摘要
在实测样本的基础上,用人工神经网络建立了熔敷金属力学性能的预测模型.该模型预测的结果同实验值之间有很好的对应关系.利用该模型研究了杂质元素S,P,O,N和合金元素C,Mn,Ti对熔敷金属低温韧性的影响,并采用正交实验的方法得出了较佳的熔敷金属化学成分.
A mechanical property prediction model for deposited metals was built upon the experimental data with the aid of artificial neural network (ANN). There are good correlations between the predicted results and the experimental data. Using this prediction model, the effects of alloying elements C, Mn, Ti and impurity elements S, P, O, N on the low temperature toughness of deposited metals were studied, and by using orthogonal designed experiment, a good chemical constitution for deposited metal was obtained. The technique proposed can be served as a reliable tool for deposited metals property control and design.
出处
《金属学报》
SCIE
EI
CAS
CSCD
北大核心
2001年第9期947-951,共5页
Acta Metallurgica Sinica
基金
国家973基金资助项目 G1998061511