Fatigue Life Prediction Method Research Articles

Probabilistic LCF life prediction framework for turbine discs considering random load history

AbstractThe load history exerts a considerable impact on the low cycle fatigue (LCF) life of turbine discs. Thus, oversimplifying the load history leads to substantial errors in fatigue life prediction. This study introduces a probabilistic fatigue life prediction method for turbine discs, accounting for the randomness inherent in LCF load history. The method involves quantifying the randomness of load history through numerical simulation and employing a surrogate model enhanced with learning functions to balance computational efficiency and accuracy. The probabilistic LCF life prediction of full‐scale turbine disc was conducted, demonstrating that the fatigue life scatter predicted by the proposed method more closely aligns with experimental data compared to the original approach. By refining the numerical simulation process, the proposed method better accounts for uncertainties in load history while maintaining computational efficiency, offering significant insights for the fatigue reliability design of turbine discs.

Master S-N Curve Fatigue Life Prediction of a Railroad Car Bogie Based on Mesh Insensitive Structural Stress Method

Fatigue life prediction of a welded structure is a complex phenomenon due to the nature of fatigue and the welding process. Additionally, Finite Element Method (FEM) results are extremely sensitive to the size of elements. Therefore, it is difficult to adopt a method to estimate the fatigue life, especially for welded structures. Besides, mesh size independence is a critical issue to perform fatigue life prediction methods that eliminates the need for excessive element numbers in the mesh. This paper investigates the Master S-N Curve Approach (MCA) using the output parameters of the mesh insensitive Structural Stress Method (SSM). MCA based on SSM employs structural stresses recovered from nodal forces and nodal moments. To recover these inputs, FEM model should be established properly. Thus, boundary conditions and applied loads were prepared for the model according to the BS EN 13749:2021. The submodeling technique in ANSYS software was used to analyze the bogie structure. To justify the mesh independence for the model, different mesh sizes were tested. In a specific range for shell bodies, SSM was shown to provide sufficient mesh independence feature. Furthermore, MCA was compared with Hot Spot Stress Method and Nominal Stress Method based on their fatigue life estimations.

Probabilistic gear fatigue life prediction based on physics-informed transformer

Gear fatigue testing is a complex process that is both time-consuming and labor-intensive, and often produces dispersed results. Traditional statistical analysis methods for probability-stress-life (P-S-N) and fatigue limit require large sample sizes. This study proposes a novel probabilistic gear fatigue life prediction method based on physics-informed Transformer. This method seeks to overcome the limitations of traditional statistical analysis methods and data-driven models by integrating stress-life physical knowledge with the data-driven model. Specifically, based on the obtained gear fatigue limit, this method predicts the probabilistic fatigue life at each stress level by using only a set of probability distribution characteristics of fatigue life under constant amplitude load. The proposed method was validated through multiple cases of gear bending fatigue tests with varying materials, treatments, and geometries, and prediction errors were within an error factor of 2. Compared with other traditional machine learning methods, the accuracy of life prediction was improved by 19.74% to 29.44%. Additionally, feature analysis was conducted to identify key parameters influencing the accuracy of probabilistic fatigue life prediction and provide a range of optimal parameter selection, which is expected to significantly promote the efficiency of fatigue basic data construction in the gear industry.

Fatigue life prediction of composite bolted joints based on finite element model and machine learning

AbstractThis study proposes a fatigue life prediction method for composite bolted joints, which combines algorithm optimization‐based hybrid neural networks with finite element modeling. First, based on the Hashin failure criterion of physical mechanism, a finite element model for fatigue life prediction of composite bolted joints is established, and the simulation calculations have been conducted using various initial conditions. Then, by integrating the simulation and experiment data, we have established a fatigue life database that serves machine learning training and prediction. Finally, the data undergo a comprehensive process of deep feature extraction through the utilization of a convolutional neural network (CNN). The resulting deep features are utilized as inputs for training the backpropagation neural network (BPNN) to predict fatigue life. The results indicate this synergistic combination of CNN and BPNN results in a substantial improvement in prediction accuracy and has remarkable superiority in predicting the fatigue life of composite bolted joints.

