论文部分内容阅读
摘要:为了获取SA06铝合金板材不同温度及应变率条件下本构模型及韧性断裂性能，对SA06铝合金板材进行了热态单向拉伸试验，结合热态单向拉伸试验和韧性断裂试验结果，提出了一种修正Misiolek模型;利用修正模型的外插性能预测颈缩后板材流变应力，应用径向基函数神经网络算法建立了Cockroft-Latham韧性断裂阂值预测模型，并对该模型进行了预测精度评估.结果表明，流变应力对温度及应变速率敏感，对比径向基函数网络模型预测误差小于10.6%.
Abstract: In order to obtain the constitutive model and the ductile fracture behavior of SA06 aluminum alloy sheet at different temperatures and strain rates, the hot unidirectional tensile test of SA06 aluminum alloy sheet was carried out. Combined with hot unidirectional tensile test and ductile fracture test As a result, a modified Misiolek model was proposed. The flow stress of the necked plate was predicted by the extrapolation performance of the modified model. The Cockroft-Latham toughness fracture threshold prediction model was established by radial basis function neural network algorithm. The prediction accuracy is evaluated.The results show that the flow stress is sensitive to temperature and strain rate, and the prediction error of the radial basis function network model is less than 10.6%.