Digital twin-based multi-view energy efficiency prediction for machining systems

Energy efficiency prediction (EEP) of machining systems is essential for energy management and planning. However, the structural complexity and coupled energy consumption across hierarchical levels of machining systems present major challenges for EEP. Current prediction methods, largely reliant on historical data and empirical models, struggle to maintain accuracy when machining condition changes. To address this challenge, this study constructs a multi-view EEP model for machining systems based on the digital twin technology. First, a four views (equipment, workpiece, process, and system) EEP framework is constructed containing a physical machining system, virtual machining system, and twin system. Second, three key technologies for realising multi-view EEP, data acquisition and information interactive perception, energy efficiency state identification and dynamic feature extraction, and low-latency real-time predictive model integration are proposed. Finally, a digital twin platform for multi-view EEP of machining systems was developed and validated. The results demonstrate that the proposed multi-view model and platform enable real-time, accurate EEP across machining systems, offering methodological support for energy savings and emissions reduction with broad application potential.