Datengetriebene diagnose von regelarmaturen zur steigerung der anlagenverfügbarkeit

Unplanned downtimes cause production losses in chemical process plants despite an already high degree of availability. One example is the unplanned replacement of control valves that cannot fulfill their safety and control requirements any more. Due to the large number of signals from process plants, the application of Industrie 4.0 principles and datadriven analysis provide a possible solution. With data analysis, defects can be identified early and preventive measures be taken. For industrial application a number of challenges has to be solved. This contribution presents the requirements for the use-case of diagnosis of control fittings. Obstacles, like the heterogeneity and closeness of the IT infrastructure, the com plexity of physical phenomena and problems during data analysis are discussed and solutions are provided. The application of data analysis techniques in existing production plants is in focus. Furthermore the benefit of shared knowledge pools and results for demonstration of the potential of data-driven analysis are shown.