基于机器学习算法对铝合金的应力集中系数预测
Prediction of Stress Concentration Coefficient of Aluminum Alloy Based on Machine Learning Algorithm
摘要: 本文将6061铝合金试件作为疲劳拉伸实验对象，得到不同损伤程度的铝合金试件，借助工业CT扫描系统和AVIZO软件获取试件内部的缺陷特征数据。通过ABAQUS仿真分析，得出疲劳循环周次与应力集中系数呈线性关系，因此应力集中系数可以作为疲劳寿命的表征参数。最后，借助机器学习算法准确预测不同缺陷特征信息的应力集中系数。
Abstract:
In this paper, 6061 aluminum alloy specimens are used as fatigue tensile test objects to obtain aluminum alloy specimens with different damage degrees. With the help of industrial CT scanning system and AVIZO software, defect feature data inside the specimens are obtained. Through ABAQUS simulation analysis, it is concluded that the fatigue cycle has a linear relationship with the stress concentration coefficient, so the stress concentration coefficient can be used as a characteris-tic parameter of the fatigue life. Finally, the stress concentration coefficient of different defect fea-ture information was accurately predicted by machine learning algorithm.
参考文献
|
[1]
|
刘静安, 潘伟津, 罗立新, 等. 铝合金锻件主要缺陷的特征及产生原因[J]. 轻合金加工技术, 2014, 42(1): 54-56.
|
|
[2]
|
Romano, S., Brandao, A., Gumpinger, J., et al. (2017) Qualification of AM Parts: Extreme Value Statistics Ap-plied to Tomographic Measurements. Materials and Design, 131, 32-48. [Google Scholar] [CrossRef]
|
|
[3]
|
Fedor, F., Manfred, H., Norbert, H., et al. (2018) Probabilistic Fa-tigue-Life Assessment Model for Laser-Welded Ti-6Al-4V Butt Joints in the High-Cycle Fatigue Regime. International Jour-nal of Fatigue, 116, 22-35. [Google Scholar] [CrossRef]
|
|
[4]
|
宋哲, 吴圣川, 胡雅楠, 等. 冶金型气孔对熔化焊接7020铝合金疲劳行为的影响[J]. 金属学报, 2018, 54(8): 811-820.
|
|
[5]
|
谢建新, 宿彦京, 薛德祯, 等. 机器学习在材料研发中的应用[J]. 金属学报, 2021, 57(11): 1343-1361.
|
|
[6]
|
Wigant, C.C. and Stephens, R.I. (1987) Report No. 870096 Society of Au-tomotive Engineers. Warrendale, PA.
|
|
[7]
|
Hasan, A. and Alshibli, K.A. (2010) Experimental Assessment of 3D Parti-cle-to-Particle Interaction within Sheared Sand Using Synchrotron Microtomography. G’eotechnique, 60, 369-379. [Google Scholar] [CrossRef]
|