Search

Neural networks (NNs) are increasingly used in design to construct the objective functions and constraints, which leads to the needs of optimization of NN models with respect to design variables. A Neural Optimization Machine (NOM) is proposed for constrained single/multi-objective optimization by appropriately designing the NN architecture, activation function and loss function. The NN's built-in backpropagation algorithm conducts the optimization and is seamlessly integrated with the additive manufacturing (AM) process-property model. The NOM is tested using several numerical optimization problems. It is shown that the increase in the dimension of design variables does not increase the computational cost significantly. Next, a brief review of the physics-guided machine learning model for fatigue performance prediction of AM components is given. Finally, the NOM is applied to design processing parameters in AM to optimize the mechanical fatigue properties through the physics-guided NN under uncertainties. One novel contribution of the proposed methodology is that the constrained process optimization is integrated with physics/knowledge and the data-driven AM process-property model. Thus, a physics-compatible process design can be achieved. Another significant benefit is that the training and optimization are achieved in a unified NN model, and no separate process optimization is needed. This article is part of the theme issue 'Physics-informed machine learning and its structural integrity applications (Part 1)'. [ABSTRACT FROM AUTHOR]

Neural network (NN) models have significantly impacted fatigue‐related engineering communities and are expected to increase rapidly 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. [ABSTRACT FROM AUTHOR]

A critical plane-based mixed-mode delamination growth model for composite materials under fatigue loadings is proposed in this study. A brief review for mixed-mode delamination is given and the special focus in the study is on a critical plane-based fracture criterion since it can automatically adapt for different local failure modes. An equivalent energy release rate range suitable for fatigue analysis is proposed for the delamination growth rate prediction under general proportional and non-proportional multiaxial loadings. The proposed methodology is validated with extensive experimental data available in the open literature. A general good agreement is observed between model predictions and experimental observations. Finally, some discussion and conclusions are drawn based on the proposed model. [ABSTRACT FROM AUTHOR]

An experimental testing procedure is proposed for generating creep-dominant creep-fatigue interaction in Alloy 617 at 950 °C. Most experimental studies on Alloy 617 at elevated temperatures employ a purely strain-controlled loading waveform with a tensile hold period. Strain-controlled loading can only generate interaction in the fatigue-dominant regime of the damage interaction diagram, due to the saturation effect of increasing hold time on cycle life. This saturation is caused by the rapid stress relaxation of Alloy 617 at high temperatures. Design codes for components undergoing creep-fatigue interaction at high temperatures require creep-fatigue testing data in the creep-dominant regime. This study investigates multiple creep-fatigue loading waveforms with the objective of producing creep-dominant interaction. The custom loading waveforms are implemented on a servo-hydraulic load frame with a furnace. A hybrid-control scheme with strain-controlled ramps and load-controlled hold periods is proposed due to its flexibility in generating varying proportions of creep and fatigue damage. Experimental data is presented and different loading schemes are compared. Special emphasis is on the creep and fatigue damage contributions by the newly proposed testing method. Several conclusions are discussed and an extension of this current work is suggested to improve the damage interaction diagram for design purposes. [ABSTRACT FROM AUTHOR]

In this paper, a real-time composite fatigue life prognosis framework is proposed. The proposed methodology combines Bayesian inference, piezoelectric sensor measurements, and a mechanical stiffness degradation model for in-situ fatigue life prediction. First, the composites stiffness degradation is introduced to account for the composites fatigue damage accumulation under cyclic loadings and a new growth rate-based stiffness degradation model is developed. Following this, the general Bayesian updating-based fatigue life prediction method is discussed. Several sources of uncertainties and the developed stiffness degradation model are included in the prognosis framework. Next, an in-situ composites fatigue testing with piezoelectric sensors is designed and performed to collected sensor signal and the global stiffness data. Signal processing techniques are implemented to extract damage diagnosis features. The detected stiffness degradation is integrated in the Bayesian inference framework for the remaining useful life (RUL) prediction. Prognosis performance on experimental data is validated using prognostics metric. Finally, some conclusions and future work are drawn based on the proposed study. [ABSTRACT FROM AUTHOR]

