基于TL-DBN方法的CNC机床节能策略设计  

Energy-Saving Strategy Design of CNC Machine Tools Based on TL-DBN Method

作　　者：钟旭佳 李阳 宋佳佳 Zhong Xujia;Li Yang;Song Jiajia(School of Intelligent Manufacturing,Zhengzhou City Vocational College,Xinmi Henan 452370,China;Jiangsu Shuangliang Cooling System Co.,Ltd.,Wuxi Jiangsu 214444,China)

机构地区：[1]郑州城市职业学院智能制造学院,河南新密452370 [2]江苏双良冷却系统有限公司,江苏无锡214444

出　　处：《现代工业经济和信息化》2024年第9期135-136,140,共3页Modern Industrial Economy and Informationization

摘　　要：为了提高数控机床节能性能,设计一种利用迁移学习方法实现数字控制机床(Computer numerical control,CNC)机床的节能控制过程。建立以随机森林(Random forest,RF)实现驱动功能的离散CNC机床能耗调控模型,减小等待阶段所消耗的能量,促进机床决策效率的快速提升。研究结果表明,时间间隔对机床节能状态影响显著,当间隔增加后,机床由9次切换减少至6次切换,时间间隔设定在25 min最优。各机床都可以在等待阶段通过合适策略达到节能的控制过程,采用该节能控制方案为实现生产过程管控提供了理论依据。In order to improve the energy-saving performance of numerical control machine tools,a transfer learning method is designed to realize the energy-saving control process of Computer numerical control(CNC)machine tools.The energy consumption control model of discrete CNC machine tool based on Random forest(RF)was established to reduce the energy consumed in the waiting stage and promote the rapid improvement of the decision-making efficiency of the machine tool.The research results show that the time interval has a significant effect on the energy saving state of the machine tool.When the time interval is increased,the machine tool is reduced from 9 times to 6 times,and the time interval is set at 25 min.Each machine tool can achieve the energy-saving control process through the appropriate strategy in the waiting stage.The energy-saving control scheme in this paper provides a theoretical basis for the realization of production process control.

关 键 词：迁移学习 节能控制 CNC机床 深度置信网络 

分 类 号：TH166[机械工程—机械制造及自动化]