%0 Journal Article
%T Expert Network for Die Casing Defect Analysis
%A Jiadi WANG
%A Yongfeng JIANG
%A Chen LU
%A Wenjiang DING
%A <br>JiadiWANG
%A YongfengJIANG
%A ChenLU
%A WenjiangDing
%J 材料科学技术学报
%D 2003
%I 
%X Due to the competition and high cost associated with die casting defects, it is urgent to adopt a rapid and effective method for defect analysis. In this research, a novel expert network approach was proposed to avoid some disadvantages of rule-based expert system. The main objective of the system is to assist die casting engineer in identifying defect, determining the probable causes of defect and proposing remedies to eliminate the defect. 14 common die casting defects could be identified quickly by expert system on the basis of their characteristics. BP neural network in combination with expert system was applied to map the complex relationship between causes and defects, and further explained the cause determination process. Cause determination gives due consideration to practical process conditions. Finally, corrective measures were recommended to eliminate the defect and implemented in the sequence of difficulty.
%K Neural network
%K Expert system
%K Die casting
%K Defect analysis
%K Back propagation<br>神经网络
%K 专家系统
%K 拉模铸造
%K 铸件
%K 缺陷分析
%U http://www.alljournals.cn/get_abstract_url.aspx?pcid=5B3AB970F71A803DEACDC0559115BFCF0A068CD97DD29835&cid=84529CA2B2E519AC&jid=324DC3CAB22330DB19158AE0A9B7BFA5&aid=0FEBE406471A32BA&yid=D43C4A19B2EE3C0A&vid=2A8D03AD8076A2E3&iid=E158A972A605785F&sid=CA9ED1AB4D9E3E04&eid=892C6E385D640C1E&journal_id=1005-0302&journal_name=材料科学技术学报&referenced_num=0&reference_num=16