A carbon emission modeling and low carbon optimization method for machining based on power data state identification

Abstract

With the intensification of global warming, the issue of massive energy consumption in industrial activities and the carbon emissions it triggers is receiving widespread attention. However, existing carbon emission models usually lack comprehensiveness, making it difficult to accurately identify and effectively reduce carbon emissions in machining processes. In this study, a carbon emission model based on state and carbon source is proposed to systematically reveal the effects of different machining parameters on the operating state, energy consumption and carbon emission. First, the semantic segmentation method is innovatively improved and utilized to achieve high-precision recognition of machine tool states, which is significantly higher than the existing methods. Subsequently, by analyzing the coupling relationship between carbon flow and kinetics as well as the state during the machining process, the carbon emission models of energy carbon source, material carbon source and non-desired carbon source of the state are established. Finally, the coupling relationships among processing parameters, operating states, energy consumption and carbon emissions were revealed through response surface methodology (RSM), which visualized the effects of cutting parameters on energy consumption and carbon emissions in different states. In order to quantitatively assess the low carbon level, the carbon efficiency of material removal rate (CEMRR) index is proposed. Optimization of CEMRR significantly reduces the carbon emission while ensuring the machining efficiency. The experimental results show that the machine state recognition accuracy based on semantic segmentation reaches 99.9 %, the optimized CEMRR is reduced by 51.48 %, and the specific cutting energy (SCE) is reduced by 1.97 %; even if the amount of material removed increases by 6.98 %, the carbon emission is still reduced by 40.87 %.