Article
Materials Science, Composites
C. Hemanth Kumar, R. P. Swamy
Summary: This study evaluated the fatigue life of glass fiber reinforced epoxy composites under combined rotating bending loads, showing a decrease in fatigue life with increasing stress levels. An artificial neural network approach was utilized to demonstrate the impact of stress levels on fatigue strength and failure modes, with high correlation between experimental data and ANN outputs. Three failure modes were observed during low and high cycle fatigue tests, successfully classified by ANN with 100% accuracy using a conjugate gradient backpropagation algorithm.
COMPOSITES COMMUNICATIONS
(2021)
Article
Engineering, Chemical
Ali Bakhshizade, Ahmad Ghasemi-Ghalebahman, Mohammad Ali Hajimousa
Summary: This paper discusses the fatigue life analysis of a blend of natural rubber and styrene-butadiene rubber with and without nanoclay particles using finite element analysis. The study finds that damage parameters based on strain can effectively estimate the fatigue lives, and the predicted fatigue life matches well with the measured fatigue results. Additionally, the effect of various parameters and the fracture surface of the nanocomposite are investigated.
POLYMER ENGINEERING AND SCIENCE
(2023)
Article
Engineering, Mechanical
Hao Gong, Zeng-gui Jin, Feng-peng Yang, Wen-tao Mao
Summary: To investigate the effects of the stop-hole on fatigue crack path and life under mixed I-II cyclic loads, a series of tests were conducted using compact tension-shear specimens with different stop-hole diameters and various loading angles. The results showed that fatigue life increased with loading angle and hole diameter. Numerical simulation was used to enhance the experimental database, and a variable-length recurrent neural network was proposed based on the database to remember crack segment information. The results were encouraging and provided inspiration for evaluating component safety.
THEORETICAL AND APPLIED FRACTURE MECHANICS
(2023)
Article
Engineering, Mechanical
Yanju Wang, Zhenyu Zhu, Aixue Sha, Wenfeng Hao
Summary: This paper proposes a novel approach for estimating the low cycle fatigue (LCF) life of titanium alloy structural parts based on the continuous damage mechanics (CDM) model. The genetic algorithm-optimized back-propagation artificial neural network (GABP-ANN) accurately predicts the LCF life of titanium alloy structural parts.
INTERNATIONAL JOURNAL OF FATIGUE
(2023)
Article
Materials Science, Composites
Hui Qian, Jincheng Zheng, Yusheng Wang, Dong Jiang
Summary: Ceramic matrix composites are widely used in the aerospace field due to their excellent mechanical properties. However, analyzing their fatigue life is challenging due to the complex microstructure and failure mechanism. To address this issue, a fatigue life analysis method based on Artificial Neural Network (ANN) is proposed, which utilizes material parameters and loading parameters to predict fatigue life. The study compares different neural networks and finds that Elman Network (ENN) and Convolutional Neural Network (CNN) provide high-precision predictions using simulation data sets. The Generalized Regression Neural Network (GRNN) fails to meet the requirements. Experimental data from literature are used to train ENN and CNN, and good predictions are achieved using only 4 S-N curves as the training set.
APPLIED COMPOSITE MATERIALS
(2023)
Article
Engineering, Mechanical
Xiangnan Liu, Xuezhi Zhao, Wen-Bin Shangguan
Summary: Uniaxial tension fatigue tests are conducted on vulcanized natural rubber specimens to establish a BPNN model for estimating fatigue life. An improved sine-cosine algorithm (ISCA) is proposed to optimize model parameters, showing better prediction accuracy and efficiency compared to other algorithms.
FATIGUE & FRACTURE OF ENGINEERING MATERIALS & STRUCTURES
(2022)
Article
Engineering, Mechanical
Xiangnan Liu, Wen-Bin Shangguan, Xuezhi Zhao
Summary: In this study, the residual fatigue life of natural rubber components under variable amplitude loads is predicted using a support vector machine (SVM) model and a nonlinear fatigue cumulative damage (NFCD) model. The experimental results demonstrate that the prediction accuracy of these models is higher than that of the traditional Miner criterion.
