Article
Mathematics, Interdisciplinary Applications
Serkan Balli
Summary: The Covid-19 pandemic is the most important health disaster the world has faced in the past eight months, predicting its trend has become a challenge. A study analyzed COVID-19 data and proposed a time series prediction model, estimating the global pandemic will peak at the end of January 2021 with approximately 80 million people cumulatively infected.
CHAOS SOLITONS & FRACTALS
(2021)
Article
Computer Science, Interdisciplinary Applications
Yongtao Bai, Deborah C. Nardi, Xuhong Zhou, Ricardo A. Picon, Julio Florez-Lopez
Summary: Fatigue resistance is crucial for the life-cycle sustainability of materials and structures, yet predicting the fatigue resistance of structural members subjected to flexural forces remains a challenge. Developing a general lumped damage simulation model can offer a new perspective for quantifying fatigue-induced remaining life for engineering structures across elastic and plastic amplitudes.
COMPUTERS & STRUCTURES
(2021)
Article
Thermodynamics
Congzhi Huang, Mengyuan Yang
Summary: Photovoltaic power is stochastic, intermittent, and volatile, posing challenges to the safe and stable operation of the power grid. To improve the accuracy of PV power forecasting, a MLSTNet model is proposed, using temporal and spatial feature extraction to achieve higher accuracy in ultra-short-term forecasting.
Article
Engineering, Multidisciplinary
Mehmet Hamarat, Sakdirat Kaewunruen
Summary: This paper proposes a novel approach based on the physical meaning of the bond network to evaluate damage in Peridynamic simulations and identify crack initiation and propagation. The commonly used nodal damage value is questioned for neglecting the relevance of broken bonds to cracks. The proposed method is shown to be robust, effective, and provide a single outcome for Peridynamic simulations.
APPLIED MATHEMATICAL MODELLING
(2023)
Article
Engineering, Civil
Liyun Yang, Fei Zhang, Changyu Lin, Huanzhen Xie, Ping Fan
Summary: This study investigates the failure characteristics of the rock-fiber concrete composite structure under compressive load through experimental research. The deformation characteristics, damage characteristics, and failure modes of the composite layer are analyzed. The test results show that the binding site of the composite can play a certain effect of crack arrest.
Article
Engineering, Marine
Ke Zhan, Chuanqing Li, Renchuan Zhu
Summary: This paper introduces a novel lightweight machine learning forecasting architecture called FreMixer, which leverages frequency domain information. Through noise reduction and frequency domain feature extraction, FreMixer demonstrates superior performance in experiments, with low deployment costs and high training efficiency.
Article
Economics
Irina Grossman, Tom Wilson, Jeromey Temple
Summary: Local and state governments rely on small area population forecasts for decision-making, but current methods are often inaccurate. Recent developments in machine learning for time series forecasting offer potential for improvement, but little research has been done in the field of demography. This paper describes the development of two machine learning methods for small area population forecasts and evaluates their performance.
SOCIO-ECONOMIC PLANNING SCIENCES
(2023)
Article
Mechanics
Haifeng He, Huaiju Liu, Andrea Mura, Caichao Zhu
Summary: In this study, a comprehensive finite-element-based method is proposed to predict the total bending fatigue life of gears, including the initiation and propagation of cracks, using the continuum damage mechanics theory and the linear elastic fracture mechanics theory. The developed gear finite element model is validated according to the ISO standard. The results indicate that the gear bending fatigue life, especially the initiation life of cracks, is highly influenced by the loading level, with the initiation life dominating the total life.
Article
Mechanics
Wenyuan Li, Youmin Rong, Yu Huang, Long Chen, Zhihui Yang, Guojun Zhang
Summary: This paper investigates the dynamic and static mechanical behavior of short pulse laser cutting CFRP and explores the effect of cutting damage on mechanical properties. The study finds that cutting damage weakens the tensile and bending strength of CFRP plate by 11.5% and 6.2% respectively. Furthermore, large cutting damage aggravates tensile fatigue crack propagation and fiber protrusion. The research results provide important insights for damage suppression and improvement of material mechanical properties.
ENGINEERING FRACTURE MECHANICS
(2023)
Article
Mechanics
M. Ciavarella
Summary: This article discusses the applicability of energy balance for large cracks and introduces the influence of liquid materials and viscoelasticity on flaw sensitivity. A comparison is made between different theories using a simple Dugdale model, revealing significant differences.
