International Journal of Fatigue最新文献

The incorporation of fibers into concrete enhances its fatigue resistance while also increasing the variability in flexural fatigue life, necessitating the development of advanced predictive models. To address this challenge, this study innovates by developing a stacking ensemble prediction model aimed at accurately predicting the flexural fatigue life of fiber-reinforced concrete (FRC). This model integrates the Deep Autoencoder Network, XGBoost, Random Forest, and Deep Neural Networks optimized by the Grey Wolf Optimizer algorithm. Initially, a dataset on the flexural fatigue of FRC was meticulously established. Subsequently, the model was rigorously evaluated using this dataset, with the results demonstrating its effectiveness in accurately predicting the fatigue life of FRC. Furthermore, SHAP analysis was utilized to interpret the relationship between input features and the fatigue life of FRC. In essence, this research offers a comprehensive and flexible predictive framework for the fatigue life of FRC, enhancing the comprehension and practical utilization of this material in construction and engineering projects, and presenting a promising avenue for future advancements in materials science and engineering.

This paper questions the recommendation regarding the use of standard specimen geometries, (Type I, Type II, and Type IV), for estimating the tensile quasi-static and fatigue properties of structural epoxy adhesives. The work presents results from an experimental program investigating the performance of structural epoxy adhesives indicating a significant effect of the specimen geometry, especially when referring to fatigue loading. Simple finite element models are also developed to facilitate the comparison of the stress distribution along the three specimen geometries. The fatigue experimental results allowed the derivation of probabilistic S-N curves, showing higher fatigue sensitivity of Type I specimens compared to Type II and IV. Furthermore, probability distribution function (PDF) curves of the equivalent static strength estimated by using Sendeckyj’s wear-out model attributed lower mean strength and higher variance for Type I specimens validating the fatigue data.

Corrosion fatigue damage occurs when metallic materials are subjected to cyclic loading in a corrosive medium. In this study, a phase field framework is proposed to predict the corrosion fatigue of carbon steels. The coupling effect of fatigue and corrosion is explicitly implemented in the proposed phase field framework by coupling the displacement field, electrochemical field and phase field. A degradation function of the interface free energy density with the consideration of elastic and plastic strain energies is introduced to account for the fatigue damage accumulated during the corrosion fatigue process. The applicability of this framework is validated by accurately capturing the pure fatigue and corrosion fatigue behaviors of compact tension specimens, particularly the acceleration effect of corrosion on the fatigue crack growth. The propagation morphology and rate of the corrosion fatigue crack in single pit and multiple pit models are studied. The distribution of stress state and strain energy density induces the directionality of crack propagation. The influence of loading frequency on the corrosion fatigue process is discussed in detail. Due to the corrosion-fatigue coupling effect, the corrosion rate increases with increasing of the loading frequency, resulting in an accelerated corrosion fatigue process. Moreover, the significance of the plasticity in the prediction of corrosion fatigue is emphasized.

The asymmetric cyclic loading process occurs in aero-engine turbine discs. A constitutive model that accurately describes the cyclic elastoplastic behaviour of the material is important for structural design and low cycle fatigue life prediction of turbine discs. In this paper, the low cycle fatigue test of FGH95 was carried out at 620℃ under 0.5%, 0.6%, 0.7%, 0.8%, 0.9%, 1.0% and 1.1% strain amplitude with strain ratio equal to 0. The material exhibited cyclic hardening followed by cyclic softening at high strain amplitudes and cyclic softening at low strain amplitudes. The mean stress relaxation rate was similar for each strain amplitude. In addition, the evolution of effective stress and back stress was obtained through the method of internal stress division. Then, the relationship between the change in stress amplitude and the change in internal stress was discussed. The results showed that with cyclic loading, cyclic hardening/softening of FGH95 was affected by the competition mechanism of back and effective stresses. Considering that slip deformation and crystal lattice rotation coexist in plastic deformation, the dynamic recovery term of the Abdel-Karim and Ohno model was used. In order to characterize the different magnitudes of back stress change in materials at different plastic strain intervals, a dynamic recovery term coefficient was introduced to the dynamic recovery term and the critical surface of the back stress. The modified model was used to compare with experimental results. Then, it gives a good description of the material’s mean stress relaxation and strain amplitude variation and gives good agreement on the hysteresis loop.

