PSI - Issue 68

2

Ela Marković et al. / Structural Integrity Procedia 00 (2025) 000–000

Ela Marković et al. / Procedia Structural Integrity 68 (2025) 345 – 350

346

1. Introduction Surface hardening is commonly employed to enhance durability and performance of highly stressed parts of components. Critical regions are selectively heat-treated to increase their hardness while preserving the integrity and ductility of the rest of the part, Cordovilla et al. (2016). At an early stage of the development and engineering design process, it is important to assess the load and durability of parts and components and to carry out computer simulations of their behaviour during exploitation. Modelling the mechanical behaviour of surface-hardened components has been a subject of research ever since the first analytical methods presented in papers by Delale and Erdogan (1982); Gu and Asaro (1997) and continues to be developed through the use of advanced computer-supported numerical methods in works by Aravind et al. (2018); Singh et al. (2022). Solving problems involving surface-hardened components usually requires either approximate analytical solutions derived from analytical solutions for homogeneous materials or numerical techniques such as the finite element method. In most commercially available numerical analysis software, analysing components with homogeneous properties is relatively straightforward, whereas components with variable properties require more complex modelling approaches. This challenge still makes the modeling of surface-hardened components a subject of research. Developed computational model enables the estimation of the expected number of load reversals to crack initiation i.e. failure (fatigue life) for surface-hardened steel components with stress concentrators. The model consists of two key parts, finite element analysis (FEA) of a component, based on which stress and strain amplitudes are determined, and a computational part, which enables capturing the number of load reversals to failure at each node of the component. The finite element (FE) model captures the hardness distribution in surface-hardened materials and includes the gradient in material strength from the high strength surface to the lower strength core. Approximation methods are used to define the cyclic stress-strain curves in relation to the hardness distribution, followed by the application of axial loading boundary conditions. After obtaining the distribution of strain amplitudes in the component, the number of load reversals to failure is estimated using a strain-based approach. 2. Finite element model A 2D parametric finite element model is developed in ANSYS ® to analyse the impact of surface hardening and different stress concentrators on the mechanical behaviour of components, Ansys Inc. (2013). The model is developed using APDL, a scripting language that facilitates the creation of parametric models, enabling customization of the simulation process and generation of various geometries. 2.1. Geometry and boundary conditions The steel specimens analysed in this study are thin strips with stress concentrators. The stress concentrators examined are semi-circular opposite single notches within finite-width thin elements, as classified by Pilkey (1997). Stress concentration refers to the area in a loaded component where stress is significantly higher than in the surrounding material due to geometric discontinuities or material inconsistencies. The presence of features such as threads, shoulders, holes, grooves, changes in cross-section, and welds alters the stress distribution, leading to localized high stresses, Pilkey (1997). The configuration of the specimen geometry under investigation is illustrated in Fig. 1a. Given that the specimen structure is thin and flat, with length and width much larger than its thickness, and loaded by in-plane forces (in this case, displacements), the model can be further simplified to 2D by applying the plane stress condition as was done in Kurowski (2017); Pilkey (1997). The boundary conditions causing the plane stress state are shown in Fig. 1b.