论文部分内容阅读
研究了稀土元素La对AZ91镁合金力学性能的影响。随La含量增加,抗拉强度和伸长率显著增大,而合金的屈服强度则未发生显著变化。根据BP神经网络的相关原理和方法,建立了关于AZ91镁合金的力学性能预测模型。结果表明,训练输出值与实验值的相关系数达到0.99909,相对误差均在±3%以内,表明采用BP神经网络法来研究相关问题是合理可行的。
The effect of rare earth La on the mechanical properties of AZ91 magnesium alloy was studied. With the increase of La content, the tensile strength and elongation increased significantly, while the yield strength of the alloy did not change significantly. According to the relevant principles and methods of BP neural network, the prediction model of mechanical properties of AZ91 magnesium alloy was established. The results show that the correlation coefficient between the training output and the experimental value reaches 0.99909, and the relative errors are both within ± 3%. It shows that it is reasonable and feasible to study the related problems by BP neural network.