Accurate modeling of mechanical machining is crucial for process monitoring, outcome prediction, and parameter optimization. However, traditional static models fail to account for the dynamic characteristics of machining, leading to significant errors. This paper presents a comprehensive review of dynamic data-model fusion modeling technology to address the issue of neglecting machining dynamics in process modeling. It introduces the fundamentals and methodologies of fusion modeling techniques, categorizing existing fusion strategies into three frameworks: data-mechanism model fusion, data-empirical model fusion, and data-driven model dynamic modeling. These approaches are evaluated across three dimensions: accuracy, interpretability, and real-time performance. Through a comprehensive analysis of fusion modeling applications in tool wear monitoring, surface quality prediction, and process parameter optimization, the study demonstrates that fusion models significantly outperform static models. Finally, the paper explores future development trends in fusion modeling technology.
Xiao et al. (Fri,) studied this question.