Article
Engineering, Industrial
Seunggyu Lee
Summary: In this study, a support vector machine was used as a metamodel to reduce the numerical cost of large numerical models in Monte Carlo simulations. A modified margin of the support vector machine was utilized for active learning, with the proportion in the design space serving as the criterion to end the learning process. The proposed method was successfully applied to various numerical examples in the study.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2021)
Article
Engineering, Multidisciplinary
Hua-Ping Wan, Jia-Rui Gan, Yi-Kai Zhu, Zeng Meng
Summary: The study focuses on accurately calculating the failure probabilities of high-dimensional complex systems using surrogate models and subset simulation (SS), with the aim of enhancing computational efficiency and accuracy through the use of multi-class adaptive support vector machine (MASVM). The proposed SS-MASVM method proves to be efficient and accurate for assessing the reliability of complex systems.
COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING
(2024)
Article
Engineering, Industrial
Atin Roy, Subrata Chakraborty
Summary: In this study, a three-stage adaptive support vector regression (SVR) model is built to alleviate the scarcity of samples in the reliability evaluation of structures with implicit limit state functions (LSFs). The model employs sequential and importance sampling techniques to ensure a sufficient number of simulation points near the failure plane for accurate estimation of reliability.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2022)
Article
Engineering, Industrial
Nick Pepper, Luis Crespo, Francesco Montomoli
Summary: This work demonstrates how to approximate the failure probability of an expensive computational model with reliability requirements using Support Vector Machines. An algorithm is proposed to select informative parameter points to improve the approximation accuracy iteratively. Additionally, a method is provided to quantify the uncertainty in the Limit State Function and estimate an upper bound to the failure probability using geometrical arguments.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2022)
Article
Computer Science, Interdisciplinary Applications
Chunyan Ling, Zhenzhou Lu
Summary: The proposed method introduces a novel two-stage meta-model importance sampling based on support vector machine (SVM) to efficiently estimate structural failure probability. It provides an algorithm to efficiently deal with multiple failure regions and rare events, with the SVM model accurately recognizing the states of samples. Several examples are performed to show the feasibility of the proposed method.
STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION
(2021)
Article
Computer Science, Interdisciplinary Applications
Bin Xie, Yanzhong Wang, Yunyi Zhu, Fengxia Lu
Summary: This paper proposes a novel reliability analysis method, RVM-SS, which combines relevance vector machine (RVM) and subset simulation (SS). It improves the efficiency and accuracy of reliability analysis by using RVM to approximate limit states and performing SS based on the constructed RVM. The updated RVM has high prediction accuracy, resulting in accurate failure probability estimation.
STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION
(2023)
Article
Engineering, Civil
Atin Roy, Subrata Chakraborty
Summary: The dual metamodeling approach is used to address the stochastic nature of earthquakes, while avoiding prior distribution assumption, a direct response approximation approach is attempted here. Furthermore, an adaptive support vector regression-based metamodeling is proposed for selecting new training samples near the failure boundary with consideration to accuracy and efficiency. The effectiveness of the approach is demonstrated by comparing it with direct Monte Carlo simulation technique and an active learning-based Kriging approach.
JOURNAL OF EARTHQUAKE ENGINEERING
(2023)
Article
Engineering, Environmental
Xueyou Li, Yadong Liu, Zhiyong Yang, Zhenzhu Meng, Limin Zhang
Summary: The paper proposes an efficient slope reliability analysis method based on active learning support vector machine (SVM) and Monte Carlo simulation (MCS), which updates the model by selecting appropriate training samples to improve efficiency and accuracy. The effectiveness of the method is demonstrated using four slope examples and compared with other surrogate models, showing better computational efficiency and similar estimation accuracy.
BULLETIN OF ENGINEERING GEOLOGY AND THE ENVIRONMENT
(2021)
Article
Construction & Building Technology
Yutai Yang, Weizhe Sun, Guoshao Su
Summary: In this paper, a support-vector-machine-based grasshopper optimization algorithm (GOA) is proposed for structural reliability analysis of large and complex structures. The reliability problem is transformed into an optimization problem, and a surrogate model of the performance function is constructed using the support vector machine model. The GOA is used to search for the most probable point (MPP), and an iterative method is constructed to improve the accuracy of the surrogate model. Numerical cases and a long-span bridge application demonstrate the significant advantages of the method in computational accuracy and efficiency.
