Abstract: Intelligent machining, as the core direction of manufacturing industry's transformation to high precision and autonomy, is highly dependent on data enablement technology for in-depth analysis of heterogeneous data from multiple sources and intelligent decision support. In this study, we systematically sort out the research progress of data enablement technology in the field of intelligent achining, put forward a full-process enabling framework covering “On-machine measurement-Signal pre-processingFeature extraction-Multi-source fusion-Data governance”, and analyze three key technologies, namely, multi-source sensing information fusion cloud-fog-edge collaborative computation, and dynamic migration of process knowledge. The study shows that the data enabling technology is effective and efficient. The study shows that data enablement technology effectively enhances the real-time response and robustness of the process, significantly strengthens the guarantee of process control and quality consistency, and promotes the leap of intelligence level in multiple industrial scenarios. However, bottlenecks such as the difficulty of modeling multi-physical field coupling in complex process scenarios, the lack of fidelity of virtual-reality mapping, and the low efficiency of knowledge migration from small samples still restrict the application of the technology. Aiming at the above challenges, this paper proposes a future technology path that integrates dynamic federated learning and causal reasoning to provide a systematic research perspective for the manufacturing industry to evolve to a higher-order stage of cognitization.

Key words: intelligent manufacturing, data enablement, digital twin, intelligent algorithms, machine learning