摘要
基于人工神经元网络原理,对影响球铁力学性能的主要参数进行了选取,介绍了在球铁力学性能预测系统中开发应用的人工神经元网络技术,对人工神经元网络的结构选型和训练进行了说明。现场运行结果表明,该系统预测结果准确,可定量得到球墨铸铁的力学性能指标。
Some main factors are selected by researching the relationship between mechanical properties and influence factors. The techniques of neural network applied to the prediction of mechanical properties of ductile iron are introduced based on the method of artificial neural network. The selection of neural net structures and training of neural network are illustrated. The results from plant test show that the system has good performance and mechanical properties of ductile iron are predicted by prediction results of the system.
出处
《机械工程材料》
CAS
CSCD
北大核心
2003年第5期51-53,共3页
Materials For Mechanical Engineering
关键词
球墨铸铁
力学性能
人工神经元网络
ductile iron
mechanical property
artificial neural network
