摘要
采用近等温压缩试验获得AZ31镁合金变形温度在550～750K,应变速率为0.01～10s-1条件下的流动应力。采用BP神经网络原理,建立了ZA31镁合金流动应力与工艺条件的神经网络模型,对在不同变形温度、应变速率和真应力下获得的流动应力实验数据进行训练。结果表明,实验值与预测值的误差很小,误差均在5%以内,为进一步研究AZ31镁合金相关性能与工艺条件的制定提供切实可行方法。
Using compression test, the experimental values of flow stress of AZ31 alloy were obtained at different deformation temperature of 550-750 K and different rates of 0.01-10 s-t. Based on BP neural network theory, an neural network between flow stress of the alloy and technological condition was established and the experimental values of flow stress were trained by neural network. The results show that the predicted values are very precision and the errors is within 5% for the experimental values, and the BP neural network is a good method for further study of AZ31 alloy.
出处
《热加工工艺》
CSCD
北大核心
2013年第16期73-76,79,共5页
Hot Working Technology