Article
Environmental Studies
Kamran Mostafaei, Shaho Maleki, Mohammad Zamani Ahmad Mahmoudi, Dariusz Knez
Summary: This research used Support Vector Machine (SVM) to predict the financial perspective of the Helichal granite mine in Iran for a duration of 30 years. The study collated financial data from the previous ten years of exploitation operations and created 100 simulations of net present value (NPV) using Monte Carlo technique. The results showed a high correlation (96%) between the SVM-predicted NPVs and the Monte Carlo-simulated NPVs, indicating the reliability of the SVM approach in anticipating the financial profitability of mining projects.
Article
Engineering, Multidisciplinary
Zhengran Lu, Chao Guo
Summary: The fastener, as a core component of the formwork support system, has a significant influence on the system's bearing capacity. The increase in failure probability of the fastener leads to a significant decrease in the magnitude and probability of the decrease in the bearing capacity. Deterministic analysis based on the integrity of all fasteners is impractical and unsafe.
AIN SHAMS ENGINEERING JOURNAL
(2023)
Article
Engineering, Geological
Mohammad Aminpour, Reza Alaie, Navid Kardani, Sara Moridpour, Majidreza Nazem
Summary: This paper presents a highly efficient machine learning-aided reliability technique for stochastic reliability analysis in geotechnical engineering. The proposed technique accurately predicts the probability of failure with significantly reduced computational time compared to traditional methods.
Article
Engineering, Mechanical
Axay Thapa, Atin Roy, Subrata Chakraborty
Summary: An adaptive Kriging based metamodeling approach for tunnel reliability analysis is proposed, which improves prediction accuracy by iteratively selecting new training points and updating the model until no points are left in the reduced space.
PROBABILISTIC ENGINEERING MECHANICS
(2022)
Article
Computer Science, Interdisciplinary Applications
Michael D. Shields, Dimitris G. Giovanis, V. S. Sundar
Summary: The paper proposes the use of an affine invariant ensemble MCMC sampler for conditional sampling to address extreme cases where subset simulation breaks down. The algorithm automatically varies step size and is particularly effective for estimating failure probabilities in strongly non-Gaussian and lower effective dimension scenarios.
COMPUTERS & STRUCTURES
(2021)
Article
Engineering, Civil
Hongyuan Guo, You Dong, Paolo Gardoni
Summary: This paper presents a novel method for time-dependent reliability analysis based on point evolution kernel density and adaptive subset simulation. The proposed method captures the cumulative density function of the first failure time to achieve time-dependent reliability analysis with high computational efficiency and accuracy.
Article
Engineering, Geological
H. Alhajj Chehade, D. Dias, M. Sadek, O. Jenck, F. Hage Chehade
Summary: This paper examines the seismic behavior of earth retaining walls using the upper bound theorem of limit analysis and discretization technique, with pseudo-static approach to model seismic effects. The extended methodology proposed is validated against conventional limit analysis results and a parametric study is conducted to identify the effects of key parameters on required reinforcement strength.
Article
Chemistry, Multidisciplinary
Hung Van Pham, Daniel Dias
Summary: This paper investigates the behavior of rigid inclusion-improved soil under different loading conditions through laboratory tests and numerical simulation. The study shows that the HYP model can better simulate the soil arching mechanism, providing good agreement with experimental data.
APPLIED SCIENCES-BASEL
(2021)
Article
Engineering, Geological
Hoang-Giang Bui, Jelena Ninic, Ngoc-Anh Do, Daniel Dias, Guenther Meschke
Summary: This technical note presents a consistent finite element formulation for tunnel linings design by introducing a variational consistently linearized formulation to consider nonlinear interaction between lining structure and ground. The proposed VHRM model allows for more efficient solution of nonlinear system equations and is applicable for various types of finite elements.
