In-Process Quality Monitoring During Turning Based on High Frequency Machine Data

Abstract

The increasing digitalization of production lines enables an increasing availability of large amounts of data across the whole process chain. Machine tools in particular are increasingly being equipped with edge computing solutions to record internal drive signals with high frequency. This fact enables new approaches to monitor and improve cutting processes in machining production. This paper presents a new approach to detect workpiece defects during turning of a control disc for hydraulic pumps. For this purpose, real-time drive data provided by an edge computing solution was recorded and examined which drive signals are influenced through machining of various workpiece defect categories. This correlation analysis provides the base for the subsequent data processing and model creation using approaches of machine learning. The developed system was evaluated by applying it to the next turning process in the considered process chain to ensure the transferability of the new in-process quality monitoring system.