A data and mechanism hybrid driven cutting parameter optimization method considering the machine tool and coolant condition flexibility

Machining configurations, as well as machine tools and coolant conditions, have a significant impact on the selection of cutting parameters. Cutting parameter selection varies with the dynamics of the machining configuration. However, most of the existing approaches carry out cutting parameter optimization under the fixed machining configuration and cannot have the capability to dynamically optimize cutting parameters according to the machining configuration. To this end, a data and mechanism hybrid-driven cutting parameter optimization method is proposed considering machine tool and coolant condition flexibility. Specifically, the effect of various machining configurations and cutting parameters on energy consumption is analyzed, and the cutting parameter optimization model is proposed considering the machine tool and coolant condition flexibility. Secondly, the proximal policy optimization-based cutting parameter energy-efficiency optimization method is developed to select cutting parameters based on the dynamic change of machining configurations. Finally, a case study is conducted to validate the proposed cutting parameter optimization method, and the results show that (1) the proposed data-mechanism hybrid-driven modeling approach can integrate the advantage of data-driven and mechanism-driven models, improving its prediction accuracy compared with other methods; (2) the energy consumption characteristics varies with the dynamic change of machining configurations, and the proposed cutting parameter optimization method can efficiently select cutting parameters adapted to the dynamic change of machining configurations.