Abstract In this study, we attempt to build a statistical (machine) learning model to predict the parabolic rate constant $$(k_{\text{P}} )$$ ( k P ) for the high-temperature oxidation of Ti alloys. Exploring the experimental studies on high-temperature oxidation of Ti alloys, we built our dataset for machine learning. Apart from the alloy composition, we included the constituent phase of the alloy, temperature of oxidation, time for oxidation, oxygen and moisture content, remaining atmosphere (gas except O 2 gas in dry atmosphere), and mode of oxidation testing as the independent features while the parabolic rate constant $$(k_{\text{P}} )$$ ( k P ) is set as the target feature. We employed three different ML models to predict the ‘ $$k_{\text{P}}$$ k P ’ for Ti alloys. Among the regression models, the gradient boosting regressor yields the coefficient of determination ( R 2 ) of 0.92 for $$k_{\text{P}}$$ k P . The knowledge gained from this study can be used to design novel Ti alloys with excellent resistance towards high-temperature oxidation. Graphic Abstract

中文翻译：

---

使用机器学习预测钛合金高温氧化的抛物线速率常数

摘要 在本研究中，我们尝试建立一个统计（机器）学习模型来预测 Ti 合金高温氧化的抛物线速率常数 $$(k_{\text{P}} )$$ ( k P )。探索钛合金高温氧化的实验研究，我们建立了我们的机器学习数据集。除了合金成分外，我们还包括合金的组成相、氧化温度、氧化时间、氧气和水分含量、剩余气氛（干燥气氛中除O 2 气体外的气体）和氧化测试方式作为独立特征而抛物线速率常数 $$(k_{\text{P}} )$$ ( k P ) 被设置为目标特征。我们采用了三种不同的 ML 模型来预测 Ti 合金的“$$k_{\text{P}}$$ k P ”。在回归模型中，对于 $$k_{\text{P}}$$ k P ，梯度提升回归器产生的决定系数 (R 2 ) 为 0.92。从这项研究中获得的知识可用于设计具有优异耐高温氧化性能的新型钛合金。图形摘要