In this study, a novel non-contacting electromagnetic cold expansion process was proposed to improve the fatigue performance of hole component using a single power supply and a single coil. The stress state and deformation of the 6063-T6 aluminum alloy hole component during the electromagnetic strengthening process were investigated through numerical simulation. The residual stress around the hole edge was measured using XRD. Fatigue testing was performed to verify the effectiveness of the process on improving the fatigue life of the hole component. Results showed that the proposed electromagnetic strengthening process could effectively improve the fatigue life of the hole component. Fatigue life of the specimen via electromagnetic treatment is 3.42 times of that of the original specimen at a maximum stress of 120 MPa. Simulation results indicated that the generation of compressive residual stress was attributed to the falling stage of the pulse current, and the maximum compressive residual stress was −102 MPa.

Pitting corrosion notably increases stress concentration, thus accelerating fatigue crack initiation and propagation at the pit base. This research quantifies the relationship between pit dimensions and fatigue crack development by introducing crack extension ratios (γ_a​ and γ_c)​. Employing a series of finite element models, the influence of pit characteristics on the stress intensity factor (K) at the crack tip is assessed. Based on these assessments, a formula for computing the K-value is proposed. Furthermore, a two-stage fatigue life prediction model for both partial and complete penetration is developed using the Paris equation. Results indicate that pits substantially affect the K-value during partial penetration. Specifically, as γ_a approaches zero, the K-value approaches zero, and as γ_a approaches one, it aligns with the K-value of a semi-elliptical surface crack. Conversely, in the complete penetration phase, the influence of the pit on the K-value is negligible, and the K-value can be calculated according to the plate with a central penetrating crack. Experimental validation confirms that the model generally maintains a prediction error within 10%.

Additive manufacturing (AM) has broad application prospects in the field of aviation, whereas it remains a challenge to fully eliminate the manufacturing defects. As a result, the fatigue performance is poor and has a large scatter. The fatigue performance and life prediction method have become a major hindrance for the application of AM in aviation field. In this paper, the microstructure and defect characteristics of SLM Al-Mg-Sc-Zr alloy that built in transverse direction (TD) and parallel direction (PD) were studied. Then the fatigue performance was investigated, and the relationship between critical defect characteristics and fatigue life was analyzed qualitatively and quantitatively based on fracture analysis. The microstructure exhibits a bimodal grain structure with fine grain size. The defect study shown that there are more large-size defects and higher porosity in the PD specimens. The fatigue life is significantly influenced by the location, size and circularity of critical defects and stress amplitude. Correlation analysis suggested that the defect location is the most influential factor on fatigue performance, followed by stress amplitude, defect circularity and defect size. Finally, a fatigue life prediction model based on stress amplitude and defect characteristics of the critical defects was proposed.

Additive manufacturing (AM) and in particular laser-powder bed fusion has become a popular manufacturing techniques in recent years due to its significant advantages; however, the mechanical behavior of AM components often varies from components fabricated using conventional processes. For example, the fatigue behavior of components made by AM processes is heavily influenced by process-related defects and residual stresses in addition to applied stress amplitudes, stress ratio and surface conditions. Accounting for the interaction of these effects in fatigue design is difficult by means of traditional fatigue assessment concepts. Machine learning algorithms offer a possibility to account for such interactions and are easily applied once trained and validated. In this study, machine learning algorithms based on gradient boosted trees with the SHapley Additive exPlanation framework are used to predict defect location and fatigue life of additive manufactured AISI 316L specimens in as-built and post-treated manufacturing states, while also facilitating the understanding of the importance and interactions of various influencing factors.

Reduced Activation Ferritic-Martensitic (RAFM) steels are the candidate structural steels for In-Vessel Components of fusion reactors. Since the operation of a tokamak-type fusion reactor is cyclic by its nature, thermomechanical fatigue will be one of the limiting factors defining the life of the plasma-facing components exposed to neutron irradiation. The assessment of fatigue life requires considerable efforts in terms of low cycle fatigue experiments, which is extremely complicated on neutron-irradiated specimens inside the hot cell environment. Performing tensile tests is instead much faster, technically easier and more cost effective than performing fatigue tests especially on neutron-activated specimens. Therefore, a method for predicting fatigue life based on the tensile properties would be an important asset to assess the design of IVC when experimental data on fatigue are not available. Here, several fatigue life prediction methods based on Universal Slopes Equation were assessed based on the fatigue test results of RAFM steels in the irradiated and non-irradiated conditions to choose the best method. Analysis of the available fatigue database showed an effect of test medium and specimen size on the fatigue life. This effect was quantified and added to the selected method by means of scaling factors. The chosen method with the scaling factors was able to predict the fatigue life of irradiated and non-irradiated RAFM steels with an accuracy of 95% within the sleeve of factor three. The modified equations were then used to predict the fatigue life of irradiated RAFM steels at irradiation doses for which the fatigue data is not available.