ENGINEERING FRACTURE MECHANICS
(2022)
Article
Engineering, Civil
Cyril Furtlehner, Jean-Marc Lasgouttes, Alessandro Attanasi, Marco Pezzulla, Guido Gentile
Summary: The probabilistic forecasting method described in this study is designed to leverage the spatial and temporal dependency of urban traffic networks, to provide accurate short-term predictions and meaningful forecasts up to several hours in advance. The method can handle missing data, predict the entire network's state in one pass, and has an execution time that scales linearly with the network size. By carefully analyzing the model, it is possible to distinguish modeling bias from the intrinsic noise of traffic phenomena and its measurement process.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Review
Engineering, Multidisciplinary
Intisar Omar, Muhammad Khan, Andrew Starr
Summary: This paper comprehensively reviews the influence of damage-sensitive features on machine learning algorithms in structural health monitoring and assessment (SHMA). It emphasizes the importance of selecting appropriate features and algorithms based on raw data and structure material type for accurate crack prediction.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2022)
Article
Multidisciplinary Sciences
Moiz Qureshi, Nawaz Ahmad, Saif Ullah, Ahmed Raza ul Mustafa
Summary: Forecasting is a popular topic in various disciplines due to the uncertainty of underlying phenomena, which can be estimated using mathematical functions. As technology advances, algorithms are updated to capture the ongoing phenomena. Machine learning algorithms, such as MLP, ELM, ARIMA, and ES models, are utilized to model and predict the real exchange rate data set. The study split the data into training and testing, and the model that best meets the KPI criteria is selected for predicting the behavior of the real exchange rate data set.
Article
Engineering, Mechanical
Jie Li, Bin Zhang, Dong Lyu, Jingbo Guo, Kang Su, Bo Hu
Summary: Based on a fatigue load spectrum, this paper establishes a model to calculate the fatigue propagation life of the cutterhead with different reliability, and analyzes the main factors affecting its reliability. The results show the dangerous failure positions and emphasize the importance of load stress amplitude and initial crack size in determining the crack propagation life and reliability of the cutterhead. The research provides a scientific basis for crack detection, life prediction, and reliability evaluation of the cutterhead.
ENGINEERING FAILURE ANALYSIS
(2022)
Article
Engineering, Civil
Yani Lian, Jungang Luo, Wei Xue, Ganggang Zuo, Shangyao Zhang
Summary: Reasonable runoff forecasting is crucial for water resource management. This study proposes a cause-driven runoff forecasting framework using linear-correlated reconstruction and machine learning model. Four experiments were conducted to validate the accuracy and efficiency of this framework. The results show the importance of including streamflow, ERA5L, and meteorology data as inputs, the superiority of linear-correlated feature reconstruction, and the effectiveness of the LSTM model.
WATER RESOURCES MANAGEMENT
(2022)
Article
Computer Science, Interdisciplinary Applications
Pham Toan Thang, Dieu T. T. Do, Jaehong Lee, T. Nguyen-Thoi
Summary: This paper presents an in-depth study on the influence of nanoscale parameters on the bending and free vibration responses of functionally graded carbon nanotube-reinforced composite nanoshells. Mathematical formulas and numerical calculations are used to investigate the effect of nanoscale parameters, material properties, and shell shapes on the deflection and fundamental frequency parameters of the nanoshells.
ENGINEERING WITH COMPUTERS
(2023)
Article
Mechanics
Seunghye Lee, Minhee Seo, Sun-Myung Kim, Thuc P. Vo, Jaehong Lee
Summary: This study presents a parametric study for a two-way beam string structure to provide optimal design parameters. Numerical examples validate the effects of different parameters on the structure, including the number of struts, deflections and stresses, as well as cable curvatures. The study also investigates the stress sharing ratio of the beam's top/bottom and the sagging cable under positive and negative pressure.
MECHANICS BASED DESIGN OF STRUCTURES AND MACHINES
(2023)
Article
Engineering, Marine
Bao-Loi Dang, Hung Nguyen-Xuan, Magd Abdel Wahab
Summary: In this paper, a systematic and time-efficient approach is proposed to calibrate 2D VARANS-VOF models for simulating wave interaction with a porous plate. A data-driven approach combined with numerical and experimental data is developed to identify the optimal empirical coefficients associated with drag force coefficients. Advanced gradient boosting decision trees algorithms are used to accurately predict the model parameters. The developed model is validated using available experimental data, showing a high level of agreement.