INTERNATIONAL JOURNAL FOR NUMERICAL AND ANALYTICAL METHODS IN GEOMECHANICS
(2022)
Article
Chemistry, Multidisciplinary
Chi Thanh Nguyen, Ngoc Anh Do, Daniel Dias, Van Vi Pham, Gospodarikov Alexandr
Summary: This paper presents an improved HRM method for estimating the internal forces induced in square and rectangular tunnel linings. The results show that the lateral earth pressure coefficient and the tunnel lining flexibility ratio have a significant impact on the internal forces induced.
APPLIED SCIENCES-BASEL
(2022)
Article
Engineering, Civil
Van Vi Pham, Ngoc Anh Do, Daniel Dias, Chi Thanh Nguyen, Van Kien Dang
Summary: This paper focuses on the different behavior of sub-rectangular tunnels under static loading compared to circular and rectangular tunnels. It provides designers with information to choose the appropriate tunnel shape solution and investigates the influence of various parameters on tunnel behavior.
TRANSPORTATION INFRASTRUCTURE GEOTECHNOLOGY
(2023)
Article
Materials Science, Multidisciplinary
Xiancheng Mei, Chuanqi Li, Qian Sheng, Zhen Cui, Jian Zhou, Daniel Dias
Summary: This study proposes a novel rubber-sand concrete (RSC) as an aseismic material and uses an artificial intelligence model, the back propagation neural network (BPNN), to predict its uniaxial compressive strength (UCS). The results show that the LSO-BPNN hybrid model has the best prediction capability. Sensitivity analysis reveals that rubber and sand have the greatest impact on UCS prediction.
MECHANICS OF ADVANCED MATERIALS AND STRUCTURES
(2023)
Article
Engineering, Geological
Dianchun Du, Daniel Dias, Ngocanh Do
Summary: This paper presents a simple and effective calculation process for analyzing the interaction between rock mass and support structure in tunnel lining design. By considering the stress redistribution and convergence caused by delayed installation, the method combines the hyperstatic reaction method and convergence confinement method. The validity of the method is verified through the comparison of results, and the influence of various parameters is discussed.
PROCEEDINGS OF THE INSTITUTION OF CIVIL ENGINEERS-GEOTECHNICAL ENGINEERING
(2022)
Article
Chemistry, Multidisciplinary
Jiamin Zhang, Daniel Dias, Qiujing Pan, Chunjing Ma, Cristina de Hollanda Cavalcanti Tsuha
Summary: This paper presents numerical simulations to assess the long-term performance of thermo-active pile systems in tropical environments for different energy demands, highlighting the significance of groundwater flow in improving thermal balance.
APPLIED SCIENCES-BASEL
(2022)
Article
Chemistry, Multidisciplinary
Duc Phi Do, Xiangfeng Guo, Daniel Dias
Summary: In this paper, the effects of uncertainties of time-dependent rock behavior and anisotropy of initial stress state on the stability of a tunnel supported by a double flexible/concrete liner are investigated. Subset simulation and Sobol global sensitivity analysis are used to estimate the failure probability of the tunnel supports and quantify the importance of different random parameters. The results show that considering the anisotropy of initial stresses increases the failure probability, especially in the concrete support elements. Furthermore, the parameters of initial stresses are found to be of main importance for both liners according to the Sobol indices.
APPLIED SCIENCES-BASEL
(2022)
Article
Chemistry, Multidisciplinary
Chuanqi Li, Jian Zhou, Daniel Dias, Yilin Gui
Summary: A novel machine learning model, KELM-GWO, was proposed for predicting rock compressive strength, with the best performance indices. Porosity was identified as the most important parameter for predicting UCS using the mean impact value (MIV) technique.
APPLIED SCIENCES-BASEL
(2022)
Article
Chemistry, Physical
Chuanqi Li, Xiancheng Mei, Daniel Dias, Zhen Cui, Jian Zhou
Summary: This paper proposes a novel hybrid artificial neural network model optimized using a reptile search algorithm with circle mapping to predict the compressive strength of rice husk ash concrete. The proposed model achieved the most satisfactory prediction accuracy regarding R-2 (0.9709), VAF (97.0911%), RMSE (3.4489), and MAE (2.6451), outperforming previously developed models.
