Monitoring compounding processes in the context of Industry 4.0.

Digital transformation, Automated process monitoring and material documentation

The plastic processing industry is undergoing a major upheaval. Although the topic of Industry 4.0 is still very abstract to most of the plastic processing industry and systems manufacturers, there is a great deal happening ”behind the scenes”. A guideline has been established called VDIA/DE 4000 that gives companies the opportunity to determine their own individual level of maturity in the context of Industry 4.0. This means companies are able to initiate some important steps in the context of digital transformation and to do this in a targeted way. This is essential for enabling companies to integrate themselves into a digital value-added chain in the future. To do this, Fraunhofer LBF develops methods that detect material and process characteristics, perform a preliminary evaluation of them and make them available for specific purposes.

When discussing Industry 4.0, you often hear things like: ”We’ve been doing that for years!”, ”What’s supposed to be new about it?”, and ”How will that help us?”. Although it is often difficult for many companies to understand the detail of what is currently happening in relation the digital transformation and where this is leading, it is clear that digitalization will continue apace and open up a host of new opportunities and business models. Any company that fails to pay attention at this stage of the upheaval and does not position itself correctly runs the risk of preparing for the future too late. Local networking of machinery and equipment, the integration of systems into an IT infrastructure and the recording and storage of data are important steps in the context of digitalization, but they are not fully representative of ”plastic processing 4.0”. A fundamental requirement of plastic processing 4.0 is for all machinery, equipment and testing data plus all data from business processes to be stored in a central database. A central data storage solution forms the basis for all subsequent steps in plastic processing 4.0. At the same time, this means that communication must be via open, neutral and standardized interfaces. The architecture that has emerged against this background is OPC UA. This means that all access to machinery and system data will take place via an interface of this type in future.

High-resolution torque signal of the left shaft of the twin screw extruder at a speed of 600 rpm. Polycarbonate (yellow); polyethylene (blue).

Digital fingerprint for materials

Once this step is complete, any sensor data can be accessed, stored centrally, and evaluated and correlated. This can even be done on a cross-company basis. In addition to all of the data currently available, the processes of evaluation, correlation and interpretation will play a major role in future. For example, at Fraunhofer LBF a high-resolution torque measurement was installed on a co-rotating twin screw extruder. The original goal of these experiments was to gain a better understanding of the energy input into the material during the fusion stage. However, careful evaluation of this data enables even more information to be obtained. In the simplest case, an average torque can be calculated to evaluate the utilization of the machine – this is the state of the art. But if you consider the complete spectrum of torque over time, you can evaluate torque peaks and changes to these with great accuracy. For example, it would then be possible to record and evaluate batch fluctuations in raw materials (e.g. with increasing use of recycled materials), wear status in the system and/or local individual events. If the signals recorded in the frequency range are considered, this provides even more new information about the current operating status of the system. Operating statuses in the range of harmful resonance frequencies can be avoided, for example. The information obtained is not used simply for monitoring, regulating and optimizing the system; it also acts as part of the material’s digital fingerprint.
In this way, data can be correlated along the entire value-added chain in future and interrelations along this chain can be incorporated directly into the subsequent process.