基于随机森林和支持向量机的Mo-Nb合金本构模型
Constitutive model of Mo-Nb alloy based on random forest and support vector machine
黄文杰 1王克鲁 1鲁世强 1钟明君 1李鑫 1曾权 1周潼 1汪增强1
作者信息
- 1. 南昌航空大学 航空制造工程学院,南昌 330063
- 折叠
摘要
在变形温度为900～1200℃、应变速率为0.01～10 s-1条件下,采用Gleeble-3800型热模拟试验机对Mo-Nb合金进行等温恒应变速率压缩实验,研究Mo-Nb合金的流动应力行为,并采用随机森林和支持向量机的方法建立该合金的本构关系模型.结果表明:Mo-Nb合金是负温度和正应变速率敏感型材料,其流动应力随变形温度升高和应变速率降低而减小;随机森林和支持向量机本构关系模型的训练样本的相关系数和平均相对误差分别为0.989、0.998及2.41%、0.94%,测试样本的相关系数和平均相对误差分别为0.991、0.996及2.47%、1.4%,二者都具有较好的预测能力;支持向量机本构关系模型精度高于随机森林,因此,支持向量机本构关系模型更适于预测Mo-Nb合金的流动应力.
Abstract
The flow stress behavior of Mo-Nb alloy was studied by isothermal constant strain rate compression experiments on Gleeble-3800 thermal simulation tester at the deformation temperature of 900-1200℃and strain rate of 0.01-10 s-1.The constitutive model of the alloy was established by random forest and support vector machine.The results show that Mo-Nb alloy is a negative temperature and positive strain rate sensitive material,and its flow stress decreases with increasing deformation temperature and decreasing strain rate.The correlation coefficients and average relative errors for the training samples of random forest and support vector machine constitutive models are 0.989,0.998 and 2.41%,0.94%,respectively,and the correlation coefficients and average relative errors for the test samples are 0.991,0.996 and 2.47%,1.4%,respectively,both of which have good prediction ability.The accuracy of the support vector machine constitutive model is higher than that of the random forest,so the support vector machine constitutive model is more suitable for predicting the flow stress of Mo-Nb alloy.
关键词
Mo-Nb合金/本构模型/随机森林/支持向量机Key words
Mo-Nb alloy/constitutive model/random forest/support vector machine引用本文复制引用
出版年
2024
NETL
NSTL
万方数据