to learn the exact temperature of the slab. Deviations are determined by measuring the heat at the top and bottom of the slab.
Accurate control of the deviation is crucial, otherwise, the deviation of the slab will enter the next rolling process.
These data allow us to accurately trace back to which process and where the defect of the product appeared, under what conditions, and on which day it developed in which direction. If it is not controlled, the slab will also be twisted and accidents will occur, causing irreversible damage to the entire production process.
In order to ensure that the equipment is running well, POSCO has always adopted regular maintenance, but the smart factory provides a new way of thinking. Posco is trying to infuse intelligence into the production process so that more predictive maintenance can be done, which not only means lower maintenance costs but also less unplanned downtime. As we all know, the steel plant is in continuous production 24 hours a day, seven days a week. Just as the engine cannot be replaced while an airplane is flying, it is difficult to replace one unit in the factory at one time.
Relevant persons from VAI said that digitalization has also changed the way steel companies communicate with suppliers and customers. For example, the use of new online platforms, mobile apps, and other systems to provide order tracking and other services has opened up new business models.
The Swedish Steel Company (SSAB) stated that the “smart manufacturing ecosystem” (smart manufacturing ecosystem) built by the data accumulated by the upstream and downstream industries is expected to be integrated into the entire value chain. Participants of the system will receive new services and enter new business fields.
4. Energy saving
In addition to using data to improve production efficiency, steel companies have installed smart meters and other sensors to reduce carbon dioxide emissions and save energy. Although many factories use the exhaust gas produced by industrial furnaces and rolling processes to generate electricity independently, a considerable number of them still use a grid power supply, and there is still room for power saving.
The energy consumption of different steel grades also varies. Sensors can accurately determine which steel grades are the most energy-intensive. Steel companies then analyze energy pricing and produce these steel grades on specific lines to ensure that production occurs when electricity is cheapest.
5. Measurement and improvement
Artificial intelligence can also reduce production costs when a steel strip is subjected to the hot-dip galvanizing process. After the strip is soaked, the installed blades are cut based on real-time measurements, thereby tightly controlling the weight and width of the strip coating. If the amount of coating exceeds the standard in the galvanizing process, it will increase the production cost, so it should be saved as much as possible.
With the help of artificial intelligence, Posco has significantly improved the quality deviation control of galvanized coatings, and the compliance rate has increased from 84% to 99%.
The data collected in each process provides a digital gene map for each steel plate in the factory. The massive amount of data collected by POSCO has increased significantly because it is completely taken from each production step of the product, which plays a non-negligible role in improving product quality and increasing customer satisfaction. According to Pohang, these data allow us to accurately trace which process and part the product defect exists in, under what conditions, at what time, and where it flows.
At the same time, data can enable steelmakers to eliminate defects through continuous closed-loop improvement measures. It can be seen that in order to create a smart factory, the biggest challenge is to collect accurate data. All other processes depend on this vital aspect.
6. Build the best team
It is not easy to gather people to form a team. In order to ensure the successful operation of the team and to change the mindset of the team, POSCO has been running a “smart” project to instill a “smart culture” in engineers. Customers and related suppliers, and even young job seekers can also participate in the project. Automating processes and factories requires workers who understand how robots operate and how to meet their needs. Unemployment in South Korea is currently high, so Posco is also offering courses in big data and artificial intelligence to job seekers.
Despite fears that robots will take jobs from humans, smart factories won’t reduce Pohang’s workforce. Employees often reflect their personal values by analyzing data or developing new products, rather than simple operations on the factory floor.
These changes are not surprising. Fully automated processes and factories also need employees, and technicians who understand how robots work. In fact, this is also a difficult job. In related factories, employees must go through the necessary training.
As global competition intensifies, steel companies have already felt pressure from peers or other material producers, but at the same time, Industry 4.0 has also given new opportunities, which will help improve production efficiency and achieve sustainable development.