Review
Green & Sustainable Science & Technology
Pooya Dastpak, Rita L. Sousa, Daniel Dias
Summary: Sinkholes are a major underground hazard that threatens infrastructure and lives, with significant costs and increased flood risk. Urban sinkholes are mainly manmade due to soil erosion from aging pipes. Climate change, storm surges, and urbanization contribute to increased subsidence in urban environments. Further experimental and numerical research is urgently needed to expand understanding of internal soil erosion due to defective pipes (SEDP).
Article
Chemistry, Multidisciplinary
Chuanqi Li, Daniel Dias
Summary: This paper utilizes four metaheuristic optimization algorithms to optimize the random forest model for predicting the rock elasticity modulus (EM), and evaluates the predictive performance of different models. The results show that the PRO-RF model achieves the best prediction accuracy, and porosity (Pn) is the most important variable for predicting the rock EM. This study provides a good example for the subsequent application of soft techniques in EM and other important rock parameter estimations.
APPLIED SCIENCES-BASEL
(2023)
Article
Engineering, Civil
Ngoc Anh Do, Van Vi Pham, Daniel Dias
Summary: This paper focuses on a new pseudo-static loading scheme for sub-rectangular tunnels using the Hyperstatic Reaction Method (HRM). New equations that allow for the computation of applied active loading and a variable spring stiffness coefficient are developed to depict the interaction between the soil and the tunnel lining. Through a numerical study considering various factors, the proposed loading scheme is calibrated and validated, demonstrating the effectiveness of the developed HRM method for the preliminary seismic design of sub-rectangular tunnels.
SUSTAINABLE AND RESILIENT INFRASTRUCTURE
(2023)
Article
Construction & Building Technology
Dianchun Du, Di Lei, Keqi Liu, Daniel Dias
Summary: This study investigates the lining internal forces of a quasi-rectangular tunnel designed according to the Hoek-Brown criterion in rock masses. By transforming the strength parameters and using the Hyperstatic Reaction Method (HRM), the influence of different tunnel depths, rock strengths, and geological parameters on the lining internal forces is analyzed. The results show that these parameters have a significant effect on the lining internal forces of the quasi-rectangular tunnel.
Article
Engineering, Civil
A. R. Ibrahim, D. A. Makhloof
Summary: The unavoidable heterogeneity in the mechanical characteristics of concrete is crucial to consider in the design of high-rise buildings. This study investigates the spatial variability of material properties and proposes a framework to capture the stochastic response and assess the reliability of structural components. The results demonstrate the importance of accounting for material uncertainty in ensuring the safety of high-rise buildings.
Article
Engineering, Civil
Zhiqiang Wan
Summary: This paper emphasizes the importance of global sensitivity analysis for stochastic dynamical systems with multiple uncertain parameters and proposes a global sensitivity index suitable for this purpose. The research findings demonstrate that the proposed approach exhibits high efficiency and accuracy.
Article
Engineering, Civil
Max Ehre, Iason Papaioannou, Daniel Straub
Summary: Reliability sensitivity analysis is a method to measure the influence of uncertain input parameters on the probability of failure in a system. Statistically dependent inputs pose challenges in computing and interpreting sensitivity indices. This study introduces a separation of effects between the probabilistic model and computational model to compute the independent and full contributions of all inputs. By using hierarchically structured isoprobabilistic transformations, the full set of variance-based sensitivity indices can be computed with a single set of failure samples obtained from a rare event estimation method.
Article
Engineering, Civil
Ze Yuan, Quanwang Li, Kefei Li
Summary: This paper proposes a method to determine a measurement plan for durability assessment of concrete structures, by calibrating the models using Bayesian updating and linear fitting in order to achieve the required accuracy. The paper establishes probabilistic time-dependent models for surface chloride concentration and chloride diffusion coefficient, and discusses the key factors affecting the accuracy of the models.
Article
Engineering, Civil
Ziqi Wang
Summary: This study addresses the fundamental limitation of equivalent linearization methods in nonlinear random vibration analysis, proposing a method to construct an estimator that converges on the nonlinear system solution using a limited number of nonlinear system simulations and optimizing the equivalent linear system to approach the nonlinear system solution quickly, especially for rare event probabilities.
