{"title":"Machine Learning Models to Predict the Static Failure of Double-Lap Shear Bolted Connections","authors":"H. Almuhanna, G. Torelli, L. Susmel","doi":"10.1111/ffe.70019","DOIUrl":"https://doi.org/10.1111/ffe.70019","url":null,"abstract":"<p>This study investigates the potential of machine learning models to predict the failure load and mode of double-lap shear bolted connections. Five algorithms were evaluated: adaptive boosting, artificial neural network, decision trees, support vector machines with radial basis function kernel, and k-nearest neighbors. A dataset comprising 221 experimental and numerical tests with varying input parameters, including different grades of stainless and carbon steel, was used to train the models. Unlike previous studies, the inclusion of diverse materials enabled the development of more generalizable models. To address data limitations, reduce biases associated with data split, and mitigate overfitting, k-fold cross-validation was adopted instead of the conventional 80/20 split. Results show that both regression and classification models achieved high coefficients of determination across most algorithms. Adaptive boosting delivered the most accurate failure load predictions, while artificial neural network achieved the highest accuracy in classifying failure modes. The findings highlight the potential of well-trained machine learning models to outperform traditional codified methods in accurately predicting the structural response of bolted connections, especially when trained on diverse datasets.</p>","PeriodicalId":12298,"journal":{"name":"Fatigue & Fracture of Engineering Materials & Structures","volume":"48 9","pages":"4041-4055"},"PeriodicalIF":3.2,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/ffe.70019","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144774159","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Hui Gao, Shida Wu, Donglin Wang, Li Zuo, Zhongwei Zhao
{"title":"A Comprehensive Investigation on Influence Factors of Fracture Toughness Testing Methods in Metallic Materials for Accuracy Improvements","authors":"Hui Gao, Shida Wu, Donglin Wang, Li Zuo, Zhongwei Zhao","doi":"10.1111/ffe.70016","DOIUrl":"https://doi.org/10.1111/ffe.70016","url":null,"abstract":"<div>\u0000 \u0000 <p>The accuracy of fracture toughness (FT) obtained using experimental methods can be influenced by factors, such as material properties and specimen geometries, and the extent of inaccuracy depends on the synthetic impact of these factors. Thus, the impact rule of these factors and corresponding underlying reasons require investigation. However, studies in this research field are significantly insufficient and underdeveloped, leaving risks for engineering structures. This study aims to provide a comprehensive investigation of the commonly used FT testing methods in ductile metallic materials for identifying factors that influence the accuracy of obtained FT and then to reveal underlying reasons for the inaccuracy and offer useful recommendations to assist scholars in conducting FT experiments accurately. The literature review and analysis methodology are performed, and the research scope mainly focuses on the three most commonly used testing methods. The results show that some factors can significantly influence the accuracy of tested FT, which may cause safety and cost problems for engineering structures. This study reveals the underlying reasons for FT inaccuracy, summarizes the influence regularity of factors, gives error ranges of each factor, and provides feasible recommendations to improve experimental accuracy, which fills gaps in this research field to significantly reduce the safety and cost risks of practical engineering structures.</p>\u0000 </div>","PeriodicalId":12298,"journal":{"name":"Fatigue & Fracture of Engineering Materials & Structures","volume":"48 9","pages":"4017-4040"},"PeriodicalIF":3.2,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144774158","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Teng Ma, Guoxi Jing, Xiuxiu Sun, Guang Chen, Yafei Fu, Tian Ma
