论文部分内容阅读
利用热模拟机对ZK60镁合金进行等温压缩变形实验,建立人工神经网络模型,并通过相关性系数和相对误差评估和验证模型的预测能力。结果表明,所建立的反向传播神经网络模型能够追踪实验值,可以描述ZK60合金在高温变形时各热力学参数之间高度非线性的复杂关系。
The isothermal compression deformation experiment of ZK60 magnesium alloy was carried out by using the thermal simulator to establish the artificial neural network model, and the predictive ability of the model was evaluated and verified through the correlation coefficient and the relative error. The results show that the established back propagation neural network model can track experimental values and describe the highly nonlinear complex relationship between the thermodynamic parameters of the ZK60 alloy under high temperature deformation.