The present-day interest in the use of composite-steel joints in primary marine applications requires an in-depth knowledge of the fatigue performance of thick adhesive joints. This paper reports on experimental tests, quasi-static and fatigue, performed on unaged and aged bi-material double strap joints with thick adhesive bondlines. The specimens are monitored by Digital Image Correlation and Infrared Thermography techniques to evaluate the types and extent of damage that occur in the joint during fatigue loading. The S-N curve of the aged joint is evaluated along with it's fatigue limit. It is found that the unaged specimens fail due to cohesive damage, hackles and disbond at the adhesive-adherend interface and the aged specimens failed due to delamination within the composite. The specimens that survived fatigue loading showed similar residual strength as specimens that were not fatigue loaded.

In order to study the fracture properties of dam concrete under post-peak cyclic loading, wedge splitting tests with three loading rates (0.001 mm/s, 0.01 mm/s, 0.1 mm/s) were performed on notched cubic dam concrete specimens. Meanwhile, the acoustic emission (AE) and digital image correlation (DIC) technologies were used to record the crack propagation process of specimens. Test results show that the fracture of dam concrete has a significant rate effect: with the loading rate increases, the peak load increases, the slope of the post-peak P-CMOD curve gradually decreases and the stiffness degradation of dam concrete becomes more serious. The cumulative AE count shows a step increasing trend and has a Kaiser effect. The Kaiser effect decreases with the post-peak cyclic loading procedure, and with the loading rate increases, the Kaiser effect increases. With the increasing of loading rate, AE energy fluctuates violently and b value fluctuates frequently, indicating the damage of dam concrete becomes more serious. As the loading procedure, the damage of the specimen accumulates gradually, and the strain recovery rate decreases gradually. With the loading rate increases, the strain recovery rate decreases and the permanent crack increases. Based on the fictitious crack model, the effective crack length shows a gradual and steady rising trend. As the loading rate increases, the growth rate of the effective crack length becomes large.

In this study, high-speed ballistic impact tests were conducted on GH4169 alloy samples with the aeroengine compressor blade leading edge feature to simulate the notch-type foreign object damages (FOD). Macroscopic and microscopic characterization of FOD and high cycle fatigue tests were performed to investigate the effect of FOD depth on GH4169 alloy fatigue strength along with numerical analysis using Kitagawa-Takahashi diagram. Results show the incident side of notch-type FOD is relatively smooth, whereas the exit side is rugged. The FOD depth ranges from 0.18mm to 1.33mm, and the fatigue strength of damaged samples is 37.93%~97.04% of the undamaged samples. As FOD depth increases, damage length, material losses and stress concentration coefficient of the FOD increase significantly along with the increasing adiabatic shear bands, micro voids and cracks, resulting in fatigue strength reduction. Numerical analysis indicates that the Kitagawa-Takahashi diagram can provide a basic model for the design of FOD tolerance.

This study aims to develop a high temperature LCF test method using a non-standard miniature thin-plate (MTP) specimen in order to characterize cyclic visco-plasticity behavior of component materials. For demonstration, fully reversed strain-range controlled LCF and creep-fatigue (CF) tests at 600 °C have been performed for a martensitic steel using both standard-sized full-scale (SSFS) and MTP specimens. A scaling factor is determined using cyclic visco-plastic finite element (FE) for geometry constraint evaluation and data conversion based on the reference strain approach. The equivalent energy principal is proposed to assess the geometry constraint effect that non-standard MTP specimen has. The high temperature LCF results from the MTP specimen based on the proposed testing methodology have shown a good agreement with SSFS specimen data under equivalent conditions. The methodology can therefore be used to conduct accurate transferability to achieve equivalent LCF behavior between the conventional standard specimen and the MTP specimen.

Biaxial (axial and torsional loading) static tests were performed for the first time on EN-GJS-600–3 ductile cast iron tubular specimens obtained reproducing the solidification conditions typical of thick-walled castings. The experimental results were elaborated to determine the yield and fracture loci of the material, which exhibited significant deviations from those predicted by the Von Mises and Mohr-Coulomb criteria usually adopted for steels and grey cast iron, respectively. For this purpose, several alternative criteria proposed in the technical literature, some of them specifically devised for composite materials, have been calibrated and compared to account for the peculiar mechanical properties of this natural composite material.

This paper develops a data-driven autonomous method for detection of fatigue damage and classification of the associated damage risk in mechanical structures, based on ultrasonic signal energy. The underlying concept is built upon attenuation of the signal and stability of the attenuation process. The attenuation provides pertinent information for damage quantification, whereas the stability represents resistance towards the fatigue damage growth. The proposed neural network (NN) model has been trained using the scaled conjugate-gradient back-propagation method. The NN model is capable of damage detection and damage classification into five classes of increasing risk. The Daubechies wavelet transform has been used to reduce the noisy pattern of the ultrasonic signal energy by using the associated approximation coefficients. The results show that the proposed method of approximation signal energy can detect and classify the damage with an accuracy of up to ∼ 9 8 . 5 % .

