相关参考文献
注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article
Engineering, Multidisciplinary
Qingzhao Kong et al.
Summary: This study proposes novel semi-supervised networks integrating deep autoencoder and pseudo-labels propagation for condition assessment with insufficient labeled data. The proposed method achieves a powerful effectiveness and robustness in identifying post-earthquake damage scenarios based on numerical simulations and full-scale laboratory tests.
Article
Computer Science, Interdisciplinary Applications
Yuchen Liao et al.
Summary: This paper proposes an innovative attention-based recurrent neural network model for predicting bridge responses under dynamic loads such as earthquakes. The model utilizes the recent advances in deep learning and demonstrates improved accuracy and reliability compared to traditional models through validation with numerical and real-world bridge data.
COMPUTERS & STRUCTURES
(2023)
Article
Automation & Control Systems
Qingsong Xiong et al.
Summary: A novel denoising approach based on deep convolutional image-denoiser networks is proposed in this study, which demonstrates powerful capability and efficiency in denoising signal.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Automation & Control Systems
Qingsong Xiong et al.
Summary: This study proposes a novel General Tuple Recognition Framework (GTRF) that can be applied to different learning paradigms in the monitoring field. The pattern-sensitive features (PSFs) are quantitatively defined through a feature extractor intervened by deep autoencoder, and downstream label propagation is performed using an optimized fuzzy clustering algorithm. Multi-level experimental tasks, including vibration SL recognition, damage UL detection, and condition SSL assessment, were conducted to validate the effectiveness and generality of the proposed GTRF.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Engineering, Multidisciplinary
N. Sukumar et al.
Summary: In this paper, a new approach based on distance fields is introduced to accurately impose boundary conditions in physics-informed deep neural networks. By using geometry-aware trial functions and concepts from constructive solid geometry and generalized barycentric coordinates, the method constructs an approximate distance function to the boundary of a domain and exactly satisfies homogeneous and inhomogeneous boundary conditions. This approach eliminates modeling errors and ensures kinematic admissibility.
COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING
(2022)
Article
Engineering, Civil
Huan Luo et al.
Summary: This article proposes a novel data-driven method to predict lateral seismic response based on estimated lateral stiffness, integrating machine learning approach with the hysteretic model. The method significantly outperforms classical methods in both prediction capabilities and computational efficiency.
EARTHQUAKE SPECTRA
(2022)
Article
Engineering, Civil
Zekun Xu et al.
Summary: This paper proposed a recursive LSTM network for predicting nonlinear structural seismic responses with lower computational cost compared to traditional methods, demonstrating good accuracy and generalization capability.
ENGINEERING STRUCTURES
(2022)
Article
Engineering, Civil
He Zhang et al.
Summary: This study proposes a meta-model method based on the time delay neural network for efficient structural seismic response simulation. An accuracy evaluation method considering drift peak amplitudes and maximum amplitudes is also proposed, ensuring a balance between accuracy and training time. The results show that the meta models can explain over 80% of the samples, including training and testing models, of these performance parameters. The average simulation time using this method is 0.08s, which is faster than the average 24 minutes spent by the finite element method.
KSCE JOURNAL OF CIVIL ENGINEERING
(2022)
Article
Engineering, Mechanical
A. Sofi et al.
Summary: Structural Health Monitoring has gained popularity due to advancements in technology and increasing repair and rehabilitation needs. Smart sensors paired with Artificial Intelligence tools can automate monitoring systems. Despite advancements in academia, there is a lag in real-world execution, emphasizing the need for standards to bridge the gap.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Computer Science, Information Systems
Xiangyu Zhao et al.
Summary: This paper investigates the issues in feature learning methods based on CNN and proposes a new module based on wavelet attention for image classification. Experimental results demonstrate significant improvements in accuracy using this approach.
MULTIMEDIA SYSTEMS
(2022)
Review
Engineering, Civil
Yating Zhang et al.
Summary: This article reviews the current state of seismic risk mitigation policies in ten earthquake-prone countries, providing best practice recommendations organized around key stages of the seismic retrofit process. The recommendations cover risk assessment, knowledge transfer, target setting, implementation, and monitoring, with a focus on the United States' local government responsibility for seismic risk mitigation.
JOURNAL OF EARTHQUAKE ENGINEERING
(2022)
Article
Engineering, Electrical & Electronic
Ibrahim Halil Ozcan et al.
