基于BP神经网络对CuNiSiCr合金时效后性能预测  

Prediction model for aging properties of CuNiSiCr alloy based on BP neural network

作　　者：雷静果[1] 刘平[2] 井晓天[1] 田宝红[2] 

机构地区：[1]西安理工大学材料科学与工程学院,陕西西安710048 [2]河南科技大学材料科学与工程学院,河南洛阳471003

出　　处：《功能材料》2005年第4期522-524,528,共4页Journal of Functional Materials

基　　金：国家高技术研究发展计划(863计划)资助项目(2002AA331112).

摘　　要：采用改进的BP 网络算法———Levenberg Marquardt算法,建立了固溶态Cu Ni Si Cr合金的时效温度和时间参数关于时效后性能的非线性映射模型。结果表明所建立的BP神经网络模型可对Cu Ni Si Cr合金时效后的性能进行有效的预测和分析。A supervised artificial neural network to model the nonlinear relationship between parameters of aging temperature and time with respect to properties was proposed for solution Cu-Ni-Si-Cr alloy. The improved model was developed by the Levenberg-Marquardt training algorithm. The results show that the BP neural network was an effective way and can be successfully used to predict and analyze the properties of Cu-Ni-Si-Cr alloy.

关 键 词：Cu—Ni—Si—Cr合金 时效 BP神经网络 L—M算法 

分 类 号：TG146.1[一般工业技术—材料科学与工程]