Search

[Display omitted] • A fatigue life prediction approach considering crack growth environment is developed. • A crack length criterion for environmental transition is determined. • Marker load method retrieves crack growth data from the fracture surfaces. • Small crack growth rate in pseudo-vacuum is estimated by inverse method. Porosity defects are the main cause of the fatigue failures of directed energy deposited Ti-6Al-4V. Porosity defect-induced cracks initiate in pseudo-vacuum and switch to air upon reaching the specimen surface. In view of this failure characteristic, this study establishes a fatigue life prediction approach based on the fracture mechanics framework considering crack growth environment. Firstly, high cycle fatigue tests at room temperature under atmospheric conditions and fracture surface observation were carried out to obtain the fatigue lives, key defect parameters and crack initiation types. Then, long and small crack growth tests in air were conducted, and crack growth rate curve in air was fitted by the Hartman-Schijve equation variant. Next, the crack growth data in the fracture surfaces were retrieved by the marker load method, which were compared with the crack growth data in vacuum in the existing literature to determine the long crack growth data in pseudo-vacuum. Small crack growth data in pseudo-vacuum were determined by inverse method. And thus, the crack growth rate curve in pseudo-vacuum was established. Finally, a crack length criterion for environmental transition was proposed to segmentally predict the fatigue lives which were compared with the experimental lives to verify the effectiveness of the approach. [ABSTRACT FROM AUTHOR]

A wide range K I solution has been developed for an eccentrically cracked middle tension specimen with clamped ends. In the derivation of the K I solution, an equivalent model was developed; weight functions for eccentric cracks in a finite-width strip were utilized, with approximate closed form K I and plate-end displacement solutions obtained. Good agreements between the approximate solutions and corresponding FEM results were observed over a wide range of eccentricities and crack lengths. The accuracy of the K I solution was analyzed under different aspect ratio conditions, with the applicability of the approximate solutions given. The equivalent model was also discussed. [ABSTRACT FROM AUTHOR]

• A new approach is proposed to diagnose and predict fatigue damage based on strain monitoring. • Fatigue experiments for 2024 aluminum plates with center holes are conducted to verify the method. • The fatigue damage diagnosis and prognosis results agree well with the experimental results. A fatigue damage diagnosis and prognosis approach has been developed based on strain monitoring. The fatigue experiment is conducted on 2024 aluminum specimens with center holes to obtain the strain/crack information. The occurrence of cracks is detected by a thresholding method. A data-driven Gaussian process regression (GPR) algorithm is used to establish the relationship between the crack length and the strain characteristic parameters. A dynamic Bayesian network (DBN) is assembled by combining the GPR algorithm and Paris formula for real-time crack propagation predictions. An experiment has been carried out to illustrate the accuracy of the approach. The results indicate that the threshold method can detect fatigue cracks effectively. The GPR algorithm can accurately identify fatigue crack length and has higher recognition accuracy than other algorithms. The crack propagation prediction error and time consumption of DBN are much smaller than mechanics-based crack growth analysis. [ABSTRACT FROM AUTHOR]

A probabilistic approach for predicting the economic life of an aircraft fleet is proposed with variation in load spectrum and structural property taken into account. Specimens of TA15M titanium alloy were fatigue tested under three individual load spectra of different damage severities. By using the fatigue test results, a generic equivalent-initial-flaw-size distribution was obtained, and a stochastic crack growth model was developed including the load spectrum variation and the material crack resistant variation. With the number of cracks exceeding the economic repair limit as the economic life criteria, a simple expression was derived for the probability of crack exceedance. [ABSTRACT FROM AUTHOR]

Highlights: [•] Variability of load spectrum of a fleet is analyzed. [•] Fatigue tests are conducted to test the variability of load spectrum. [•] Load spectrum damage follows a lognormal distribution. [ABSTRACT FROM AUTHOR]

Abstract: A nonsymmetric fatigue crack growth (FCG) test is specially designed and performed to study whether FCG test data from nonsymmetric cracking which is unavoidable in practice can be used to determine valid FCG rate information for fatigue life evaluation. Two types of alloys were used in this test and the FCG data were obtained via visual measurements. The effects of boundary conditions and central hole were examined using FEA software and the results showed that within engineering allowable error margin, the theoretical solution of stress intensity factors (SIFs) under the remote uniform tension is valid for nonsymmetric crack in this test. As a result, a method for analyzing the FCG rate of nonsymmetric through-thickness crack is proposed. It is found that the FCG rate parameters in the stable stage have no difference between two crack tips and the crack growth data can be used together to determine the Paris parameters. The proposed method, if supported by further research, can be adopted to validate FCG test data involving nonsymmetric cracking in standard tests. [Copyright &y& Elsevier]

An increment learning scheme based on fully-connected neural network is proposed to predict the fatigue–crack growth in middle tension, M(T), specimens of 7B04 T6 aluminum and TA15 titanium alloy under constant amplitude stress. Usually, the measurement of fatigue–crack growth rates is labor-intensive and time-consuming, and the dataset of fatigue–crack growth is small. Neural networks with back-propagation algorithm are not good at training on small dataset. Here we design network inputs which employ multiple increment information to overcome this shortage. Given the first part data points of crack growth in a specimen, the trained network can predict the rest for both aluminum alloy and titanium alloy without any prior knowledge. The trained network learns the underlying rules in experimental data of crack growth. Our method shows superiority to conventional fitting formulas and common neural networks such as recurrent neural network and long short-term memory method. Our work demonstrates the capacity of neural network and provides an alternative method to predict fatigue–crack growth. • An NN-based increment learning scheme is developed to predict fatigue–crack growth. • The trained network can give satisfactory predictions without any prior knowledge. • NN-based increment learning scheme can work good with small amount of data. [ABSTRACT FROM AUTHOR]

• Fatigue crack growth behavior was illustrated for corner cracks at holes. • A crack-tip constraint independent fatigue crack growth rate was proposed. • A fatigue crack growth equation was proposed for corner cracks at holes. The fatigue crack growth (FCG) behavior of a corner crack at the edge of a hole has not been well understood. To address this issue, we first conducted fatigue tests under constant amplitude loading. The results show that the FCG rate along the thickness direction varies nonlinearly with the stress intensity factor range (Δ K) in a log-log scale. This is because the crack-tip stress relaxation at the depth point gets increasingly pronounced as the crack propagates. Finally, we proposed a crack-tip constraint independent FCG rate and the corresponding FCG equation to correlate the crack-tip constraint independent FCG rate with Δ K. [ABSTRACT FROM AUTHOR]