钢材力学性能的机器学习
Machine Learning of Mechanical Properties of Steels
摘要：
机械性能是结构材料的基本性能。从NIMS数据库中选择关于钢的四个机械性能（即，疲劳强度、拉伸强度、断裂强度和硬度）的分析的360个数据，包括碳钢和低合金钢。对360个数据应用五种机器学习算法来预测机械性能，随机森林回归说明了最佳性能。通过随机森林和符号回归进行特征选择，得到回火温度和碳、铬和钼合金元素对钢的机械性能的四个最重要的特征。此外，通过符号回归生成数学表达式，表达式明确预测四个机械性能中的每一个如何定量地随四个最重要的特征而变化。目前的工作证明了符号回归在发现新的先进材料中的巨大潜力。
英文摘要：
The mechanical properties are essential for structural materials. The analyzed 360 data on four mechanical properties of steels, viz. fatigue strength, tensile strength, fracture strength, and hardness, are selected from the NIMS database, including carbon steels, and low-alloy steels. Five machine learning algorithms were applied on the 360 data to predict the mechanical properties and random forest regression illustrates the best performance. The feature selection was conducted by random forest and symbolic regressions, leading to the four most important features of tempering temperature, and alloying elements of carbon, chromium, and molybdenum to the mechanical properties of steels. Besides, mathematic expressions were generated via symbolic regression, and the expressions explicitly predict how each of the four mechanical properties varies quantitatively with the four most important features. The present work demonstrates the great potential of symbolic regression in the discovery of novel advanced materials.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
