钛合金相变点预测模型的构建和评估

钛工业进展, 2016, 33(6): 28-31.

钛合金相变点预测模型的构建和评估

任军帅 ^1, , 张英明 ^2, , 吝靖玉 ^3, , 席恩平 ^4,

1.西北有色金属研究院,陕西西安,710016

2.西北有色金属研究院,陕西西安,710016

3.西北有色金属研究院,陕西西安,710016

4.西北有色金属研究院,陕西西安,710016

关键词：钛合金 , 相变点 , 合金元素 , 多元线性回归模型 , 人工神经网络模型

Construction and Evaluation of the Prediction Models of Titanium Alloys’ (α +β)/βTransus Temperature

Keywords： titanium alloys , transus temperature , alloying elements , multiple linear regression , artificial neural network

基于西北有色金属研究院实际生产中统计的321组钛合金铸锭化学成分与相变点数据,构建了预测钛合金(α+β)/β相变点的人工神经网络模型和多元线性回归模型,并对模型的准确性进行了评价分析。结果显示,多元线性回归模型的训练值及预测值与(α+β)/β相变点实际值的相关性系数分别为0．76105和0．80993,而人工神经网络模型的相关性系数分别为0．92721和0．81851,具有更好的相关性。人工神经网络模型的平均绝对误差为4．02℃,相比多元线性回归模型(平均绝对误差为5．11℃)具有更高的精度,可以更好地描述合金元素与钛合金(α+β)/β相变点之间的非线性关系。

In this paper, artificial neural network ( ANN ) and multiple linear regression ( MLR ) model were developed and analyzed based on 321 sets of data, incuding alloying elements and (α + β)/β transus data from Northwest Institute for Nonferrous Metal Research, and the accuracy of both models were evaluated. Results show that the correlation coefficients between the MLR model’ s training data and (α + β)/βtransus, the prediction data and (α +β)/βtransus were 0. 761 05 and 0. 809 93, respectively. The correlation coefficients between the ANN model’ s training data and (α + β)/β transus, the prediction data and (α + β)/β transus were 0. 927 21 and 0. 818 51, respectively. ANN predictions are in better agreement with the experimental data. The mean absolute error of the MLR model is 5. 11 ℃, and the mean absolute error of the ANN model is 4. 02 ℃. ANN model can more accurately describe the non-linear relationship between (α +β)/βtransus temperature and alloying elements.

参考文献

[1]	鲁海洋;孙宇;朱艳春;曾卫东.基于神经网络的BT20钛合金本构关系模型[J].钛工业进展,2010(4):20-24.
[2]	Wenwen Peng;Weidong Zeng;Qingjiang Wang;Hanqing Yu.Comparative study on constitutive relationship of as-cast Ti60 titanium alloy during hot deformation based on Arrhenius-type and artificial neural network models[J].Materials & design,2013Oct.(Oct.):95-104.
[3]	孙宇;曾卫东;赵永庆;张学敏;马雄;韩远飞.基于BP神经网络的置氢TC21合金力学性能预测[J].稀有金属材料与工程,2012(6):1041-1044.
[4]	岳洋;朱景川;刘勇;王洋;杨夏炜.基于BP神经网络的TA15钛合金热变形工艺-性能预报[J].稀有金属材料与工程,2009(10):1811-1814.
[5]	邵一涛;曾卫东;韩远飞;周建华;王晓英;周义刚.基于BP人工神经网络的TC17钛合金显微组织-力学性能关系预测[J].稀有金属材料与工程,2011(2):225-230.
[6]	段园培;李萍;薛克敏;甘国强;曹婷婷.TB8钛合金固溶组织研究及神经网络预测[J].稀有金属材料与工程,2012(8):1426-1430.
[7]	孙宇;曾卫东;赵永庆;韩远飞;邵一涛;周义刚.应用人工神经网络研究化学元素对钛合金相变点的影响[J].稀有金属材料与工程,2010(6):1031-1036.
[8]	Guo Z;Malinov S;Sha W.Modelling beta transus temperature of titanium alloys using artificial neural network[J].Computational Materials Science,20051(1):1-12.

上一张下一张

计量

下载量()
访问量()

作者相关

文章评分

您的评分：
1

0%
2

0%
3

0%
4

0%
5

0%

材料期刊网