Electroplating faces the challenge of transitioning to Industry 4.0, a particularly complex task due to the combination of mechanical system components and electrochemical process technology. Compared to highly automated mechanical manufacturing, there is a significant need for catching up in production automation and the implementation of Industry 4.0 concepts. This presents great potential for increasing efficiency and reducing costs.
Process monitoring in electrochemical coating is particularly demanding, as the complex electrolyte compositions are often not fully known, and technical equipment is subject not only to mechanical wear but also to severe corrosion. Initial approaches to digitalization using "multi-sensor" systems already exist, but customers—especially in the automotive industry—are increasingly demanding more sustainable, productive, and reliable solutions in electroplating as well.
The aim of the project is to develop an approach for predictive maintenance based on high time resolution performance measurement and the matching of data from the Manufacturing Execution System (MES) for electroplating. Electroplating processes are characterized by a multitude of work steps, processes and equipment as well as materials and energy sources used. Using innovative data mining methods, the effort for data acquisition in Industry 4.0 is to be reduced to a single power measurement in order to make approaches for predictive maintenance available at low cost for both existing and new plants. The inclusion of the control data from the MES in the evaluation enables a clear assignment of effects in the time high-resolution power signal, so that a single measurement can be used for numerous aggregates. For a continuous productive use, the approach shall be converted into a cyber-physical production system. The use of an artificial intelligence makes it possible to develop complex self-learning evaluation algorithms that can be easily applied to further plants.
