摘要
在制造业零件加工过程中,精确高效的机械零件加工质量预测模型,可以帮助机械生产工艺过程有更高的生产效率,减少时间和成本。但在目前机械零件加工质量预测模型研究中,预测算法存在着计算复杂、精度低等问题,限制了机械零件加工质量的提升。因此,基于XGBoost机器学习,通过人工蜂群算法与遗传算法结合的方式进行参数寻优,建立机械零件加工质量目标优化函数。实验结果表明,在迭代训练所提方法的预测精确度和收敛速度上都有更好的表现,通过该方法,整体零件加工平均误差为0.0045 mm,相较于优化前,误差有60.87%的降低。
In the process of manufacturing parts processing,an accurate and efficient prediction model for machining quality of mechanical parts can help the mechanical production process to have higher production efficiency and reduce time and cost.However,in the current research on the prediction model of machining quality of mechanical parts,the prediction algorithm has some problems such as complex calculation and low accuracy,which limit the improvement of machining quality of mechanical parts.Therefore,based on XGBoost machine learning,this paper optimizes the parameters by combining artificial bee colony algorithm with genetic algorithm,and establishes the objective optimization function of machining quality of mechanical parts.The experimental results show that the prediction accuracy and convergence speed of the iterative training method are better.Through this method,the average machining error of the whole part is 0.0045mm,which is 60.87%lower than that before optimization.
作者
汪再如
WANG Zai-ru(Anhui Grain Engineering Vocational College,Heifei 230011,Aahui,China)
出处
《贵阳学院学报（自然科学版）》
2023年第1期91-96,共6页
Journal of Guiyang University：Natural Sciences
关键词
零件加工
质量预测
机器学习
人工蜂群算法
遗传算法
Parts processing
Quality prediction
Machine learning
Artificial bee colony algorithm
Genetic algorithm