{"title":"Low-Cycle and Thermomechanical Fatigue Life Prediction Method for Compacted Graphite Iron Based on Small-Sample Physics-Informed Neural Networks","authors":"Teng Ma, Guoxi Jing, Xiuxiu Sun, Guang Chen, Yafei Fu, Tian Ma","doi":"10.1111/ffe.70002","DOIUrl":"https://doi.org/10.1111/ffe.70002","url":null,"abstract":"<div>\u0000 \u0000 <p>A physics-informed neural network (PINN) model based on deep learning has been proposed for predicting low-cycle fatigue (LCF) and thermomechanical fatigue (TMF) life. By analyzing the LCF and TMF data of compacted graphite iron (CGI), characteristic parameters were identified that can simultaneously represent both types of fatigue, achieving a unification of the parameters for the two fatigue life models. The incorporation of fatigue life physical information as a constraint in the loss function of the deep neural network enabled accurate predictions of LCF and TMF for CGI under small-sample conditions. Comparative analysis results indicated that the deep learning–based PINN model outperformed traditional machine learning models in terms of prediction accuracy. Additionally, comparisons with traditional LCF and TMF life prediction models showed that the deep learning–based PINN model achieves high prediction accuracy while possessing generalization and extrapolation capabilities unattainable by traditional models. These results demonstrate that the PINN model exhibits high accuracy and versatility.</p>\u0000 </div>","PeriodicalId":12298,"journal":{"name":"Fatigue & Fracture of Engineering Materials & Structures","volume":"48 9","pages":"3999-4016"},"PeriodicalIF":3.2,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144774157","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yiqun Hu, Zilong Zhang, Yuhang Zhang, Lei Yuan, Re Xia
{"title":"Torsional Mechanical Behavior of TPMS Porous Structures: Experimental Insights on Diamond, Gyroid, and Schwarz Primitive Designs","authors":"Yiqun Hu, Zilong Zhang, Yuhang Zhang, Lei Yuan, Re Xia","doi":"10.1111/ffe.70018","DOIUrl":"https://doi.org/10.1111/ffe.70018","url":null,"abstract":"<div>\u0000 \u0000 <p>Triply periodic minimal surface (TPMS) porous structures exhibit significant potential for industrial applications owing to their high surface area, lightweight, and tunable mechanical properties. This work explores the torsional behavior of three typical TPMS porous structures, including diamond, gyroid, and Schwarz primitive types. Monotonic torsion tests show that the torsional resistance increases with relative density. Under symmetric cyclic torsional loading, the normalized torque amplitude decreases with increasing cycles, whereas failure cycles decrease and dissipated energy increases with higher strain amplitudes. Once the strain exceeds the elastic threshold, the torque amplitude initially drops sharply and then gradually declines until fracture occurs. In asymmetric cyclic torsion tests, the fracture morphology under asymmetric torsion remains consistent with that observed under monotonic torsion. These findings contribute to a deeper understanding of the mechanical reliability of TPMS porous structures and offer guidance for their structural optimization in engineering applications.</p>\u0000 </div>","PeriodicalId":12298,"journal":{"name":"Fatigue & Fracture of Engineering Materials & Structures","volume":"48 9","pages":"4056-4070"},"PeriodicalIF":3.2,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144774160","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
K. Sanjeeviprakash, A. Rajesh Kannan, N. Siva Shanmugam
{"title":"Fatigue Life Assessment of Wire arc Additive Manufactured Duplex Stainless Steel From ER2209 Filler Wire","authors":"K. Sanjeeviprakash, A. Rajesh Kannan, N. Siva Shanmugam","doi":"10.1111/ffe.70014","DOIUrl":"https://doi.org/10.1111/ffe.70014","url":null,"abstract":"<div>\u0000 \u0000 <p>Duplex stainless steels (DSS) are known for their excellent strength and corrosion resistance in cyclic loading applications such as power plants and structural components, etc. This study examines the fatigue life of wire arc additive manufactured (WAAM) ER2209 DSS. The as-fabricated WAAM wall consists of fine and equiaxed dendrites