Automation: plants are becoming increasingly networked

These small measuring devices have been in use for decades in the highly automated plants of the steel industry. They are robust and well adapted to their environment: for example, to enable them to pick up reliable data despite the extreme heat that prevails in the blast furnace or the fine mist in the hot strip cooling section. But are they smart? And fit to play a leading role in Industry 4.0? “Fit, yes, but smart in the true sense of the word, no”, says Volker Lang, Head of Digital Transformation & Innovation at thyssenkrupp Steel. “But that doesn’t mean we have to ‘retool’ everything. We can make our production plant smart by networking the sensor technology already installed.” Digital solutions such as edge computing or IoT gateways can play a key role here.

The principle is similar to that used in the applications that almost 40 percent of the German population know from their own smart home: Users, for example, set a certain temperature and the smart radiators regulate themselves by receiving and evaluating measurement data. For this purpose, the heating system and the sensor must be networked by means of a control unit. The same principle holds true in the hot rolling mill, where constant measurement and control of the temperature is also of key importance because it has a crucial influence on the mechanical properties of the steel sheets produced.

Using sensors to control quality characteristics

Smart networking and the fact that industrial sensors have not only become significantly smaller but also cheaper have fundamentally changed the way data is handled. “As part of our digitalization measures, we have increased the number of sensors installed many times over,” Daniel Overlöper tells us. “At the same time, powerful processors ensure that the huge volumes of data generated are processed in real time.” Take tensile strength as an example: Typically, this parameter, like other mechanical properties, is measured post-production by taking an actual physical sample of the material. In addition, the desired material properties can now also be monitored continuously during production with the aid of additional sensors. The data collected is fed into a mathematical model, together with other process parameters recorded in real time.

This allows results to be simulated in advance. “Although this is not yet a viable alternative to destructive sampling, it already helps the staff in the control room to better understand the quality-relevant parameters and control them accordingly.” Avoiding disruptions

In conjunction with systematic data evaluation, sensors are also able to detect problems at an early stage and increase plant availability. “The technology has improved significantly in the area of vibration measurement, for example,” notes Overlöper. “Today’s sensors are smaller, more accurate, and cheaper. As a result, we have retrofitted them all over the place and are now able to detect very early on if something is not running smoothly.”

Digital networking for efficient steel production

Last but not least, thyssenkrupp Steel uses smart sensor technology to allow us to exploit resources more efficiently and therefore more sustainably. Gas and electricity flows are already being measured today. “It is technically feasible to refine this sensor technology so that we could also use it to record specific energy consumption levels in real time,” explains Volker Lang. Digital networking is an important component of an efficient and sustainable steel production process, which can be traced transparently right back to the smelter.

Algorithms control the cooling process

The water shower cooling has several advantages over the conventional air cooling process. “We were able to reduce the lead times significantly, from 72 on average to only five to six hours“, says Thomas Westermann, Head of Maintenance and Logistics who has played an instrumental role in the implementation of the new process. Quality defects like corrosion pits can be avoided by the controlled cooling process. Another important feature is that the coils are cooled down in the shower cooling area and the storage area preceding the pickling line only to the extent that is optimal for the pickling process.

The basis for this - apart from the consolidation of various process data - is a self-developed mathematical model on the basis of which each individual ring "knows" when it reaches the optimum temperature. The globally unique method thus has two important benefits: it optimizes warehouse logistics and increases productivity in the manufacturing process. "Something like this only works if you have a reliable database," explains Head of IT Ulrich Schneppe. "This is the result of many years of detail work, it's a bit like cleaning out the cellar. But now the data tracking and tracing helps us to link process parameters even more precisely to the material flow, and to continuously get better also in future.”

Smart optimization of occupational safety and health

The digital warehouse management system also meets with a positive response from the internal staff. It enables forklift drivers to spot or deposit coils more quickly and provides information about where the load can be deposited safely – a benefit for occupational safety and health. “The more often you use the digital tools, the more you appreciate their benefits“, says Thomas Westermann. “It starts with the fact that I can optimize my own logistic processes and ends with the controlling process, where I can get information on material stocks, turnover rates and tied up capital at the click of a button.“ “Cleaning out the basement" has paid off.