Fatigue & Fracture of Engineering Materials & Structures最新文献

The paper presents an analysis of the effect of random defects formed during additive manufacturing on fatigue behavior. A significant scatter in the fatigue strength was observed for the vertical build direction of the 316L stainless steel specimens. The multiple defects initiated the fatigue crack. The size of the selected critical defects was correlated with the failure probability. The defect criticality was defined using an analytical model of the normalized Crossland stress as a function of the defect size, location, and other defects affecting the local stress distribution. The components of the multiaxial fatigue criterion were calculated using the finite element method. The critical defect size distributions generated for variable defect density were analyzed using extreme value statistics. The Weibull distribution for fatigue strength was predicted using the simulation results and the El-Haddad model. The probabilistic modeling of multiple defects provided the predictive power of the scatter in fatigue properties.

Fatigue–creep damage-induced mechanical degradation has been a significant failure mode of proton exchange membrane (PEM) for fuel cells, but the damage behavior and intelligent assessment method are not well understood. A multitask time-series model is proposed to address this, which predicts fatigue–creep crack growth by incorporating in situ fatigue testing with mean stress effects. The results show that both fatigue and creep damage influence crack growth, with higher stress ratios accelerating it. The plastic zone size correlates with crack growth rate, highlighting the coupled fatigue–creep mechanisms. A damage assessment model is developed to quantify contributions from both factors. The approach, validated with K-fold cross-validation, demonstrates high predictive accuracy. SHAP analysis identifies stress ratio and stress intensity factor as key factors for creep and fatigue damage. This method improves prediction accuracy and enhances understanding of PEM degradation, offering insights for predictive maintenance and durable PEM design in fuel cells.

To improve the applicability of the extended finite element method (XFEM)–based cohesive zone model (CZM) for mixed-mode fracture problems, this study proposes a user-defined subroutine developed within the commercial finite element (FE) software Abaqus/Standard. A potential-based constitutive model with an explicit formulation is employed to improve convergence and describe nonlinear cohesive interactions. A high-performance parallelized computational module is implemented to enhance efficiency. Benchmark examples of pure mode I and mode II are included to verify the subroutine, while mixed-mode beam (MMB), L-shaped panel, and tension-shear specimens demonstrate its accuracy and computational performance in mixed-mode crack problems. Influencing factors, including element quality and load increment, on the accuracy of the numerical solution are analyzed. With appropriate modifications, the proposed formulation can be extended to other coupled cohesive models and FE platforms, offering broad applicability for a wide range of advanced fracture modeling scenarios.

To improve the applicability of the extended finite element method (XFEM)–based cohesive zone model (CZM) for mixed-mode fracture problems, this study proposes a user-defined subroutine developed within the commercial finite element (FE) software Abaqus/Standard. A potential-based constitutive model with an explicit formulation is employed to improve convergence and describe nonlinear cohesive interactions. A high-performance parallelized computational module is implemented to enhance efficiency. Benchmark examples of pure mode I and mode II are included to verify the subroutine, while mixed-mode beam (MMB), L-shaped panel, and tension-shear specimens demonstrate its accuracy and computational performance in mixed-mode crack problems. Influencing factors, including element quality and load increment, on the accuracy of the numerical solution are analyzed. With appropriate modifications, the proposed formulation can be extended to other coupled cohesive models and FE platforms, offering broad applicability for a wide range of advanced fracture modeling scenarios.

Comparing different fatigue analysis techniques is crucial when assessing the structural durability of complex welded structures. The structural stress, notch stress, and crack-propagation concepts were previously applied to the fatigue strength of large-scale welded K-joints commonly adopted in offshore structures under constant and variable amplitude loadings. Here, the K-nodes have been analyzed according to the peak stress method (PSM), which considers the weld toe and weld root as sharp V-notches and estimates rapidly the notch stress intensity factors (NSIFs) using rather coarse FE meshes. A fatigue design stress, the equivalent peak stress, and dedicated fatigue resistance curves are defined in the framework of the PSM, along with software to enable the automated application of the PSM in Ansys Mechanical. The accuracy of each fatigue design concept has been evaluated by comparing the relevant fatigue lifetime estimations with the experimental results. All investigated concepts delivered conservative results, however, with a different ranking.

