论文部分内容阅读
基于神经网络的非线性映射和泛化能力,采用人工神经网络方法,建立了置氢TC21合金力学性能预测的BP神经网络模型。模型的输入参数包括高温拉伸试验温度和置氢含量,输出参数为合金的常用力学性能指标,即抗拉强度和屈服强度。通过检验样本验证了ANN模型的准确性。结果表明:该模型具有容错性好、通用性强等优点,可以预测置氢TC21合金在不同拉伸温度和不同置氢含量下的机械性能。同时,将神经网络技术应用于材料制备工艺设计领域,可以明显地提高工艺设计效率,缩短实验周期。
Based on the nonlinear mapping and generalization ability of neural network, BP neural network model for predicting the mechanical properties of hydrogen-occluded TC21 alloy was established by artificial neural network. The model input parameters include high temperature tensile test temperature and hydrogen content, the output parameters of the common mechanical properties of the alloy, that is, the tensile strength and yield strength. The accuracy of the ANN model was verified by testing samples. The results show that the model has the advantages of good fault tolerance and versatility, and can predict the mechanical properties of hydrogen-derived TC21 alloy at different drawing temperatures and different hydrogen contents. At the same time, the application of neural network technology in the field of material preparation process design can obviously improve the process design efficiency and shorten the experiment period.