Abstract The demand for energy generation with low carbon emissions evoked the development of ultra-super critical technology that allows operating steam turbines at high temperature and pressure conditions. However, operating at extreme conditions necessitates careful consideration of structural integrity which is affected by residual stresses. Welding is used for joining of components of steam turbines, but this process causes the formation of residual stresses of complex form. Careful investigation is necessary to understand the distribution of potentially detrimental residual stress fields. Eigenstrain theory was previously used for the development of the artificial intelligence based eigenstrain (AI-eig) contour method that allowed advanced modelling of the behaviour of Inconel alloy 740H under thermo-mechanical loading conditions. Models created using this method are capable of evaluating the residual stress fields in the whole specimen or in the parts and slices created using electric discharge machining (EDM). In the previous applications of the AI-eig contour method, the determination of the distribution of eigenstrain in as-welded and heat-treated specimens was followed by the calculation of volumetric residual stresses. In this study, long- and short-transverse components of the residual strains determined by the AI-eig contour method applied to EDM-cut surfaces of the parts of as-welded and heat-treated specimens were validated using the neutron strain scanning method. The results demonstrate the effectiveness of the integrative modelling approach that enables the determination of eigenstrains in the whole specimen and the calculation of residual strains before and after the machining process.

中文翻译：

---

中子应变扫描用于基于人工智能的特征应变轮廓方法的实验验证

摘要 对低碳排放能源的需求引发了超超临界技术的发展，该技术允许在高温高压条件下运行汽轮机。然而，在极端条件下运行需要仔细考虑受残余应力影响的结构完整性。焊接用于连接汽轮机部件，但该过程会导致形成复杂形式的残余应力。需要仔细调查以了解潜在有害残余应力场的分布。特征应变理论以前用于开发基于人工智能的特征应变 (AI-eig) 轮廓方法，该方法允许对 Inconel 合金 740H 在热机械载荷条件下的行为进行高级建模。使用这种方法创建的模型能够评估整个试样或使用放电加工 (EDM) 创建的零件和切片中的残余应力场。在 AI-eig 轮廓方法的先前应用中，在确定焊态和热处理试样中的本征应变分布之后，会计算体积残余应力。在这项研究中，使用中子应变扫描方法验证了由 AI-eig 轮廓法应用于焊接和热处理试样零件的 EDM 切割表面的残余应变的长和短横向分量。