Article
Engineering, Civil
Van-Thien Tran, Trung-Kien Nguyen, H. Nguyen-Xuan, Magd Abdel Wahab
Summary: This paper proposes an algorithm for vibration and buckling optimization of functionally graded porous microplates and investigates the effects of material distribution, length scale, porosity density, and boundary conditions on their characteristics.
THIN-WALLED STRUCTURES
(2023)
Article
Engineering, Aerospace
Nam V. Nguyen, Kim Q. Tran, P. Phung-Van, Jaehong Lee, H. Nguyen-Xuan
Summary: In this study, an efficient numerical framework is proposed to explore the responses of functionally graded triply periodic minimal surface (FG-TPMS) microplates. The static bending, free vibration, and buckling characteristics of these structures are thoroughly presented for the first time. The study utilizes refined plate theory and isogeometric analysis to study these mechanical responses. It also takes into account the size effect with the modified couple stress theory. The findings contribute to the development and application of TPMS geometry in microscale structures.
AEROSPACE SCIENCE AND TECHNOLOGY
(2023)
Article
Engineering, Multidisciplinary
Yun Lu Tee, H. Nguyen-Xuan, Phuong Tran
Summary: This paper investigates the bending behavior of porcupine quills and bioinspired Voronoi sandwich panels, aiming to understand the influence of geometrical design on their bending performance. X-ray micro-computed tomography is used to examine the internal morphology of the quill, revealing a functionally graded structure. Inspired by this, Voronoi sandwich panels are designed with Voronoi seed distribution and gradient transition configurations. Experimental results and simulations show that the graded panels exhibit better bending performance than the uniform panels. This study provides valuable insights for engineering applications in aerospace and automobile industries.
BIOINSPIRATION & BIOMIMETICS
(2023)
Article
Mechanics
Hau T. Mai, Seunghye Lee, Donghyun Kim, Jaewook Lee, Joowon Kang, Jaehong Lee
Summary: This paper proposes a robust deep neural network-based parameterization framework to directly solve the optimum design for geometrically nonlinear trusses subject to displacement constraints. The integration of DNN into Bayesian optimization allows for finding the best optimum structural weight, which is further optimized through hyperparameter optimization. The experimental results demonstrate that this approach can overcome the drawbacks of machine learning applications in computational mechanics.
EUROPEAN JOURNAL OF MECHANICS A-SOLIDS
(2023)
Article
Mathematics, Applied
Thang Le-Duc, H. Nguyen-Xuan, Jaehong Lee
Summary: In this study, a novel deep learning model called Finite-element-informed neural network (FEI-NN) is proposed for parametric simulation of static problems in structural mechanics. The approach uses supervised training to consider the parametric variables of structures as input features and implicitly embeds the spatial variables into the loss function through a soft constraint called finite element analysis (FEA) loss. The training process minimizes the empirical risk function while partially respecting the mechanical behaviors through the FEA loss, which is defined as a residual calculated from the weak form of the surrogate system scaled from the actual structure. Additionally, a technique based on batch matrix multiplication is introduced to reduce the time complexity for estimating the FEA loss. The method is demonstrated to outperform traditional data-driven approaches in terms of faster convergence and better DNN models for both generalization and extrapolation performance through several experiments.
FINITE ELEMENTS IN ANALYSIS AND DESIGN
(2023)
Article
Computer Science, Artificial Intelligence
Thang Le-Duc, Quoc-Hung Nguyen, Jaehong Lee, H. Nguyen-Xuan
Summary: In this article, the advantages of heuristic mechanisms are explored, and a new optimization framework named sequential motion optimization (SMO) is devised to enhance gradient-based methods. Inspired by balancing composite motion optimization (BCMO), SMO establishes a sequential motion chain of gradient-guided individuals to improve parameter updates. Experimental results show that SMO outperforms vanilla stochastic gradient descent (SGD) implemented via backpropagation (BP) algorithm in terms of training quality on various benchmark datasets. It is suggested that SMO has the potential to be combined with other gradient-based variants for improving its effectiveness in solving large-scale optimization problems.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2023)
Article
Green & Sustainable Science & Technology
Thinh Huynh, Anh Tuan Pham, Jaehong Lee, H. Nguyen-Xuan
Summary: In this paper, a method for optimizing the component parameters of fuel cell hybrid electric vehicles (FCHEVs) is proposed to improve their performance and reduce operating costs. The balancing composite motion optimization (BCMO) algorithm is used to design the polymer electrolyte membrane fuel cell system, lithium-ion battery, electric motor, and differential unit. The proposed method takes into account the desired performance, hydrogen consumption, fuel cell system efficiency, and power source lifespan through a single cost function. Comparative studies validate the effectiveness of the method.
