Compilation of LCF and TMF Data for Single-Crystal and Directionally-Solidified Ni-base Superalloys (All Orientations)

This Excel file contains a compilation of low cycle fatigue (LCF) and thermomechanical fatigue (TMF) data for single-crystal (SX) and directionally-solidified (DS) Ni-base superalloys that are typically used for blades and vanes in gas turbines. Most of the data is generated by strain-controlled uniaxial fatigue tests. Both continuous cycling and cycles with dwells (hold times) at either the maximum or minimum of the cycle are included in the dataset. Several different alloys with similar mechanical behaviors are included in the dataset. The dataset only includes materials that have undergone a solution and age heat treatment typical of that applied to blade and vane components. All the fatigue data in this dataset is limited to uniaxial cyclic loading. The loading direction with respect to the crystal orientation or solidification direction varies. The data in this dataset has been extracted from the public domain. The sources of the data are included in the Excel file. This data is used to train a Probabilistic Physics-guided Neural Network (PPgNN) that outputs a strain-life curve, including the mean response and 95% confidence intervals. The INPUTS and OUTPUTS used in the PPgNN are highlighted in the Excel file.本Excel文件汇集了适用于燃气轮机叶片及导叶的单晶（SX）和定向凝固（DS）镍基高温合金的低周疲劳（LCF）与热机械疲劳（TMF）数据集。本数据集绝大多数数据由应变控制单轴疲劳试验生成，涵盖连续循环加载以及在循环最大或最小载荷处带有保载时间的循环加载两种加载模式。 数据集包含多种力学行为相近的合金，且仅纳入经过叶片、导叶组件典型固溶及时效热处理的材料。本数据集收录的所有疲劳数据均限定为单轴循环加载，加载方向相对于晶体取向或凝固方向存在差异。所有数据均取自公开领域，数据来源已在Excel文件中列明。 该数据集用于训练概率物理引导神经网络（PPgNN），该网络可输出应变寿命曲线，包括平均响应结果与95%置信区间。Excel文件中已高亮标注该PPgNN所用的输入项与输出项。