A hybrid approach to integrate machine learning and process mechanics for the prediction of specific cutting energy

Specific cutting energy is an important concept because it affects not only surface integrity but also process sustainability. However, the predictive power of traditional analytical models for specific energy is significantly limited by the complex mechanical-thermal coupling in cutting. This paper has proposed a new hybrid approach to integrate data-driven machine learning and process mechanics for the prediction of specific cutting energy. Compared to traditional analytical models, the accuracy of the hybrid approach has been validated in milling of H13 tool steel and Inconel 718. The predictive model is also transferable to other cutting processes.