Abstract:  Aiming at the problems of high carbon emission efficiency and poor surface quality caused by using unreasonable processing parameters during the working processes of CNC milling machines, an optimization method of CNC milling machine processing parameters oriented to low carbon and high quality was proposed. Initially, carbon emission factors in the milling processes were analyzed, and target functions of carbon emission efficiency, surface roughness, and processing time were defined. Prediction models for carbon emission efficiency and surface roughness for CNC milling machines were subsequently established, utilizing the support vector regression improved by grey wolf optimizer. Then, with spindle speed, feed rate, and cutting width designated as optimization variables, an improved egret swarm optimization algorithm was applied to optimize the cutting parameters. This resulted in obtaining Pareto front solutions for processing parameters that were low in carbon emissions, high in quality and efficient. Suitable processing parameters were selected using the EW-TOPSIS method. Finally, an experimental platform for monitoring carbon emissions in CNC milling machines was established, and the feasibility and validity of the proposed method were verified by the experimental results.

Key words:  , low-carbon and high-quality, improved support vector regression, improved egret swarm optimization algorithm, entropy weight-technique for order preference by similarity to ideal solution（EW-TOPSIS）

摘要： 针对数控铣床生产过程中工艺参数不合理导致碳排放量高、表面质量差等问题，提出了一种面向低碳高质的数控铣床工艺参数优化方法。分析了铣削过程碳排放因素，给出碳排放效率、表面粗糙度和加工时间的目标函数，构建基于灰狼算法改进支持向量回归的数控铣床碳排放效率和表面粗糙度预测模型。以主轴转速、进给速度和切削宽度为优化变量，采用改进白鹭算法进行切削参数优化，获得了低碳高质高效工艺参数Pareto前沿解，通过熵权逼近理想解排序法选择合适的工艺参数。搭建了数控铣床碳排放监测实验平台，实验结果验证了所提方法的可行性和有效性。

关键词: 低碳高质, 改进支持向量回归, 改进白鹭算法, 熵权逼近理想解排序法