Digital twin-based tool wear monitoring of micro-milling process

In-situ prediction and intelligent monitoring of micro-milling tool wear remain challenging problems in micro-milling technology. Digital twin technology offers a promising solution to address these issues. This study proposes a digital twin-based monitoring system for micro-milling tool wear. First, micro-milling tool wear experiments were conducted, with the tool diameter reduction rate and flank wear land width used as evaluation indicators. An intelligent prediction model based on a Long Short-Term Memory (LSTM) network was developed to achieve high-precision forecasting of tool wear. Subsequently, the overall architecture design of the digital twin-based monitoring system was completed. A micro-milling motion simulation model was established to enable real-time acquisition, preprocessing, and transmission of micro-milling force data. Leveraging the Unity 3D software development platform and C# programming language, a digital twin-based micro-milling tool wear monitoring system was developed. This system realizes twin reproduction of the micro-milling process and tool wear monitoring, providing critical support for intuitive visualization of the micro-milling process and accurate assessment of tool wear status.