基于神经元网络的Al-Mg-Si合金动态再结晶预测  

Prediction on dynamic recrystallization behavior of Al-Mg-Si alloy based on artificial neural network

作　　者：刘建勃[1] 王智毅[1] 马雄[2] 

机构地区：[1]榆林学院能源工程学院,陕西榆林719000 [2]钢铁研究总院高温材料研究所,北京100081

出　　处：《塑性工程学报》2017年第4期167-172,共6页Journal of Plasticity Engineering

基　　金：榆林学院高层次人才科研启动基金项目(11GK57)

摘　　要：利用Gleeble1500D热模拟实验机对Al-0.62Mg-0.73Si合金进行等温压缩实验,分析热变形参数(温度、应变速率、应变)对流变曲线特性的影响规律,并利用BP人工神经网络模型构建了合金热变形过程中动态再结晶行为动力学模型。实验结果表明:BP人工神经网络模型能够较好的描述Al-0.62Mg-0.73Si合金热变形时的动态再结晶行为,此外,通过对BP人工神经网络模型中不同隐含层节点数条件下的预测精度分析得到,当隐含层节点数大于等于9时,BP人工神经网络模型的预测效果最佳。本研究结果可用于优化Al-0.62Mg-0.73Si合金热变形工艺参数,并为全面地研究铝合金热变形行为提供理论依据。The isothermal compression test of Al-0. 62Mg-0. 73 Si alloy was performed on Gleeble 1500 D thermal simulation testing machine. The influence of temperature,strain rate and strain on the rheological curve features was analyzed. The dynamic recrystallization kinetics model was established by BP artificial neural network model. Results show that the established BP artificial neural network model can well exhibit the dynamic recrystallization behavior of Al-0. 62Mg-0. 73 Si alloy during hot deformation. In addition,influence of the number of neuron on the prediction accuracy of BP neural network model was analyzed. The ideal BP artificial neural network model is considered to be determined when the number of hidden layer nodes is greater than or equal to 9. The results can be used to optimize the hot deformation parameters and provide a theoretical basis for the study of hot deformation process of Al-0. 62Mg-0. 73 Si alloy.

关 键 词：AL-MG-SI合金 等温压缩 动态再结晶 BP神经元网络 模型评价 

分 类 号：TG319[金属学及工艺—金属压力加工]