with predominantly austenitic phase due to complex thermal cycles and filler wire chemistry. Tensile tests reveal anisotropy between build and deposit directions, with notable differences in yield strength, tensile strength, and ductility. Fatigue behavior along the build direction resembles austenitic stainless steel SS316, enduring up to 1.5 million cycles without failure below 40% of yield stress (188.7 MPa), but shows significant differences compared to wrought DSS2205. Post-fatigue EBSD analysis identifies crack initiation from persistent slip bands, propagating along austenite-ferrite grain boundaries. Fracture morphologies show smooth, widely spaced striations at low-stress amplitudes, transitioning to rough, closely spaced striations at higher stress amplitudes.</p>\u0000 </div>","PeriodicalId":12298,"journal":{"name":"Fatigue & Fracture of Engineering Materials & Structures","volume":"48 9","pages":"3967-3979"},"PeriodicalIF":3.2,"publicationDate":"2025-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144774020","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Li Qian, Enlong Liu, Ru Zhang, Jianhai Zhang, Tianzhi Yao, Gaofeng Ma, Xi Lu
{"title":"Fracture Evolution and Mesoscopic Damage in Granite: A Macro–Mesoscopic Constitutive Model Under True Triaxial Stress","authors":"Li Qian, Enlong Liu, Ru Zhang, Jianhai Zhang, Tianzhi Yao, Gaofeng Ma, Xi Lu","doi":"10.1111/ffe.70015","DOIUrl":"https://doi.org/10.1111/ffe.70015","url":null,"abstract":"<div>\u0000 \u0000 <p>This study develops a macro–mesoscopic constitutive model to investigate fracture evolution in granite under true triaxial stress. Granite is modeled as a binary composite of bonded and frictional media, with microcrack propagation driving transitions between phases. Mesoscopic flaw evolution is tracked using CT scanning, enabling quantification of flaw volume and distribution during loading. A homogenization-based framework links mesoscale damage to nonlinear macroscopic behavior, explicitly capturing the influence of intermediate principal stress. A novel parameter, the flaw closure degree, is introduced to characterize compaction, providing a direct indicator of internal damage. The model accurately reproduces stress–strain responses and failure modes observed in tests, overcoming the limitations of phenomenological approaches. By incorporating fracture mechanics and internal flaw dynamics, this work offers a predictive tool for assessing structural integrity in brittle rock systems subjected to complex stress paths.</p>\u0000 </div>","PeriodicalId":12298,"journal":{"name":"Fatigue & Fracture of Engineering Materials & Structures","volume":"48 9","pages":"3980-3998"},"PeriodicalIF":3.2,"publicationDate":"2025-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144774021","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Max Ahlqvist, Kenneth Weddfelt, Daniel Leidermark, Viktor Norman
{"title":"Very High Cycle Fatigue Performance of Ductile Cast Iron With Different Microstructures","authors":"Max Ahlqvist, Kenneth Weddfelt, Daniel Leidermark, Viktor Norman","doi":"10.1111/ffe.70011","DOIUrl":"https://doi.org/10.1111/ffe.70011","url":null,"abstract":"<p>The very high cycle fatigue performance of four ductile cast irons, a solid solution strengthened ferritic, ferritic-pearlitic, and two austempered ductile cast irons, was investigated by ultrasonic fatigue testing under fully reversed loading conditions. The step-stress fatigue testing method incorporating re-use of runouts was employed, with a \u0000<span></span><math>\u0000 <mn>3</mn>\u0000 <mo>×</mo>\u0000 <mn>1</mn>\u0000 <msup>\u0000 <mrow>\u0000 <mn>0</mn>\u0000 </mrow>\u0000 <mrow>\u0000 <mn>8</mn>\u0000 </mrow>\u0000 </msup></math> cycles to failure runout criteria. The guarantee of failed specimens enables extensive fractography and characterization of fatigue-initiating defects, facilitating extreme value analysis of defect distributions. Furthermore, the fatigue and defect data were used for evaluation of stress–cycle–defect relationships as well as fatigue strength distributions. The obtained relationships between applied stress and fatigue-initiating defect size were used to adjust the data with respect to a given defect size, to compare and evaluate fatigue-initiating defect size equivalent fatigue strength distributions, emphasizing the role of the different matrix microstructures.