Fatigue performance analysis of precast segmental assembled concrete beams

In recent years, assembled bridges have become widely utilized in bridge construction, raising concerns about durability-related bridge diseases over time. These issues significantly impact the fatigue life of assembled bridges, necessitating an in-depth exploration of their fatigue performance. While existing research primarily concentrates on the transverse connection of multiple longitudinal beams, there is a notable dearth of studies on longitudinal precast segmental assembled bridges. This paper addresses this gap by establishing a fatigue benchmark finite element model for segmental assembled concrete beams, building upon existing experiments. The study employs numerical simulation to analyze the entire fatigue process, examining stress distribution, damage development, and considering the influence of reinforcement corrosion. Furthermore, a fatigue life prediction method, based on fatigue residual strength (R), is proposed for predicting the fatigue life (N) of concrete in precast segmental assembled beams. Results reveal that prestressed and ordinary reinforcements experience increasing stress with loading cycles, peaking around 100,000 cycles. Throughout fatigue loading, compressive stress in concrete remains low, preventing fatigue compression failure. However, tensile stress near joints gradually rises, initiating cracks at the mid-span beam’s bottom. With continued cyclic loading, these cracks propagate towards the loading point. The upper and lower limits of fatigue life predicted by the fatigue life prediction method closely align with the compressive fatigue test values of concrete, proposed fatigue life prediction method is efficient and accurate.

Multi-Stage Hybrid Loading Fatigue Life Prediction Study

Abstract Fatigue is a damage accumulation process in which material properties are continuously deteriorated and degraded under cyclic loading. Damage accumulation analysis plays a key role in preventing the occurrence of fatigue failure, and the damage evolution mechanism is an important parameter of interest for fatigue behavior. The fatigue cumulative damage life prediction model is established by focusing on the damage evolution mechanism, and an improved Manson–Harford cumulative damage nonlinear model is developed to calculate fatigue damage life by considering the effect of load ratio of multi-stage loading on fatigue life. A fatigue life prediction method considering high-low cycles and multi-stage mixed loads is proposed. The fatigue life prediction accuracy of the model is verified by fatigue tests on structural samples. The accuracy of the fatigue life prediction error is the highest among similar models under the same test, the model proposed in this paper.

An improved J-integral based adhesive joint fatigue life estimation method for automotive structural durability analysis

ABSTRACT To estimate the fatigue life of lightweight automotive structures in compliance with top-down design principles, there is a need for an accurate adhesive joint fatigue life prediction method that is compatible with industrial finite element modeling practices. This paper proposes a method to split the analytical J-integral solution based on the fracture mode of joints. The split mode I and II J-integrals correspond to the opening mode and sliding mode of the bonded joint, respectively. By splitting the J-integral and introducing the concept of the mixed mode ratio, an improved approach for estimating adhesive joint fatigue life is presented, considering the influence of loading modes. The proposed fatigue life prediction method is validated through fatigue tests on a sub-component level bonded structure. The results of the validation demonstrate that the split J-integral analytical solution method, considering the mixed mode ratio, provides better predictions compared to its predecessor approach.

High-cycle and low-cycle fatigue life prediction under random multiaxial loadings without cycle counting

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.

Pore-induced fatigue failure: A prior progressive fatigue life prediction framework of laser-directed energy deposition Ti-6Al-4V based on machine learning

Pores are major cause of fatigue failure in laser-directed energy deposition (L-DED) titanium alloy. For the safe application of L-DED titanium alloys, it is essential to establish a fatigue life prediction method based on pore-induced fatigue. This paper proposes a prior progressive fatigue life prediction framework based on ridge classification and kernel ridge regression algorithms. The fatigue life prediction was carried out on L-DED Ti-6Al-4V alloy in three steps: critical pore identification, fine granular area existence prediction and final fatigue life prediction. The fatigue life prediction method adopted in the current study outperform the others with a correlation coefficient as high as 0.951, followed by a comparison with the results derived from different machine learning algorithms. The results show that the proposed fatigue life prediction framework can predict the fatigue life of L-DED Ti-6Al-4V alloy based on computed tomography tests and microstructure features. Due to its strong generalization ability and effectiveness, the proposed prediction method is expected to be valuable for fatigue-resistant design of L-DED Ti-6Al-4V alloy.

Damage-Accumulation-Induced Crack Propagation and Fatigue Life Analysis of a Porous LY12 Aluminum Alloy Plate.

