Tchoupe, ElioElioTchoupeHeidemanns, LukasLukasHeidemannsKüpper, UgurUgurKüpperHerrig, TimTimHerrigKlink, AndreasAndreasKlinkBergs, ThomasThomasBergs2023-07-112023-07-112022https://publica.fraunhofer.de/handle/publica/44540610.1016/j.procir.2022.09.1892-s2.0-85143909217Increasing environmental and economic concerns have increased the use of high-strength materials and superalloys, such as the well-known nickel-based alloys. Due to their very high strength, such materials push established machining methods to their economic limits. As a result, non-traditional machining methods such as precise electrochemical machining (PECM) are gaining importance. PECM has shown significant advantages such as high material removal rates and lower surface roughness. Despite these advantages, some challenges still need to be overcome, such as the implementation of an in-process monitoring system and the determination of optimal process parameters to achieve maximum material removal rates and good surface roughness. While the analytical determination of the optimal parameters is difficult due to the high complexity of the process, the empirical determination shows better results. However, a hitherto very unused approach in this respect is the data-driven model. Based on process data, such models have shown great results in other fields, especially for highly complex systems. The first step in developing such a model is to figure out what process data are needed. In this paper, the potential of a data driven model for PECM is presented. Furthermore, current waveforms for different PECM setups are analyzed and various parameters for the in-process monitoring of the process are formulated. The ability of these parameters to differentiate between different processes was investigated. It was found that the formulated parameters could be used to monitor the efficiency of flushing in the working gap.enECMProcess monitoting, Machinie learningjournal article