摘要
通过应用人工神经网络技术 ,用获得的时效实验数据进行训练 ,建立Cu 0 30Cr 0 15Zr合金硬度和导电性与时效时间和时效温度的映射模型 ,从而可预测铜合金在一定时效条件下的硬度和导电性。该神经网络算法采用BP算法 ,网络结构采用 2 3 30 2形式。结果表明 ,神经网络用于铜合金的时效性能预测是可行的。
A prediction model for aging properties of copper alloys(Cu-0.15Cr-0.30Zr) under aging environment is developed based on artificial neural net technology, and the non-linear relationship between alloys hardness, conductivity and aging time, aging tempreture is established. The hardness and conductivity of the copper alloys can be predicted by means of the trained neural net from the aging data.The learning algorithm for neural net is BP (back-propagation) algorithm with 2-3-30-2 structure. The results show that the predical model based BP learning algorithm for aging properties performances of copper alloys is feasible and effective.
出处
《材料热处理学报》
EI
CAS
CSCD
北大核心
2003年第4期78-80,共3页
Transactions of Materials and Heat Treatment
基金
国家高技术研究发展计划 (863计划 ) (2 0 0 2AA331 1 1 2 )
河南省重大科技攻关项目 (0 1 0 2 0 2 1 30 0 )