Article
Engineering, Civil
Min Li, Srinivasan Arunachalam, Seymour M. J. Spence
Summary: This paper presents a multi-fidelity approach for computing small failure probabilities in engineering systems. By integrating information from different levels of model fidelity, the required number of high-fidelity model runs is reduced while maintaining accuracy in estimating failure probabilities.
Article
Engineering, Civil
Aritra Chatterjee, Trisha Chakravorty, Baidurya Bhattacharya
Summary: This paper presents a methodology to determine the system reliability of commonly used steel moment connections that have been designed according to current element based procedures. A general expression is derived to modify element resistance factors and meet specified system reliability targets.
Article
Engineering, Civil
Yuanqin Tao, Kok-Kwang Phoon, Honglei Sun, Jianye Ching
Summary: This study derives theoretical and approximate variance reduction functions (VRFs) for a potential inclined slip line in a spatially variable soil. The study investigates one-dimensional (1D) VRFs and proposes approximate VRFs for the one-dimensional Whittle-Mate 'rn (WM) model and the one-dimensional cosine Whittle-Mate 'rn (CosWM) model. The paper also derives theoretical scales of fluctuation and VRFs for commonly used two-dimensional autocorrelation models and proposes general approximations for the VRF over an inclined line.
Article
Engineering, Civil
Amir H. Khodabakhsh, Seid H. Pourtakdoust
Summary: The Fokker-Plank-Kolmogorov (FPK) equation is a crucial model for understanding and improving the performance of engineering systems. However, its solution is challenging due to its high dimensionality. This study introduces FPK-DP Net, a physics-informed network that can effectively solve high-dimensional FPK equations and demonstrates its applicability and accuracy through numerical implementations on benchmark problems.
Article
Engineering, Civil
Wouter Jan Klerk, Vera van Bergeijk, Wim Kanning, Rogier Wolfert, Matthijs Kok
Summary: This paper examines the reliability of flood defence systems under shock-based degradation and compares different maintenance concepts. The results show that the current maintenance concept fails to meet the reliability requirements for revetment failure. Additional inspections and targeted interventions can significantly reduce total cost and improve the robustness of flood defence systems.
Article
Engineering, Civil
Xuejing Wang, Xin Ruan, Joan R. Casas, Mingyang Zhang
Summary: This paper proposes a probabilistic Gaussian mixture model for simulating heavy vehicle scenarios on long-span bridges under free-flow conditions. The study utilizes a non-stationary Poisson process to simulate the uneven occurrence of heavy vehicles in different lanes, considering the correlation of gross vehicle weights within close range. The results show that the correlation and stationarity of vehicle distribution location significantly affect the structural responses.
Article
Engineering, Civil
Jianhua Xian, Ziqi Wang
Summary: This study presents an importance sampling formulation based on adaptively relaxing parameters, providing a unified framework for various existing variance reduction techniques and laying the foundation for creating new importance sampling strategies. It proposes two importance sampling strategies for low-dimensional and high-dimensional problems, which are crucial for fragility analysis in performance-based engineering.
Article
Engineering, Civil
Chi Xu, Jun Chen, Jie Li
Summary: This study proposes a new algorithm to determine the probability distributions of the live load duration and compares the results with Monte Carlo simulation. The algorithm allows the exact determination of design live loads based on a predefined exceeding probability, providing guidance for engineering design.
Article
Engineering, Civil
Hongyuan Guo, You Dong, Emilio Bastidas-Arteaga
Summary: This paper presents a general reliability assessment framework for RC structures based on a Mixed Bayesian network, taking into account environmental parameters, chloride transport, and concrete crack inspection. The case study reveals that early detection of large cracks may lead to an overestimation of failure probability by about 500%.
Article
Engineering, Civil
Changle Peng, Cheng Chen, Tong Guo, Weijie Xu
Summary: Reliability Analysis (RA) is critical in structural design and performance evaluation. This study proposes a novel learning function, SEUR, for surrogate model-assisted RA to improve efficiency and accuracy. The SEUR function is demonstrated to be more effective and efficient in dealing with nonlinear problems, small probabilities, and complex limit states.