论文部分内容阅读
以800H合金的热压缩实验为基础,分析800H合金在不同温度和应变速率下800H合金的流动应变行为。基于800H合金变形温度、应变率、应变和应力的实验数据,建立关于800H合金热变形的GRNN神经网络预测模型。依据GRNN神经网络训练结果,选择平滑因子为0.2的网络。应力预测值和实验结果的相关性分析表明,建立的800H合金热变形行为GRNN神经网络模型稳定性高、泛化能力很强,可应用于其他合金的热变形行为预测。
Based on the hot compression experiment of 800H alloy, the flow strain behavior of 800H alloy at different temperature and strain rate was analyzed. Based on the experimental data of deformation temperature, strain rate, strain and stress of 800H alloy, a prediction model of GRNN neural network for 800H alloy was established. According to the training results of GRNN neural network, the network with smoothing factor of 0.2 is selected. The correlation analysis between the predicted stress and the experimental results shows that the established GRNN neural network model of the hot deformation behavior of the 800H alloy has high stability and generalization ability and can be applied to predict the thermal deformation behavior of other alloys.