Rivets are usually used to connect the skin of an aircraft with joints such as frames and stringers, so the skin of the connection part is a porous structure. During the service of the aircraft, cracks appear in some difficult-to-detect parts of the skin porous structure, which causes great difficulties in the service life prediction and health monitoring of the aircraft. In this paper, a secondary development subroutine in PYTHON based on ABAQUS-XFEM is compiled to analyze the cracks that are difficult to monitor in the porous structure of aircraft skin joints. The program can automatically analyze the stress intensity factor of the crack tip with different lengths in the porous structure, and then the residual fatigue life can be deduced. For the sake of safety, the program adopts a more conservative algorithm. In comparison with the physical fatigue test results, the fatigue life of the simulation results is 16% smaller. This project provides a feasible simulation method for fatigue life prediction of porous structures. It lays a foundation for the subsequent establishment of digital twins for damage monitoring of aircraft porous structures.

Fatigue life prediction considering individual modulus for unidirectional needled C/SiC composites under variable amplitude loading

Needled C/SiC composites are widely used in the aerospace field, making the evaluation of fatigue damage critically important. A fatigue damage model associated with the individual modulus is proposed. Compared to the fatigue damage model by S-N curve, the proposed fatigue damage model can better characterize the fatigue damage behavior of UN-C/SiC composites. By combining the rainflow counting method, the Goodman equation, and the Palmgren-Miner damage accumulation rule, a fatigue life prediction method considering the individual modulus is developed for UN-C/SiC composites under variable amplitude loading. Compared with the experimental life, the predicted fatigue life shows a good result.

Fatigue life prediction method of Q235 steel single lap thin-walled bolted connections under shear loading

Fatigue life prediction method of Q235 steel single lap thin-walled bolted connections under shear loading

A generalization ability-enhanced image recognition based multiaxial fatigue life prediction method for complex loading conditions

A generalization ability-enhanced image recognition based multiaxial low-cycle fatigue life prediction is proposed for complex conditions, including multiaxial variable amplitude fatigue, fatigue-creep interaction, and anisotropic materials. Notably, a novel modified dynamic hysteresis image plotting method is introduced for multiaxial variable amplitude loading. Building on a pre-trained convolution neural network (CNN) model, fine-tuning is used across three cases. It makes the error decrease to 25%. Besides, fine-tuning improves training efficiency and generalization. Based on fine-tuning results, each layer of the model is attributed with physics significance. Furthermore, regression activation map (RAM) is used for interpretation, illustrating that the model focuses on the inner and edges of hysteresis images.

Thermal fatigue life prediction method for die casting mold steel based on the cooling cycle

Thermal fatigue life prediction method for die casting mold steel based on the cooling cycle

Notch effect identification and fatigue life prediction of undercut weld

Undercut, as a typical surface imperfection at the weld toe, significantly impacts the weld fatigue life. Neither the ISO5817 standard for imperfection level definition nor the traditional fatigue life prediction methods for welds can accurately identify the undercut effect on weld fatigue life. To solve this problem, a new stress concept and life prediction method are proposed, which can predict the fatigue life of undercut welds in both linear elastic and elastic–plastic conditions. Its prediction accuracy is verified by test data and engineering examples.

Predicting asphalt mixture fatigue life via four-point bending tests based on viscoelastic continuum damage mechanics

A more technically accurate approach to predict fatigue life of asphalt mixtures can enhance the precision of performance evaluation for asphalt pavements. This study applied the viscoelastic continuum damage mechanics (VECD) to investigate the fatigue behavior of asphalt mixtures. The fatigue life of typical asphalt mixtures was evaluated through four-point bending tests under various conditions. The strain-modulus empirical equation was firstly employed to fit the experimental data. However, it was noted that this equation was not entirely suitable for all conditions, as factors like temperature and loading frequency can influence fatigue life not only through the dynamic modulus but also via the damage properties of asphalt mixtures. To address this limitation and achieve a comprehensive fatigue life prediction equation, the Time-Strain Superstition Principle (TSSP) and Frequency-Strain Superstition Principle (FSSP) were introduced to characterize the impact of temperature and loading frequency on fatigue life, which can reduce fitting errors. Additionally, a mechanical prediction equation was also derived by the VECD approach, and its applicability under different conditions was explored. The results revealed that the damage characteristic curve (C-S curve) remained unaffected by loading frequency and strain level but was highly dependent on temperature. Significant fitting errors were observed in different temperatures when employing the VECD method for fatigue life prediction, as the VECD method did not account for the influence of temperature on the viscoelastic properties of asphalt mixtures, which in fact plays a critical role in fatigue life. In response, temperature adjustment coefficients were introduced into the derived equation. The predicted results demonstrated that the newly modified method significantly improved prediction precision, yielding an error of less than 20%.

