基于Elman神经网络的铸造7050铝合金的均匀化研究  

Study on Homogenization of Casting 7050 Aluminum Alloy Based on Elman Neural Network

作　　者：沈娴[1] 江春然 闫晶[1] 赵艳[2] SHEN Xian;JIANG Chunran;YAN Jing;ZHAO Yan(Hebci Polytechnic Institute,Shijiazhuang 050091,China;College of Information Science and Engineering,Hebei University of Science ＆Technology,Shijiazhuang 050000,China)

机构地区：[1]河北工程技术学院,河北石家庄050091 [2]河北科技大学信息科学与工程学院,河北石家庄050000

出　　处：《热加工工艺》2018年第17期99-101,共3页Hot Working Technology

摘　　要：未热处理的7050铝合金铸锭常伴有严重的成分偏析,需要在进一步加工前进行均匀化处理。而铸锭的均匀化处理工艺直接影响着铝合金的热加工性能。本文引入Elman神经网络对7050铝合金铸锭的均匀化过程中试样导电率和硬度随均匀化温度和时间的变化进行了研究。试验和Matlab仿真的结果表明：本文构建的Elman神经网络对各组检验样本的误差低于5%,具有较高的预测精度。7050 aluminum alloy ingots without heat treatment are often accompanied by severe components segregation,which needs to be homogenized before further processing. The homogenization treatment of the ingot directly affects the hot working properties of the aluminum alloy. The variation of conductivity and hardness with homogenization temperature and time during homogenization of 7050 aluminum alloy ingot was studied by Elman neural network. The experiment and Matlab simulation results show that： the error between the constructed Elman neural network and each group of test samples is less than 5%,which has higher prediction accuracy.

关 键 词：7050铝合金 铸锭 ELMAN神经网络 均匀化 

分 类 号：TG292[金属学及工艺—铸造]