Data-driven Discrete Simulation-based Dynamic Modeling and Continuous Optimization for Comprehensive Carbon Efficiency of Batch Hobbing

Abstract

Low-carbon manufacturing is an inevitable requirement for the green transformation of enterprises. For batch hobbing, continuous improvement of process parameters is an important way to achieve low-carbon optimization under the constraints of limited data and time-varying machining configurations. This is the research gap that needs to be filled. Therefore, in this paper, a dynamic modeling and continuous optimization method for comprehensive carbon efficiency (CCE) of hobbing based on data-driven discrete simulation is proposed. Specifically, the study integrates ML (meta-learning) and DEVS (discrete event system specification) in the hobbing process to create a dynamic model of CCE. The dynamic model combines the generalization of the data-driven approach and the capability to abstract events of the discrete simulation approach, which can autonomously adapt to the current machining configuration and output machining results in real time. On this basis, a modified multi-objective seagull optimization algorithm (MOSOA) is used for the continuous optimization of CCE in batch hobbing. Finally, the effectiveness and superiority of the proposed method are verified by a case study and comparative analysis. Moreover, this paper analyzes the effect of process parameters on CCE under different working conditions and provides guidance for gear hobbing.