A general framework for probabilistic prognosis and uncertainty management under fatigue cyclic loading is proposed in this article. First, the general idea using the Bayesian updating in prognosis is introduced. Several sources of uncertainties are discussed and included in the Bayesian updating framework. An equivalent stress level model is discussed for the mechanism-based fatigue crack growth analysis, which serves as the deterministic model for the lap joint fatigue life prognosis. Next, an in situ lap joint fatigue test with pre-installed piezoelectric sensors is designed and performed to collect experimental data. Signal processing techniques are used to extract damage features for crack length estimation. Following this, the proposed methodology is demonstrated using the experimental data under both constant and variable amplitude loadings. Finally, detailed discussion on validation metrics of the proposed prognosis algorithm is given. Several conclusions and future work are drawn based on the proposed study. [ABSTRACT FROM PUBLISHER]

Abstract: A novel in situ scanning electron microscope (SEM) fatigue testing is proposed in this paper to investigate the fatigue crack growth mechanisms within one cyclic loading under plane stress conditions. The objectives of the experimental study are to verify the hypotheses of a small time scale fatigue crack growth model and to develop a new experimental methodology for the detailed mechanism investigation of fatigue crack growth. During the testing, one loading cycle is uniformly divided into a certain number of steps. At each step, high resolution images are taken around the crack tip region. Imaging analysis is used to quantify the crack growth kinetics and crack tip deformation behavior at any time instant in a loading cycle. Crack closure phenomenon is directly observed during the crack growth process in the current investigation. It is also observed that crack growth is not uniformly distributed within a loading cycle and only happens during a small portion of the loading path. Multiple mechanisms exist within one cyclic loading, which is not able to be captured using the classical cycle-based approaches. A detailed discussion is given based on the current investigation to explain some well-known issues in the classical fatigue theory. [Copyright &y& Elsevier]

Abstract: An experimental study is proposed to investigate the effect of random loading sequence effect on the fatigue crack growth behavior of Al 7075-T6. The testing matrix includes different overload cycle percentage, overload ratios, and deterministic and random loading sequences in the current investigation. Multiple specimen tests and statistical data analysis are performed to show the effect of random loading sequence on the median and scatter behavior of fatigue crack growth. The proposed experimental study suggests that extreme value distribution is a good approximation of fatigue life distribution. It is observed that the effect of uncertain loading is different under different loading spectrums. For high overload cycle percentage spectrums, the random loading sequence has no major impact on the probabilistic crack growth behavior compared to the deterministic loading sequence with identical load cycle distributions. For low overload cycle percentage spectrums, the random loading sequence has huge impact on the probabilistic crack growth behavior compared to the deterministic loading sequence with identical load cycle distributions, for both the median and the scatter of the fatigue crack length curves. Finally, all experimental observations are reported in table format in for future numerical model development and validation for interested readers. [Copyright &y& Elsevier]

Abstract: Mechanical components have stress risers, such as notchs, corners, welding toes and holes. These geometries cause stress concentrations in the component and reduce the fatigue strength and life of the structure. Fatigue crack usually initiates at and propagates from these locations. Traditional fatigue analysis of notched specimens is done using an empirical formula and a fitted fatigue notch factor, which is experimentally expensive and lacks physical meaning. A general methodology for fatigue limit prediction of notched specimens is proposed in this paper. First, an asymptotic interpolation method is proposed to estimate the stress intensity factor (SIF) for cracks at the notch root. Both edge notched and center notched components with finite dimension correction are included into the proposed method. The small crack correction is included in the proposed asymptotic solution using El Haddad’s fictitious crack length. Fatigue limit of the notched specimen is estimated using the proposed stress intensity factor solution when the realistic crack length is approaching zero. A wide range of experimental data are collected and used to validate the proposed methodology. The relationship between the proposed methodology and the traditionally used fatigue notch factor approach is discussed. [Copyright &y& Elsevier]

Abstract: A new methodology is proposed in this paper to calculate the equivalent initial flaw size (EIFS) distribution. The proposed methodology is based on the Kitagawa–Takahashi diagram. Unlike the commonly used back-extrapolation method for EIFS calculation, the proposed methodology is independent of applied load level and only uses fatigue limit and fatigue crack threshold stress intensity factor. The advantage of the proposed EIFS concept is that it is very efficient in calculating the statistics of EIFS. The developed EIFS methodology is combined with probabilistic crack growth analysis to predict the fatigue life of smooth specimens. Model predictions are compared with experimental observations for various metallic materials. [Copyright &y& Elsevier]

