增材制造工艺过程在线监测与诊断系统开发  

Development of on-Line Monitoring and Diagnosis System for Additive Manufacturing Process

作　　者：都腾飞 邵波 谢雪松 刘新 DU Teng-fei;SHAO Bo;XIE Xue-song;LIU Xin(Changjiang Intelligent Manufacturing Research Institute Co.,Lit.,Jiangsu Changzhou 213000,China)

机构地区：[1]江苏长江智能制造研究院有限责任公司,江苏常州213000

出　　处：《机械设计与制造》2023年第5期292-295,共4页Machinery Design & Manufacture

基　　金：国家重点研发计划资助(2018YFB1308110)。

摘　　要：3D增材制造产品成型过程中,各个关键参数的实时监控具有十分重要的现实意义。这里搭建了3D打印数采分析系统,提出了以ABB机器人、DeviceNet模块等为主体硬件,DeviceNet为实时通讯协议,C#为开发语言及SQL Server为数据库的设计方案,同时通过ARIMA时序机器学习模型,对采集的数据进行自相关函数分析。研究表明,数采分析诊断系统可以对机器人在3D打印过程中的各项数据进行实时监控,通过ARIMA模型的训练可以在焊接过程中对异样参数进行提前预警,从而对工艺参数的优化和产品质量的提升都有实际参考价值。In the process of3D additive manufacturing,the real-time monitoring of key parameters is of great practical significance.A 3D printing data acquisition and analysis system is built,which takes ABB Robot and DeviceNet module as the main hardware,DeviceNet as the real-time communication protocol,C#as the development language and SQL Server as the database design scheme.At the same time,the ARIMA sequential machine learning model is built to analyze the collected data by auto correlation function.The research shows that the data acquisition analysis and diagnosis system can monitor the data of the robot in the 3D printing process in real time.Through the training of ARIMA model,the abnormal parameters can be forewarned in advance in the welding process,which has practical reference value for the optimization of process parameters and the improvement of product quality.

关 键 词：自相关分析 数据采集 3D打印 模式识别 

分 类 号：TH16[机械工程—机械制造及自动化] TP2[自动化与计算机技术—检测技术与自动化装置]