INTERNATIONAL JOURNAL OF FATIGUE
(2022)
Article
Chemistry, Multidisciplinary
Jinna Shi, Wenxiu Zhang, Yanru Zhao
Summary: In order to improve the prediction accuracy of the machine learning model for concrete fatigue life using small datasets, a GRW-DBA data augmentation method is proposed. The method significantly improves the prediction accuracy of the ANN model, reaching 98.6% R-2 index, and is verified in different data distributions. A graphical user interface is created for easy application in engineering.
APPLIED SCIENCES-BASEL
(2023)
Article
Nanoscience & Nanotechnology
Zhiwei Li, Dong An, Rizheng He, Zhijian Sun, Jiaxiong Li, Zhiyi Zhang, Yaqing Liu, Chingping Wong
Summary: Different crosslink networks were used to prepare carbon black/graphene oxide/natural rubber composites (CB/GO/NR) through the latex co-precipitation approach. The influence of different types of crosslinks on crack propagation resistance and fatigue life in vulcanized systems was investigated. The study demonstrated that the crosslink network and polysulfide-based crosslink structure in the conventional vulcanization (CV) system were the key factors to improve the crack propagation resistance, leading to improved tear strength and lower crack growth rate.
ADVANCED COMPOSITES AND HYBRID MATERIALS
(2023)
Article
Engineering, Mechanical
Zheng Pan, Yaling Lai, Yanping Wang, Wuwei Duan, Yu Qiao, Yuanwen Liu, Chengkai Song
Summary: The experimental study on rubber specimens revealed that the multiaxial fatigue life increases with decreasing strain amplitude. The loading path significantly influences the fatigue life of vulcanized natural rubber. The SWT model and CXH model provide accurate predictions for both materials.
FATIGUE & FRACTURE OF ENGINEERING MATERIALS & STRUCTURES
(2021)
Article
Engineering, Mechanical
Kun Zhou, Xingyue Sun, Shouwen Shi, Kai Song, Xu Chen
Summary: A machine learning method integrating ANN and PLS algorithm was proposed to identify genetic features by optimizing fatigue life prediction. Successful development of a predicting model and identification of the significance of early fatigue data for predicting fatigue life were demonstrated in this study.
FATIGUE & FRACTURE OF ENGINEERING MATERIALS & STRUCTURES
(2021)
Article
Materials Science, Characterization & Testing
Ganfang Luo, Chi Zhang, Yan Huan, Yuan Yin, Hongguo Sun, Zhaobin Chen, Xiaoniu Yang
Summary: The influence of fatigue on the adhesion of cord/rubber composites was investigated using H-pull tests. Fatigue was found to enhance the diffusion of sulfur from rubber to RFL and lead to a homogenization of modulus. Higher frequencies resulted in a decrease in fatigue life due to self-heating, but this effect was negligible at low displacement amplitudes. Longer fatigue life was achieved at higher R ratios, with an exceptional increase observed at the highest R ratio. A universal CLD model was developed to predict fatigue life, and it was found that there exists a critical R ratio beyond which H-pull tests no longer accurately evaluate dynamic adhesion.
Article
Mechanics
GaoYuan He, YongXiang Zhao, ChuLiang Yan
Summary: This paper proposes a physics-informed neural network framework to improve the accuracy of multiaxial fatigue life prediction.
ENGINEERING FRACTURE MECHANICS
(2023)
Article
Materials Science, Composites
Ali Bakhshizade, Ahmad Ghasemi-Ghalebahman, Mohammad Ali Hajimousa
Summary: In this study, multiple regression models were established to assess the fatigue life of nanoclay nanocomposites. The strain amplitude was found to have the most significant effect on fatigue life, while the nanoclay loading and test frequency had negligible effects. The interactions between factors also did not have a significant impact on fatigue life.
POLYMER COMPOSITES
(2023)
Article
Engineering, Mechanical
Su Liu, Wenjing Shi, Zhixin Zhan, Weiping Hu, Qingchun Meng
Summary: A novel method combining the error trained back propagation artificial neural network (BP-ANN) technique with the continuum damage mechanics (CDM) model is proposed for predicting the high cycle fatigue (HCF) life of aluminum alloys. Experimental data and numerically computed fatigue lives are used to train the ANN model, and the predicted errors are used to adjust the numerical results for final fatigue life prediction. The proposed technique shows better accuracy and stability compared to other methods.
INTERNATIONAL JOURNAL OF FATIGUE
(2022)