Abstract: A general methodology for stochastic fatigue life prediction under variable amplitude loading is proposed in this paper. The methodology combines a nonlinear fatigue damage accumulation rule and a stochastic S–N curve representation technique to achieve this objective. The nonlinear damage accumulation rule proposed in this paper improves the deficiencies inherent in the linear damage accumulation rule and still maintains its simplicity in the application. The covariance structure of the stochastic damage accumulation process under variable amplitude loading is included into the methodology by a new stochastic S–N curve approach. A wide range of fatigue data available in the literature is used to validate the proposed methodology, which covers different type of metallic materials under constant and variable amplitude loadings. The prediction results are also compared with existing fatigue models. [Copyright &y& Elsevier]

A new damage accumulation model for fatigue life prediction of composite laminates is proposed in this study. The model is constructed on the ply level and uses a new multiaxial damage index to consider the damage caused by different stress components. The fatigue life is predicted based on the proposed model and experimental results of the unidirectional laminates. The numerical results are compared with the experimental data for balanced laminates. The predicted fatigue lives agree with the experimental observations very well. The methodology for probabilistic fatigue life prediction and reliability calculation is also presented. [Copyright &y& Elsevier]

A concurrent simulation and experimental validation for curvilinear fatigue crack growth (FCG) analysis under both constant amplitude and overload spectrums is proposed in this paper. The simulation methodology is based on a small time-scale fatigue crack growth model and the extended finite element method (XFEM) to calculate the stress intensity factor solution of an arbitrary curvilinear crack. Parametric studies are used to determine the algorithm parameters in the numerical fatigue crack growth simulation. Following this, experimental testing on modified compact specimens is performed under both constant amplitude and overload loadings for model validation. Experimentally measured crack growth orientations and lengths are compared with numerical simulations. Both the experimental and simulation results show the overload retardation behavior for curvilinear cracks under overload loadings. The investigated periodic overload loading has no significant impact on the crack growth orientations. Several conclusions and areas of future work are identified based on the proposed numerical and experimental investigations. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents a method for fatigue damage propagation model selection, updating, and averaging using reversible jump Markov chain Monte Carlo simulations. Uncertainties from model choice, model parameter, and measurement are explicitly included using probabilistic modeling. Response measurement data are used to perform Bayesian updating to reduce the uncertainty of fatigue damage prognostics. All the variables of interest, including the Bayes factors for model selection, the posterior distributions of model parameters, and the averaged results of system responses are obtained by one reversible jump Markov chain Monte Carlo simulation. The overall procedure is demonstrated by a numerical example and a practical fatigue problem involving two fatigue crack growth models. Experimental data are used to validate the performance of the method. [Copyright &y& Elsevier]

• A time-derivative fatigue crack growth formulation and no need for cycle counting in the presence of arbitrary loadings. • Unified methodology for high-cycle fatigue, low-cycle fatigue, and their interactions. • Demonstrated and validated with uniaxial and multiaxial random loadings. • Fatigue life prediction using crack growth analysis with explicit near-threshold modeling. A novel fatigue life prediction method under random multiaxial loadings is proposed in this paper. One unique benefit of the proposed method is that it is based on the time-derivative fatigue crack growth formulation and does not need cycle counting under arbitrary loadings. First, a brief review of subcycle fatigue crack growth (FCG) analysis and the equivalent initial flaw size (EIFS) concept for life prediction is introduced. Next, the existing subcycle fatigue crack growth model is extended to near-threshold conditions. An intrinsic fatigue threshold of the material is introduced, and a hypothesis is proposed that the crack only grows when the local stress intensity factor is beyond the intrinsic threshold. Following this, the analogy of stress intensity factor and strain intensity factor is used to extend the fatigue life prediction model to strain-based fatigue analysis. Correction for large plastic deformation is included to handle low-cycle fatigue life prediction. The novelties of this model are that it considers the FCG at the threshold and near-threshold region and it is able to predict both HCF and LCF conditions using FCG-based life prediction. Following this, extensive in-house and literature data for AL-7075-T6 under uniaxial and multiaxial, constant, and random loadings are used for model validation. Discussions for the effect of model parameters are provided based on parametric analysis. Conclusions and future work are mentioned. Code and data are released in public cloud service for interested readers. [ABSTRACT FROM AUTHOR]