The very high cycle fatigue (VHCF) behavior of AlSi10Mg specimens fabricated via selective laser melting (SLM) was systematically investigated under different printing strategies, including single-laser and dual-laser configurations. Experimental results revealed that fatigue cracks predominantly initiated at pores, with dual-laser SLM specimens exhibiting significantly higher pore density and reduced fatigue life compared to single-laser specimens. Among the examined factors, the depth of crack-initiating defects relative to the specimen surface emerged as the most critical determinant of fatigue performance, surpassing pore size and aspect ratio. Furthermore, this study proposed an L model, which explicitly incorporates defect location and geometry effects, and demonstrated its superior predictive accuracy compared to classical models like Murakami's. These findings provide important insights for optimizing additive manufacturing process parameters to enhance the VHCF performance of SLM-fabricated AlSi10Mg components.

The very high cycle fatigue (VHCF) behavior of AlSi10Mg specimens fabricated via selective laser melting (SLM) was systematically investigated under different printing strategies, including single-laser and dual-laser configurations. Experimental results revealed that fatigue cracks predominantly initiated at pores, with dual-laser SLM specimens exhibiting significantly higher pore density and reduced fatigue life compared to single-laser specimens. Among the examined factors, the depth of crack-initiating defects relative to the specimen surface emerged as the most critical determinant of fatigue performance, surpassing pore size and aspect ratio. Furthermore, this study proposed an L model, which explicitly incorporates defect location and geometry effects, and demonstrated its superior predictive accuracy compared to classical models like Murakami's. These findings provide important insights for optimizing additive manufacturing process parameters to enhance the VHCF performance of SLM-fabricated AlSi10Mg components.

Accurate assessment of engine block fatigue life (N_f) is critical for enhancing transportation system reliability, particularly through precise modeling of the fatigue crack propagation (FCP) rate (da/dN) in the upper main bearing wall. Existing da/dN models, however, exhibit limited N_f prediction accuracy due to neglecting structural and loading effects. This study first employs a component equivalence method to experimentally derive true da/dN curves for diverse upper main bearing wall configurations, with standard specimens as controls. Three machine learning (ML) models are then trained on these datasets, with the random forest algorithm selected for its superior accuracy to establish the final da/dN model. Incorporating crack closure effects via the virtual crack closure technique, simulated N_f predictions are generated. Fatigue tests on engine blocks validate the model's predictions, demonstrating a 13.8% improvement in accuracy over conventional approaches, attributable to the explicit integration of structural and loading parameters. The research method presented in this paper provides a more effective tool for the accurate establishment of the da/dN model for actual engineering components and the precise prediction of fatigue life, thereby laying a solid foundation for improving component reliability.

The reformer furnace tube serves as the crucial component in the steam reformer furnace. During the service exposure, the gradual accumulation of creep damage is the primary cause of the final failure of the furnace tube. To evaluate the creep damage evolution and predict the service time, a continuum damage mechanics (CDM) model is established based on the Kowalewski–Dyson model. This model integrates creep stress into the precipitate coarsening damage equation, accounting for its influence on activation energy in Ostwald ripening kinetics. Additionally, the stress-related uniaxial creep ductility is incorporated into the cavitation damage equation. The material parameter optimization procedure and the associated ABAQUS user subroutine UMAT of the proposed model are also developed. Different from the Kowalewski–Dyson model, the present model can predict the stress dependence of fracture strain and rupture time. The precipitate coarsening damage is predominant in the initial stage, while the cavitation damage accelerates the growth of total damage in the latter stage, ultimately resulting in creep fracture. Eventually, the total damage of the furnace tube made of HP40Nb alloy reaches its critical threshold at 101,450 h.

We conducted a statistical fatigue analysis of an additively manufactured aluminum alloy (AlSi10Mg) with five different sizes of specimens to establish probabilistic S-N curves and assess an anomalous behavior of strength deterioration in high-cycle and very-high-cycle regimes. For the five specimen types, S-N data were obtained by ultrasonic axial cycling at R = −1. Based on the fitted Basquin equations, the S-N data were converted to equivalent stress amplitudes at the same life. In general, the stress amplitudes can be regarded as normal, log-normal, and Weibull distributions, with the log-normal slightly overestimating the fatigue performance, whereas the Weibull slightly underestimated it, and the normal being the most suitable. Finally, we compared the normal distribution of fatigue resistance with the control volume V₉₀ of specimen types and plotted the strength deterioration from 10⁶ cycles to the extrapolated 10¹⁰ cycles, indicating that the specimen size effect diminishes as fatigue life increases.