基于人工神经网络的金属注射成形研究现状与发展  

Research Status and Development of Metal Injection Molding Based on Artificial Neural Network

作　　者：张洪涛 尚峰[2,3,4] 张鹏 陈晓秋 李化强 ZHANG Hongtao;SHANG Fcng;ZHANG Pcng;CHEN Xiaoqiu;LI Huaqiang(School of Mechanical and Electrical Engineering, China University of Mining and Technology, Xuzhou 221116, China;School of Mechanical Engineering, Huaihai Institute of Technology, Lianyungang 222005, China;Jiangsu Marine Resources Development Research Institute （Lianyungang）, Lianyungang 222001, China;Jiangsu Key Laboratory of Large Engineering Equipment Detection and Control, Xuzhou Institute of Technology, Xuzhou 221111, China)

机构地区：[1]中国矿业大学机电工程学院,江苏徐州221116 [2]淮海工学院机械工程学院,江苏连云港222005 [3]江苏省海洋资源开发研究院(连云港),江苏连云港222001 [4]徐州工程学院江苏省大型工程装备检测与控制重点建设实验室,江苏徐州221111

出　　处：《热加工工艺》2018年第12期29-31,36,共4页Hot Working Technology

基　　金：江苏省重点研发计划-产业前瞻与共性关键技术项目(BE-2015100);江苏省大型工程装备检测与控制重点建设实验室开放课题项目(JSKLEDC201506);连云港市科技计划项目(产业前瞻与共性关键技术)(CG1514)

摘　　要：人工神经网络可有效地预测金属注射成形注射阶段最佳工艺参数组合,从而减少后续工艺缺陷。介绍了国内外利用人工神经网络优化金属注射成形工艺的研究现状以及如何建立人工神经网络模型用于金属注射成形工艺参数优化。认为以MATLAB软件优化金属注射成形工艺参数能改善注射坯性能,将为金属注射成形的智能化提出新的思路。Artificial neural network can effectively predict the optimum combination of process parameters in the injection stage of metal injection molding process, thereby, the subsequent processes defects reduce. The research status of metal injection molding technology optimized by artificial neural network in China and abroad, and how to establish the artificial neural network model for the optimization of metal injection molding process parameters were introduced. It is believed that the optimization of metal injection molding process parameters by MATLAB software can improve the performance of injection blank, which can give a new idea for the intelligence of metal injection molding.

关 键 词：金属注射成形 人工神经网络 研究现状 发展前景 

分 类 号：TG14[一般工业技术—材料科学与工程]