INTERNATIONAL JOURNAL OF PRECISION ENGINEERING AND MANUFACTURING-GREEN TECHNOLOGY
(2023)
Article
Construction & Building Technology
Trung Kien Nguyen, Muhammad Shazwan Suhaizan, H. Nguyen-Xuan, Phuong Tran
Summary: This work proposes a new lightweight cellular concrete with controllable mechanical properties inspired by natural cellular structures, which is suitable for prefabricated engineering applications. 3D printed sacrificial formworks with lattice and TPMS architectures were used and infiltrated with foamed concrete of different densities. The compressive performance, air void characteristics, and failure mechanisms were investigated through numerical simulations and experimental tests. The gyroid cellular structure exhibited the highest compressive capacity and the bio-inspired architecture of the formworks significantly affected the air void distribution and compressive strength of the concrete.
CONSTRUCTION AND BUILDING MATERIALS
(2023)
Article
Automation & Control Systems
Tam T. Truong, Jaehong Lee, T. Nguyen-Thoi
Summary: Most previous studies on damage detection in civil engineering structures have focused on either element damage detection or joint damage detection, separately. This study proposes an effective data-driven approach using an attention based convolutional gated recurrent unit network (ACGRU) for real-time damage detection of both joint and element in frame structures.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Engineering, Civil
Chien H. Thai, P. T. Hung, H. Nguyen-Xuan, P. Phung-Van
Summary: In this paper, a new size-dependent meshfree method is introduced to analyze the free vibrations of magneto-electro-elastic (MEE) functionally graded (FG) nanoplates. The method combines the nonlocal strain gradient theory (NSGT), the higher-order shear deformation theory (HSDT), and meshfree method for the first time. The effective material properties of MEE-FG nanoplates are expressed using a power-law scheme. Numerical examples are given to investigate the effect of various parameters on the natural frequency of MEE-FG nanoplates.
ENGINEERING STRUCTURES
(2023)
Article
Computer Science, Interdisciplinary Applications
Nam V. Nguyen, Kim Q. Tran, H. Nguyen-Xuan
Summary: This study presents an efficient computational approach for analyzing the nonlinear static and dynamic behavior of functionally graded plates based on triply periodic minimal surface architectures. The study investigates the nonlinear behavior of three TPMS structures and evaluates the influence of various parameters and dynamic loads. The results demonstrate that FG-TPMS plates exhibit superior energy absorption capacity under geometric nonlinearity conditions.
ENGINEERING WITH COMPUTERS
(2023)
Article
Computer Science, Artificial Intelligence
Trong Nghia-Nguyen, Mamoru Kikumoto, H. Nguyen-Xuan, Samir Khatir, Magd Abdel Wahab, Thanh Cuong-Le
Summary: Soil compression parameters are crucial for ensuring the safety of civil engineering structures. Currently, evaluating these parameters through laboratory tests is time-consuming and labor-intensive, leading to increased construction costs. In this paper, machine learning models are employed to establish a reliable method for obtaining these parameters, using data from five different construction projects. Various models, including artificial neural network, deep neural network, and optimized deep neural network models, are compared, with the optimized models showing superior performance. Validations using data from a road project confirm the effectiveness and potential applications of the proposed methods.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Mechanics
Xiaolong Liu, Kelian Luo, Pengcheng Gao, Tao Cong, Xi Wang, Wenjing Wang
Summary: This paper investigates the formation mechanisms of the zig-zag crack region on the shattered rim of railway wheels. The zig-zag crack region, identified as a typical region for crack propagation in rolling contact fatigue behavior, was observed using scanning electron microscopy and transmission electron microscopy. The formation of the zig-zag morphology is attributed to the periodic deflection of the propagation path relative to the initial propagation plane, caused by the limited plastic deformation zone at the crack tip. Grain refinement and secondary cracks in the zig-zag crack region are a result of the large compressive and shear stresses induced by rolling contact loading.
