摘要
建立了神经网络模型,通过训练得到了总体平方差趋于稳定的隐含层神经元数目,对两种镍基高温合金CMSX-4和PWA1484中γ′相和γ相点阵常数进行了预测,并和电子衍射实验结果进行对比。结果表明,预测结果与实验结果具有良好的吻合度。
A neural network model was established to predict the lattice constant of phase and phase in two kinds of nickel-based superalloy CMSX-4 and PWA1484, and the prediction results were compared with the results of electron diffraction experiments. The results show that the values from both methods fit well.
出处
《铸造技术》
CAS
北大核心
2013年第9期1124-1126,共3页
Foundry Technology
关键词
镍基高温合金
神经网路
点阵常数
nickel-base superalloy
artificial neural network
lattice constant