This study presents the development and field validation of an on-site applicable process monitoring system for machining-based manufacturing environments. In actual turning and milling operations, the acquisition of stable and reliable data is often ...
This study presents the development and field validation of an on-site applicable process monitoring system for machining-based manufacturing environments. In actual turning and milling operations, the acquisition of stable and reliable data is often hindered by external disturbances such as noise, vibration, cutting fluid, and electrical interference. To overcome these challenges, this study integrates CNC interface data with analog sensor signals (acceleration, sound, and current) to establish a CNC-based automatic triggering data acquisition method, enabling stable data collection without operator intervention. The acquired signals are preprocessed and synchronized to distinguish cutting intervals and eliminate non-cutting signals, thereby ensuring data consistency. The developed monitoring system comprises an embedded PC–based field monitoring unit and a dashboard platform, offering functions such as real-time machining visualization, abnormal machining detection, process–quality correlation analysis, and automatic daily report generation. Field validation results demonstrate that the proposed system effectively detects tool wear, overload, and abnormal machining conditions, while quantitatively analyzing equipment availability, performance, and quality to enhance overall productivity. Future work will focus on developing a sensorless monitoring approach and integrating the system with upper-level platforms such as MES, thereby establishing an intelligent process management framework for machining-based manufacturing sites.