当前位置>主页 > 期刊在线 > 智能制造 >

智能制造

放射性可视化算法在车间生产数据分析中的应用
蒋彧琛,方鹏程
(北京航空航天大学,北京 100191)
摘要点击次数:134    

摘  要:车间生产数据,以零件加工为例,几乎所有的零件都具有十几项乃至几十项属性或特征,因此每个零件都可以视为高维空间上的点。由于高维空间通常是违背人类直觉的,而且当数据的维度非常高时,必然引发“维数灾”,导致许多数据挖掘算法不能正常执行,因此从原始数据中很难直观地发现数据中所包含的信息。放射性可视化技术就是将高维数据进行“降维”,运用力学原理和数学方法对高维数据进行转换,映射成二维散点图,从而对零件进行分析和可视化管理。


关键词:放射性可视化;数据挖掘


作者介绍:

蒋彧琛(1990.07-),男,汉族,天津人,硕士。研究方向:企业信息化。


中图分类号:TP391.41     文献标识码:A 文章编号:2096-4706(2018)02-0000-03

The Application of Radioactivity VisualizationAlgorithm on Analysis of Workshop Production Data

JIANG Yuchen,FANG Pengcheng

BeihangUniversity,Beijing  100191,China

AbstractWorkshopproduction data,for example of parts processing,almost allparts have more than ten or even dozens of attributes or features,so each part can be regarded as a point on high dimensionalspace.Because the high dimensional space is usually contrary to human intuition,and when the dimension of the data is very high,it will inevitably lead to "dimension disaster",which causes many data mining algorithms to not be executed normally,so it is difficult to find the information contained in the datafrom the original data.The technology of radioactivity visualization is to"reduce the dimension" of the high dimensional data,and use the mechanics principle and mathematical method to transformthe high dimensional data,map into the two-dimensionalscatter plot,so as to analyze and visualize the parts.

Keywordsradioactivityvisualization algorithm;data mining


参考文献:

[1] 张彦华.工程材料学[M].北京:科学出版社,2010:34-35+119.

[2] SP95ISA-95企业系统与控制系统集成国际标准 [S].美国:仪表、系统和自动化协会,1995.

[3] Cios,K.J.,W.Pedrycz,R.W.Swiniarski,L.A.Kurgan,DataMining:AKnowledge Discovery Approach,Spinger,New York,2007.