An experimental study is proposed to investigate the high-cycle fatigue behavior of prestressed concrete box-girders. Special focus is on the shear diagonal cracks in the webs of girders, which is rarely investigated in the open literature. Laboratory fatigue tests were designed and conducted to study the propagation of shear diagonal crack of PC box-girder specimens under cyclic loadings. The development of stirrup strain, propagation characteristics of diagonal crack and the related factors are discussed in detail. A finite element (FE) model for concrete damage of PC box girder under high-cycle fatigue loading is established to investigate the crack propagation behavior. Following this, a model for the calculation of diagonal crack widths of PC box-girder under fatigue loading is proposed based on the experimental observations. The calculation results are compared with experimental data for validation. It is observed that the maximum stirrup strain in crack zone increases 67.9% based on this investigation. The ratio of displacement during the last cycle and that during initial loading can research 40.3% in the current investigation. Finally, several conclusions are drawn based on the proposed study. [ABSTRACT FROM AUTHOR]

Abstract: A new subcycle-based delamination growth model is proposed. The key idea is to model delamination growth at any time instant within a cyclic loading, rather than the cycle-averaged growth kinetics. First, some existing models are briefly reviewed for the fatigue delamination growth analysis of composite materials. Following this, two hypotheses are given for the derivation of the subcycle delamination growth model: (1) delamination growth does not happen during the unloading path and (2) delamination growth does not happen when the applied loading is below a reference level during the loading path. Mathematical expression is given for the calculation of delamination growth rate. Next, the extension of the proposed model is discussed to use both stress intensity factor and energy release rate as the driving force parameters, which are widely used in the open literature. Finally, the model predictions are compared with extensive experimental data under different stress ratios for model validation. One of the advantages of the proposed model is that the stress ratio dependent delamination growth can be predicted. Some conclusions and future work are drawn based on the proposed method. [Copyright &y& Elsevier]

• A time-based crack growth method for fatigue damage prognosis of steel bridges is proposed. • The proposed method can perform time-based fatigue crack growth analysis without cycle counting. • Full-scale coupled vehicle-bridge interaction framework is adopted for numerical demonstration. • Case study on the fatigue damage prognosis of the I-10 Twin Span Bridge in Louisiana is presented. Traditionally, the fatigue damage assessment of steel bridges is conducted using cycle-based methods. Cycle-based methods require stress time history be transformed to cycle history first before fatigue analysis can be performed. However, well-defined cycle history does not exist for the traffic-induced stress time history of bridges as it contains numerous large and small cycles with random stress ranges embedded together. To address this inherent difficulty, a time-based fatigue crack growth (FCG) model is proposed for fatigue damage prognosis of steel bridges under traffic loading. The time-based FCG method describes the crack growth in the time domain through a crack growth kinetics function derived based on the physical mechanism of fatigue crack propagation. To demonstrate the proposed method, numerical simulations are first performed using full-scale coupled vehicle-bridge interaction framework to generate stress time histories of a typical steel bridge under traffic loading. Then, FCG analysis is performed using the time-based method and the effects of road surface roughness and vehicle weight on the FCG are studied. Finally, a case study on the I-10 Twin Span Bridge is presented to demonstrate the time-based FCG method for fatigue damage prognosis using in situ monitoring data. The results show that the proposed method can be used to directly predict the FCG of steel bridges in the time domain using stress time histories obtained either from stress analysis or in situ monitoring. [ABSTRACT FROM AUTHOR]

This paper reports an experimental and theoretical investigation on the fatigue performance enhancement of damaged steel structures by bonded carbon fiber reinforced polymer (CFRP) systems considering the temperature effects. First, thermomechanical properties of the epoxy adhesive and CFRPs were examined via dynamic mechanical analysis (DMA) and tensile tests. The appropriate curing procedure for the structural adhesive was obtained. Next, the fatigue behavior of notched steel plates strengthened by bonded CFRPs was investigated at different temperatures. During the fatigue tests, beach marking and back-face strain techniques were used to monitor the fatigue crack growth on steel components and the damage evolution within the bonding interface. The results show that the high-temperature resistance of the epoxy adhesive can be effectively improved by increasing the curing temperature and duration. The properly cured bonded CFRP patches can effectively enhance the fatigue performance of damaged steel structures. However, and the elevated temperatures significantly degrade the fatigue behavior of the CFRP-strengthened steel components. Unlabelled Image • Curing epoxy adhesive at elevated temperatures for a specific duration enhances its high-temperature resistance. • Properly cured bonded CFRP patches effectively enhances the fatigue performance of damaged steel structures. • Elevated temperatures can cause failure mode transform of bonding interface between CFRP and steel. • Elevated temperatures significantly degrade the fatigue behavior of the CFRP-strengthened steel components. [ABSTRACT FROM AUTHOR]