熔滴过渡光谱信息的计算机自动识别  

Automatic recognition of spectrum information of metal transfer with computer

作　　者：韩国明[1] 曹新华[1] 云绍辉[2] 柳刚[1] 李俊岳[1] 

机构地区：[1]天津大学材料科学与工程学院,天津300072 [2]石油大学(华东)焊接研究所,山东东营257061

出　　处：《电焊机》2002年第6期16-19,共4页Electric Welding Machine

基　　金：国家自然科学基金资助项目(59575059)

摘　　要：利用熔滴过渡光谱信息自动识别系统,采集了熔化极气体保护焊(MIG、MAG、CO2)5种熔滴过渡形式的光谱信号作为样本,采用计算机进行预处理,创造性地提取了多个熔滴过渡光谱信号波形关键性的特征,建立了判别法则和最小距离分类器。结果表明,成功地对各种熔滴过渡形式进行模式识别,并计算出过渡频率、熔滴平均直径等与焊接质量密切相关的参数。By the creation of a automatic recognition system for metal transfer mode,this paper collects five kinds of spectrum sig nal in gas metal arc welding(MIG,MAG and CO 2 )as training samples.These samples have been pretreated by computer,and the several key characteristic parameters of the spectrum signal have been creatively extracted,also,the corresponding recognition function and a mini -mum-dis tance-classifier have been constructed.The results show that the pattern recognition of several kinds of metal transfer modes for the metal gas arc welding can be done successfully,and the relative im portant parameters in welding process,such as the frequen cy of droplet transfer and the average diameter of each droplet,can also be obtained.

关 键 词：熔化极气体保护焊 熔滴过渡 光谱信号 计算机 自动识别 

分 类 号：TG403[金属学及工艺—焊接] TP391[自动化与计算机技术—计算机应用技术]