Tensile efficiency and fatigue life of similar and dissimilar carbon steel joints subjected to rotary friction welding

Rotary friction welding (RFW) is a solid-state welding method that can address the melting point limitation of other welding types; however, the welded area remains at risk of failure due to fatigue during operation. Fatigue failure is difficult to detect and predict, making it a critical failure mode in engineering. Although many fatigue life prediction methods are available, they do not consider the impact of RFW parameters (such as pressure, speed, and friction time) on fatigue strength. Owing to their good mechanical properties and commercial viability, carbon steel of grades C35 and C45 is commonly applied to manufacture main parts in the automotive industry. This study centers on employing RFW to join C35 and C45, representing the first-ever instance of using this technique to weld them together. The primary goal of RFW for C35 and C45 is to construct optimum parts that meet specific design requirements and have a long fatigue life. Findings showed C45–C45 joints (77 %) were more tensile-efficient than C35–C35 joints (73 %). Dissimilar C35–C45 joints reached 60 % of C45 base metal and 76 % of C35 base metal in terms of tensile efficiency. RFW welds had lower tensile strength than their base metals. Optimal results came at 35 MPa friction pressure and 8–12 s friction time for both similar and dissimilar RFW joints using the Coffin–Manson method, dissimilar C33–C45 joint fatigue life improved by over 85 %, aligning C35 and C45 engineering aspects. SEM microstructure analysis showed two regions: non-deformed (NDZ) and plastically deformed (PDZ), with broader PDZ indicating better tensile strength. These findings enhance RFW's efficiency, promoting automotive part longevity and safety.

Multiaxial fatigue life prediction of notched specimens based on multidimensional grey Markov theory

It is always a challenge to accurately predict the fatigue life of notched specimens under multiaxial loading, which is a guarantee of structural integrity. Therefore, in this paper, multiple factors affecting fatigue life are considered and a multidimensional grey Markov model is used for fatigue life prediction. Firstly, the plane of maximum shear strain amplitude is determined as the critical plane. Meanwhile, the stress–strain components on the critical plane are extracted, and the relevant parameters affecting the fatigue life are calculated. Secondly, due to the defects of the grey theory itself, the prediction error is larger. Therefore, the grey model is modified with Markov theory to get the grey Markov model. The correlation degree analysis between relevant factors and fatigue life is carried out with grey theory to analyze the degree of influence of different factors on fatigue life. Finally, the grey Markov model is analyzed and verified using experimental data from Q355(D), GH4169 and Al7050 notched specimens and the results show that the grey Markov model can provide an economical and effective method for multiaxial fatigue life prediction.

A novel machine learning method for multiaxial fatigue life prediction: Improved adaptive neuro-fuzzy inference system

In this study, a neuro-fuzzy-based machine learning method is developed to predict the multiaxial fatigue life of various metallic materials. First, the fuzzy inference system and neural network are combined to identify and capture the nonlinear mapping relationship between multiaxial fatigue damage parameters and fatigue life. Non-proportionality and phase differences are introduced to characterize different loading paths. Next, the Adam algorithm is employed to update the premise parameters of the original model to achieve fast and accurate convergence. Then, subtractive clustering is applied to extract fuzzy rules between input variables and output for more efficient prediction. Moreover, the hyperparameters in the proposed model are automatically optimized by the adaptive opposition slime mould algorithm to obtain the optimal model. The predictive performance of the proposed model is verified by fatigue experimental data for six materials in published literature, which indicates that the proposed model can effectively predict the fatigue life of various materials under different loading paths. Meanwhile, compared with six classical machine learning models, it is found that the proposed model has better predictive performance and extrapolation capability.

BPNN Based Multi-factor Synthesis Prediction Method for Coal Rock Fatigue Life

BPNN Based Multi-factor Synthesis Prediction Method for Coal Rock Fatigue Life