2A70铝合金本构关系的新模型  

A new model of the constitutive relationship of 2A70 aluminum alloy

作　　者：刘芳[1] 单德彬[1] 吕炎[1] 杨玉英[1] 

机构地区：[1]哈尔滨工业大学材料科学与工程学院,黑龙江哈尔滨150001

出　　处：《哈尔滨工业大学学报》2005年第4期449-450,517,共3页Journal of Harbin Institute of Technology

基　　金：中国博士后基金资助项目

摘　　要：为了研究2A70铝合金高温变形时的流动规律,在Gleeble-1500热模拟试验机上进行了一系列等温压缩试验.在此基础上,对传统的BP算法进行了改进,建立了2A70铝合金本构关系的人工神经网络模型.在该模型中,等效应变、等效应变速率和变形温度是模型的输入向量,输出向量是材料的流动应力.模型的计算结果表明:对于样本数据,该模型的相对误差不超过±3%;对于非样本数据,模型的相对误差不超过±6%,说明此模型能准确的反映2A70铝合金热变形时的流动规律.To study the flow rules of the 2A70 aluminum alloy deformed at high temperature, a series of isothermal compression tests were performed on a Gleeble - 1500 thermal simulator. On the basis of experiment data, a new constitutive model of the 2A70 aluminum alloy was constructed using an amended BP artificial neural network. True strain, strain rates and temperatures were input parameters of the model, while flow stress was the only output parameter. Computer results showed that the relative error of the new model was less than ±3% for the sampled data while it was less than ±6% for the non-sampled data. It is evident that the constitutive model constructed by BP ANN can accurately describe the flow rule of the 2A70 alloy hot deformation.

关 键 词：2A70铝合金 本构关系 热模拟试验 BP人工神经网络 

分 类 号：TG302[金属学及工艺—金属压力加工]