npj Computational Materials

Using machine learning and a data-driven approach to identify the small fatigue crack driving force in polycrystalline materials（用机器学习-数据驱动法识别多晶材料的小疲劳裂纹驱动力）
Andrea Rovinelli, Michael D. Sangid, Henry Proudhon & Wolfgang Ludwig
npj Computational Materials 4:35 (2018)
doi:s41524-018-0094-7
Published online:16 july 2018
Abstract| Full Text | PDF OPEN

摘要:小裂纹的扩展是导致结构部件进入疲劳期的主要因素。尽管人们对此有很大的兴趣，但就裂缝扩展的方向和速度而言，小裂缝的生长标准尚未确定。本研究提出了一种识别微结构小疲劳裂纹驱动力的新方法。采用贝叶斯网络和机器学习技术可识别微机械和微结构变量，以及它们对疲劳裂纹的扩展方向和扩展速率的影响关系。多模态数据集结合了多晶聚合体内原位扩展的小裂纹的高分辨率4D实验数据和晶体塑性模拟数据，用来提供训练数据。相关变量构成解析表达式的基础，因此代表方向和速率方程的小裂纹驱动力。我们对该表达式捕获所观察实验行为的能力作了量化，并与直接来自贝叶斯网络的结果和文献中常见的疲劳指标作了比较。结果表明，采用所提出的分析模型可靠地预测了小裂纹扩展的方向，比其他疲劳指标更为有利。　　

Abstract:The propagation of small cracks contributes to the majority of the fatigue lifetime for structural components. Despite significant interest, criteria for the growth of small cracks, in terms of the direction and speed of crack advancement, have not yet been determined. In this work, a new approach to identify the microstructurally small fatigue crack driving force is presented.Bayesian network and machine learning techniques are utilized to identify relevant micromechanical and microstructural variables that influence the direction and rate of the fatigue crack propagation.A multimodal dataset, combining results from a high-resolution 4D experiment of a small crack propagating in situ within a polycrystalline aggregate and crystal plasticity simulations, is used to provide training data.The relevant variables form the basis for analytical expressions thus representing the small crack driving force in terms of a direction and a rate equation.The ability of the proposed expressions to capture the observed experimental behavior is quantified and compared to the results directly from the Bayesian network and from fatigue metrics that are common in the literature.Results indicate that the direction of small crack propagation can be reliably predicted using the proposed analytical model and compares more favorably than other fatigue metrics.

Editorial Summary

Crack propagation: machine learning identifies micromechanical variables（裂缝扩展：机器学习识别微机械变量）

机器学习技术可以识别钛合金中小裂纹扩展方向背后的复杂变量。由美国普渡大学Michael Sangid领导的团队，采用机器学习建立了两个独立的贝叶斯网络，分析了钛合金原位疲劳循环过程中获得的衍射数据和X线断层影像数据。第一主应力轴在特定方向上的取向和最大分辨剪切应力，与裂纹扩展最为相关，将其纳入关系分析中，用以描述裂纹扩展方向的概率。该分析表达式再现了实验结果，比以前文献的预测更为可靠。这种半监督机器学习方法可以帮助我们识别其他复杂工程问题中的驱动力。

A machine learning technique can identify the complex variables behind the propagation direction of small cracks in a titanium alloy.A team led by Michael Sangid at Purdue University in the U.S.A built two separate Bayesian networks using machine learning to analyse diffraction and tomography data acquired during in situ fatigue cycling of a titanium alloy.The orientation of the first principal stress axis in a specific direction and the maximum resolved shear stress were the most strongly correlated with crack propagation, and were incorporated into an analytical relationship to describe the probability of the crack propagation direction.This analytical expression reproduced experimental results and was more reliable than previous literature predictions. This sort of semi-supervised machine learning methodology may help us identify driving forces in other complex engineering problems.