一种大数据驱动的机床组群互学习精度优化方法及验证  

A Big Data-Driven Mutual Learning Precision Optimization Method for Machine Clusters and Validation

作　　者：王胜[1] 陈聪 陈建新 姜昊 陈翔飞 周明安[1] WANG Sheng;CHEN Cong;CHEN Jianxin;JIANG Hao;CHEN Xiangfei;ZHOU Mingan(Department of Mechanical Engineering,Quzhou College of Technology,Quzhou Zhejiang 324000,China;Zhejiang Herke Intelligent Equipment Co.,Ltd.,Quzhou Zhejiang 324000,China)

机构地区：[1]衢州职业技术学院机电工程学院,浙江衢州324000 [2]浙江赫科智能装备有限公司,浙江衢州324000

出　　处：《机床与液压》2023年第12期63-67,共5页Machine Tool & Hydraulics

基　　金：2021年度衢州市科技攻关项目(第二批)成果(2021F001);浙江省基础公益研究计划资助项目成果(LGC21E050002);2022年衢州市科技计划攻关项目成果(2022NC06)。

摘　　要：数控机床目前急需解决效率提高和精度保证的问题,传统数控机床难以精确、全面地预测和优化各种加工工况下的误差。提出一种通过应用制造大数据来提高数控机床组群整体性能的新思路和实现方法,使机床组群能够以单台机床无法实现的速度进行学习和精度优化,研究制造过程数据的特点和表征方法。实验表明:大数据模型驱动下的机床组群互学习精度优化加工方法,加工精度和表面质量明显优于普通传统数控加工方法,实现了数控铣削的高效高质量加工。CNC machine tools are now urgently needed to solve the problem of efficiency improvement and accuracy assurance,and it is difficult for traditional CNC machines to predict and optimize the error accurately and comprehensively under various machining conditions.A new idea and implementation method was proposed to improve the overall performance of CNC machine clusters by applying manufacturing big data,to enable machine clusters to learn and optimize precision at a rate that could not be achieved by a single machine tool,and to study the characteristics and characterization methods of manufacturing process data.The experiments show that the machine clusters mutual learning precision optimization machining method driven by big data model is significantly better than the common traditional CNC machining method in terms of machining accuracy and surface quality,and high efficiency and high quality machining of CNC milling is realized.

关 键 词：大数据 机床组群 自适应 表面质量 精度优化 

分 类 号：TG543[金属学及工艺—金属切削加工及机床] TG666