The present study is aimed at validation of notch stress/ strain estimation schemes such as classical Neuber, Hoffmann-Seeger and recently developed Ince-Glinka method for Nuclear piping material (low C-Mn steel). The study has considered different constraints, loading conditions, various hole sizes to accommodate strain gradient variations and equivalent peak strains. The notch stress field evaluated using these schemes is compared with corresponding stress using elastic-plastic Finite Element (FE) analyses. The comparisons have brought out that the Hoffmann-Seeger scheme results in reasonably accurate assessment of stress localization nearly for all constraint geometries, loadings and strain gradients. However, the classical Neuber scheme is more suitable for low constraint geometries and intermediate constraint geometries whereas it results in under-estimation of maximum principal stress for high constraint geometries, thereby leading to over-prediction of fatigue life. Further, the suitability of energy equivalence equations of Ince-Glinka model for individual stress components, has been reviewed.

Fatigue fracture is one of the most common metallic component failure cases in manufacturing industries. The observation on fractography can provide the direct evidence for failure analysis. In this study, an image segmentation method based on Fully Convolutional Networks (FCNs) was proposed to figure out the boundary between fatigue crack propagation and fast fracture regions from optical microscope (OM) fractography images. Furthermore, novel morphing-based data augmentation method was adopted to enable few-shot learning of sample images. The proposed framework can successfully segment two categories, namely the crack propagation and fast fracture regions, thus differentiating the boundary of two regions in one image. The artificial intelligence (AI) assisted fatigue analysis architecture can complete the failure analysis procedure in 0.5 second and prove the feasibility of fatigue failure analysis. The segmentation accuracy of developed network achieves 95.4% for the fatigue crack propagation region, as well as 97.2% for the fast fracture region, which possesses comparable accuracy to the segmentation results using Mask R-CNN Regional Convolutional Neural Network (Mask R-CNN), one state-of-the-art deep learning networks.

Neural network (NN) models have made a significant impact on fatigue-related engineering communities and are expected to increase rapidly soon due to the recent advancements in machine learning and artificial intelligence. A comprehensive review of fatigue modeling methods using NNs is lacking and will help to recognize past achievements and suggest future research directions. Thus, this paper presents a survey of 251 publications between 1990 and July 2021. The NN modeling in fatigue is classified into five applications: fatigue life prediction, fatigue crack, fatigue damage diagnosis, fatigue strength, and fatigue load. A wide range of NN architectures are employed in the literature and are summarized in this review. An overview of important considerations and current limitations for the application of NNs in fatigue is provided. Statistical analysis for the past and the current trend is provided with representative examples. Existing gaps and future research directions are also presented based on the reviewed articles.

Fatigue behavior of Al-Al and Al-steel refill friction stir spot welding joints Haokun Yang1a*, Detao Cai2,3a, Jiaxin Li1, Chin Yung Kwok1, Yunqiang Zhao3, Yu Li4, Haisheng Liu4, Waiwah Lai11 Smart Manufacturing Division (SMD), Hong Kong Productivity Council (HKPC), Hong Kong 999077, People’s Republic of China2 State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha, Hunan 410082, People’s Republic of China3 China-Ukraine Belt and Road Joint Laboratory on Materials Joining and Advanced Manufacturing, China-Ukraine Institute of Welding Guangdong Academy of Sciences, Guangzhou, Guangdong 510650, People’s Republic of China4 Center for Industrial Analysis and Testing, Guangdong Academy of Science, Guangzhou, Guangdong 510650, People’s Republic of Chinaa These authors contribute equallyCorresponding author: Haokun Yang, +852 27885679,[email protected]: Refill friction stir spot welding, Fatigue fracture, Aluminum, Steel

The paper discusses various partial solutions used for estimating fatigue life under variable amplitude multiaxial loading in the high-cycle fatigue domain. The concurring effects are treated, and their proposed solutions are commented upon. The major focus is on the categories of the phase shift effect and of cycle counting, and on the scope and quality of data, which support discussed theories. Results of own new experimental data set on specimens from S355 steel are provided. Fatigue life estimates for McDiarmid and Findley multiaxial methods and for two different methods of load path decomposition to cycles are shown to highlight some of the points open for discussion. It is concluded that the available experimental data are not sufficient to substantiate a clear decision to follow a definite algorithm.

The state function based on fatigue accumulation model has a great influence on accuracy of fatigue reliability analysis of components. A nonlinear fatigue accumulation model considering the interaction of loads is proposed in this paper. By introducing the square ratio of front and back loads as the new load interaction factor into the conventional toughness exhaustion model, thus it is improved to reflect the load sequence and interaction effect simultaneously. Moreover, the fatigue state function based on proposed model is constructed, which is analyzed by the probability density evolution method. The time-varying fatigue reliability curve is obtained by analyzing the generalized density evolution equation. The proposed model are verified using four experimental data, and results shows that the predication residual life fraction such obtained is more accurate. And accuracy and efficiency of the proposed time-varying fatigue reliability analysis method is validated using multi-level loading experimental data.