</p>","PeriodicalId":12298,"journal":{"name":"Fatigue & Fracture of Engineering Materials & Structures","volume":"48 9","pages":"3950-3966"},"PeriodicalIF":3.2,"publicationDate":"2025-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/ffe.70011","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144774007","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Zhongkai Ren, Lixin Liu, Haoran Li, Wei Xu, Peng Chen, Tao Wang
{"title":"A Novel Multiaxial High-Cycle Fatigue Life Prediction Model Based on Critical Plane-Intrinsic Damage Dissipation","authors":"Zhongkai Ren, Lixin Liu, Haoran Li, Wei Xu, Peng Chen, Tao Wang","doi":"10.1111/ffe.70003","DOIUrl":"https://doi.org/10.1111/ffe.70003","url":null,"abstract":"<div>\u0000 \u0000 <p>The critical plane approach identifies the crack initiation plane and propagation directions, whereas intrinsic damage dissipation quantifies energy dissipation directly correlated with fatigue damage. This study proposes a multiaxial high-cycle fatigue (HCF) failure criterion and a life prediction model by combining both theories and explicitly incorporating mean stress effects. The proposed criterion employs four critical plane parameters: maximum shear stress, shear stress amplitude, maximum normal stress, and normal stress amplitude. This methodology demonstrates clear physical significance. Validation against experimental datasets demonstrates close agreement between model predictions and empirical results in both fatigue limit determination and life estimation. Comparative evaluations against classic and recent criteria reveal statistically superior predictive accuracy of the proposed criterion. The method shows its promising potential as a tool for multiaxial HCF engineering analysis.</p>\u0000 </div>","PeriodicalId":12298,"journal":{"name":"Fatigue & Fracture of Engineering Materials & Structures","volume":"48 9","pages":"3936-3949"},"PeriodicalIF":3.2,"publicationDate":"2025-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144773990","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rujun Li, Wei Zhang, Davide Salvatore Paolino, Andrea Tridello, Filippo Berto, Yan Peng
{"title":"CALC\u0000 −\u0000 \u0000 ΔK\u0000 th\u0000 : Automatic Assessment of the Stress Intensity Factor Threshold From VHCF Fracture Surfaces With Optically Dark Area","authors":"Rujun Li, Wei Zhang, Davide Salvatore Paolino, Andrea Tridello, Filippo Berto, Yan Peng","doi":"10.1111/ffe.14701","DOIUrl":"https://doi.org/10.1111/ffe.14701","url":null,"abstract":"<div>\u0000 \u0000 <p>\u0000 <span></span><math>\u0000 <mtext>CALC</mtext>\u0000 <mo>−</mo>\u0000 <msub>\u0000 <mi>ΔK</mi>\u0000 <mi>th</mi>\u0000 </msub></math> is a assessing the stress intensity factor (SIF) threshold of materials from specimens failed in the very-high-cycle fatigue (VHCF) region and showing the typical fish-eye morphology. The developed process involves two key parameters: the size of the optically dark area (ODA), which is a typical feature region on the fracture surface of specimens failing in the VHCF life region from internal defects, and the stress amplitude at the crack initiation site. Based on the optical image of the fracture surface input from the user side, the automatic detection of the ODA feature is obtained with a deep learning method. Thereafter, the analytical stress distribution in the specimen is assessed, thus allowing to compute the critical SIF threshold for the investigated specimen. This package requires minimal scanning prerequisites on specimens, featuring notable advantages in modularity, automation, and practical usability.</p>\u0000 </div>","PeriodicalId":12298,"journal":{"name":"Fatigue & Fracture of Engineering Materials & Structures","volume":"48 9","pages":"4071-4083"},"PeriodicalIF":3.2,"publicationDate":"2025-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144774034","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}