ENGINEERING FRACTURE MECHANICS
(2024)
Article
Mechanics
Anastasia Iziumova, Aleksei Vshivkov, Ivan Panteleev, Virginia Mubassarova, Oleg Plekhov, Denis Davydov
Summary: The aim of this study was to investigate the correlation between structural, acoustic emission, and thermal characteristics of fatigue crack growth in titanium alloys. Cluster analysis of the acoustic emission signals revealed two different types of signals observed during the fatigue crack development. It was experimentally demonstrated that the stored energy tends to reach an asymptotic value at the final stage of fatigue crack growth and this is correlated with the twinning process intensification in titanium alloy Ti Grade 2. A correlation was assumed between the stages of change in heat flux, the cumulative energy of the first cluster of acoustic emission signals, and the crack length.
ENGINEERING FRACTURE MECHANICS
(2024)
Article
Mechanics
M. Vieira de Carvalho, I. A. Rodrigues Lopes, F. M. Andrade Pires
Summary: This study investigates the numerical challenges of fracture mechanics models within implicit quasi-static frameworks and proposes an instability criterion. The ratio of cohesive to internal power is identified as a crucial factor. Two strategies for handling fracture problems with instabilities are discussed and a comparative assessment is performed. The study also examines more complex material responses, including transformation-induced plasticity effects.
ENGINEERING FRACTURE MECHANICS
(2024)
Article
Mechanics
Thomas Duminy, Aurelien Doitrand, Sylvain Meille
Summary: This study conducted in situ wedge splitting tests on millimeter-size PMMA samples and proposed a method to determine the material tensile strength and critical energy release rate using digital image correlation and a full finite element implementation of the coupled criterion.
ENGINEERING FRACTURE MECHANICS
(2024)
Article
Mechanics
Xin Chang, Xingyi Wang, Chunhe Yang, Yintong Guo, Yanghui Wan
Summary: The influence of cyclic thermal shock and high-temperature acid etching on the Mode I fracture of shale was investigated in this study. It was found that cyclic thermal shock severely degrades the strength and fracture toughness of shale, while high-temperature acid etching treatment improves the fracture toughness. These findings are valuable for optimizing process parameters to reduce initiation pressure in deep shale formations.
ENGINEERING FRACTURE MECHANICS
(2024)
Article
Mechanics
Liaojun Yao, Mingyue Chuai, Zhangming Lyu, Xiangming Chen, Licheng Guo, R. C. Alderliesten
Summary: Methods based on fracture mechanics have been widely used in fatigue delamination growth (FDG) characterization of composite laminates. This study proposes appropriate similitude parameters to represent FDG behavior with different R-ratios.
ENGINEERING FRACTURE MECHANICS
(2024)
Article
Mechanics
Zesheng Zang, Zhonghui Li, Yue Niu, Shan Yin
Summary: This study conducted experiments and recorded signals to investigate the fracture behavior and damage evolution characteristics of coal samples. The results showed that as loading proceeds, the stress, electric potential (EP), and acoustic emission (AE) values increase, and EP and AE signals are excited when stress drops. The fracture behavior of coal samples is altered by flaw inclination, and the destruction mode becomes increasingly complicated. The damage evolution characteristics of coal samples can be evaluated and analyzed by defining the coefficient of variation (CV value) of EP and the b value of AE.
ENGINEERING FRACTURE MECHANICS
(2024)
Article
Mechanics
Clotilde Berdin, Nathalie Prud'homme
Summary: In this study, zirconia layers with different fractions of tetragonal phase and thicknesses were tested for multi-cracking behavior. Cracks perpendicular to the tensile direction were observed, showing a blunting effect into the substrate. The ratio of crack spacing at saturation to layer thickness decreased as the layer thickness increased. Unit cell modeling was used to establish a relationship between crack spacing and layer strength, which fell within the bounds of Hu and Evans model and was found to be insensitive to the tetragonal zirconia fraction.
ENGINEERING FRACTURE MECHANICS
(2024)
Article
Mechanics
Huadong Zhang, Weichen Kong, Y. H. Liu, Yuh J. Chao
Summary: Williams' series expansion crack tip solution in linear elasticity is modified to include a uniform crack face pressure. Practical methods to calculate T-stress from near crack tip stresses are outlined. The analytical results are consistent with numerical results.
ENGINEERING FRACTURE MECHANICS
(2024)
Article
Mechanics
Jiahao Kong, Haoyue Han, Tao Wang, Guangyan Huang, Zhuo Zhuang
Summary: This paper introduces a phase-field model for polymer foam materials by combining the phase-field method with the crushable foam model. The model is calibrated using experimental data and successfully simulates the fracture processes of polyurethane under different loading conditions. The study is important for the engineering applications of polymer foam materials.
ENGINEERING FRACTURE MECHANICS
(2024)