This work investigates the influence of temperature on the creep damage behaviors of red sandstone. The samples treated at 25 ℃, 200 ℃, 400 ℃ and 600 ℃ are selected to carry out the uniaxial compression and creep experiments. It is found that temperature has obvious influence on uniaxial compressive strength, Young’s modulus and failure modes of red sandstone. It can be also believed that the temperature can degrade the mechanical behaviors of red sandstone. However, as the temperature increases, the damage value does not always increase, the damage variable has a negative value at 25 ~ 400 ℃. It can be found that the higher the temperature, the larger the effect of loading ratio on the ratio of creep strain to instantaneous strain, that is, temperature reduces the ability of red sandstone to resist creep deformation. Acoustic emission (AE) technology has been also used in the loading process of uniaxial compression and creep tests. It is found that the probability density of AE absolute energy of different samples still satisfies the Gutenberg-Richter law well. In uniaxial compression test, as the temperature increases, the absolute value of the critical index increases exponentially. In the uniaxial creep test, as the loading ratio increases, the absolute value of the critical index first decreases and then increases. When the loading ratio is the same, as the temperature increases, the absolute value of the critical index also increases exponentially.

A mechanistic based cavitation model that considers nucleation, early-stage growth and sintering under creep-fatigue interaction is proposed. The number density of cavities ρ and their evolution during multi-cycle creep-fatigue loading are predicted. Both the cavity nucleation and early-stage growth rates, controlled by grain boundary (GB) sliding mechanism during the tension phase, are formulised as a function of local normal stress σ n. The cavity sintering that occurs during the compression phase is described as a function of σ n, but the mechanism switches to the unconstrained GB diffusion. By examining various load waveform parameters, results provide important insights into experimental design of studying the creep-dominated cavitation process under creep-fatigue interaction. First, creep-fatigue test with initial compression will promote higher ρ value compared to that with initial tension, if the unbalanced stress hold time in favour of tension is satisfied. Second, the ρ value does not have a monotonic dependence on either the compressive hold time or stress level, because of their competing effect on nucleation and sintering. Third, the optimum value of stress variation rate exists in terms of obtaining the highest ρ value due to sintering effect.

Using thin-walled cone-pipe welded joints of stainless steels, fatigue tests under bending loads were carried out. The test data were statistically analyzed with the Benard’s approximation, Gaussian, 2P-Weibull, and 3P-Weibull distributions. Stress–life curves at different failure probabilities by a constant strength scatter band model were obtained. The metallographic structures were investigated, and the stress concentration states were analyzed to elucidate the causes of the strengths and scatters. In the high-cycle fatigue regime, the 3P-Weibull distribution was mostly in agreement with the Benard’s approximation, and the coefficient of determination was 0.9642. The microstructure of the weld metal with a high weld opening angle was mainly ferrite phase with 20% austenite distribution. The crack initiation point was close to the weld interface, but the propagation direction was at a right angle, and initially penetrated the heat affected zone of the cone, leading to the high fatigue strength. The stress concentration factors depended on the weld opening angles, indicating the main factor which affected strengths and scatters.

Complex civil structures require the cooperation of many building materials. However, it is difficult to accurately monitor and evaluate the inner damage states of various material systems. Based on a convolutional neural network (CNN) and the acoustic emission (AE) time-frequency diagram, we used the transfer learning method for classifying the AE signals of different materials under external loads. The results show the CNN model can accurately classify cracks that come from different materials based on AE signals. The recognition accuracy can reach 90% just by re-training the full connection layer of the pre-trained model, and its accuracy can reach 97% after re-training the top 2 convolutional layers of this model. A realization of cracking source identification mainly depends on the differences in mineral particles in materials. This work highlights the great potential for real-time and quantitative monitoring of the health status of composite civil structures.

This work is an extension of applying a previously developed fracture mechanics cracking damage model to predict the fatigue lifetime of un-notched round specimens made of a ferrite-pearlite 0.4C-70/30 carbon steel in some cases of variable amplitude loading VAL. The model simulates the collective behavior of growing short fatigue cracks originating from the specimen surface roughness. Material grains of different phases, sizes and strengths are randomly distributed over the minimum circumference. Possible activities of surface cracks are predicted against loading cycles. Relevant published experimental data were utilized for comparison. The present predictions are in agreement with the corresponding experimental results.

In this paper a new interpretation and modification of the SWT function in terms of the total damaging energy density is proposed and discussed. The total damaging energy density is the sum of the damaging part of the strain energy density and complementary energy density corresponding to the first quadrant in damaging σD-εD axes. For cyclic loading with positive mean stress (σm≥0) the proposed function reduces to the original SWT formulation. For cyclic loading with negative mean stress (σm<0) the maximum stress is augmented by 1/3 of absolute value of the mean stress. The proposed approach shows a consistent correlation of the mean stress effects for both positive and negative mean stresses.