Comprehensive real-time information about the machining process is of undeniable importance both in research and in industrial environment due to its significant potential for increasing process understanding and ensuring machining stability. Suitable data analyses enable early detection of defects which opens possibilities of timely failure prevention. In increasingly complex interlinked production systems, potential errors, malfunctions or defects are becoming more and more diverse, which lead to a rapid increase in the necessary information content.
A complete development of causal or data-driven models in modern production conditions requires skillful recording and synchronization of various data around the process and its environment. Particular challenge here is the real-time acquisition of the diverse data from various data sources: In addition to monitoring of processing progress through direct communication with machine controls, a wide variety of sensors collect process and machine-relevant characteristics. During the process, large amounts of disordered data are generated. Data types, rates and architectures depend on the data origin and often differ greatly from one another. In order to generate intelligent high-quality and useful data from Big Data, the next step is the use of intelligent analysis algorithms.