摘要
通过试验,研究了受过循环变形、具有稳定超弹性变形性能的形状记忆合金丝在拉伸到不同应变幅值条件下卸载的超弹性变形行为。根据试验测得的结果,提出了基于神经网络的形状记忆合金超弹性本构关系模型,并把模型计算的结果和实验数据进行了比较分析,结果表明,该模型具有很高的精度。该模型避免了已有模型在参数确定上的困难,具有一定的工程应用价值,为建立形状记忆合金本构模型提供了一个新的思路。
Superelasticity is one of the most important properties of shape memory alloy. In this mation behavior of NiTi shape memory alloy subjected to cyclic loading with stable superelastic tally. According to the test data, a constitutive model for the superelasticity of shape memory paper, the superelastic defor- ity allo artificial neural network. The numerical results agree well with experimental observations, whic is investigated experimen- y is put forward based on h verifies the constitutive model has a high accuracy. With this model, the difficulties on the determination of the parameters for other models can be avoided, and which leads to a practical engineering application of this model. Thus, a new attempt is provided in the present paper for building the constitutive model of shape memory alloy.
出处
《振动工程学报》
EI
CSCD
北大核心
2006年第1期109-113,共5页
Journal of Vibration Engineering
基金
国家杰出青年科学基金资助(50025823)
海外青年学者研究基金资助(50328807)
关键词
形状记忆合金
超弹性
本构关系
神经网络
shape memory alloy
superelastieity
constitutive relationship
neural network