Summary: Electric motors are widely used in industrial applications for their stability, solidity, and ease of use, with mechanical bearing faults having the highest statistical occurrence percentage. Accurate detection of bearing faults is critical to prevent unexpected breakdowns. A proposed enhanced performance bearing fault diagnosis system based on vibration data processing from multiple accelerometers mounted on bearings has been validated for inner race and rolling element faults.
ELECTRICAL ENGINEERING
(2022)
Article
Engineering, Civil
Li-Ling Yu et al.
Summary: Unreinforced masonry, including non-confined unreinforced masonry (NCURM) and confined unreinforced masonry (CURM), is prone to collapse during earthquakes. This study develops accurate hysteretic models for NCURM and CURM walls using an experimental database and the Ibarra-Krawinkler model. A calibration procedure is established to determine the model parameters, and empirical equations are created based on regression analyses. A multi-degree-of-freedom shear model, constructed based on the wall hysteretic model, is also proposed and verified to have good accuracy.
ENGINEERING STRUCTURES
(2022)
Article
Geosciences, Multidisciplinary
Kim A. Johnston et al.
Summary: Preparation for natural hazards by communities is important for reducing health problems, PTSD, and aiding recovery. This study examines community engagement approaches in Australia and proposes a community-led engagement model for preparedness, which highlights the need for understanding local communities and aims to build resilience.
Article
Chemistry, Analytical
Xiaorui Shao et al.
Summary: This article proposes a domain adaptive and lightweight framework for fault diagnosis based on 1D-CNN, which can extract features with robustness and domain invariance through CORAL processing to minimize domain shifts, effectively improving FD performance.
Article
Mathematics, Interdisciplinary Applications
Stefano Markidis
Summary: Physics-Informed Neural Networks (PINN) are a powerful tool for solving scientific computing problems. This study investigates the design, implementation, and performance of PINNs using the Quantum Processing Unit (QPU) co-processor. A simple Quantum PINN is designed to solve the one-dimensional Poisson problem using a Continuous Variable (CV) quantum computing framework. The study highlights the difference in optimization effectiveness between quantum and classical PINNs, with the basic Stochastic Gradient Descent (SGD) optimizer outperforming adaptive and high-order optimizers.
FRONTIERS IN APPLIED MATHEMATICS AND STATISTICS
(2022)
Article
Engineering, Civil
Chunxiang Li et al.
Summary: This paper develops a fast seismic performance estimation methodology using deep learning techniques to predict the seismic demands of tall pier bridges. The efficiency of the method is verified through comparison with nonlinear time history analysis (NLTHA). The results show that, when trained properly, the deep learning models can provide satisfactory predictions for shear force, bending moment, and section curvature ductility. The time efficiency of deep learning models is significantly improved compared to NLTHA.
ENGINEERING STRUCTURES
(2022)
Article
Engineering, Civil
Weiping Wen et al.
Summary: This manuscript aims to develop a method for predicting the seismic responses of existing reinforced concrete (RC) frame buildings using limited building information. Through the design and analysis of 162 typical RC frame buildings, a convolutional neural network (CNN) is developed with ground motion records and five easily obtainable building parameters as inputs, and inter-story drift (IDR) and peak floor acceleration (PFA) as outputs. The proposed method is verified through independent cases and shows sufficient accuracy and computational efficiency.
ENGINEERING STRUCTURES
(2022)
Article
Engineering, Industrial
Jinsong Yu et al.
Summary: This paper introduces a Digital Twin approach for health monitoring, using a nonparametric Bayesian network to construct a model that represents the dynamic degradation process and uncertainty propagation. Real-time model updating based on GPF and DPMM is proposed to enhance adaptability and reduce uncertainty, ensuring the effectiveness of health monitoring.
JOURNAL OF MANUFACTURING SYSTEMS
(2021)
Article
Instruments & Instrumentation
Cheng Yuan et al.
Summary: This study proposed a percussion-based method to replace constant contact between timber structures and sensors due to moisture vulnerability, and experimental studies were conducted to verify the effectiveness of two automated moisture detection approaches. By incorporating machine learning and deep learning techniques, the proposed methods covered a comprehensive classification ability from 1D time-domain signal classification to 2D image classification.
SMART MATERIALS AND STRUCTURES
(2021)
Article
Construction & Building Technology
Dimitrios Tsarpalis et al.
Summary: A reduced-order modeling approach is proposed for the linear and nonlinear analysis of automated rack-supported warehouse structures under seismic loads, offering a compromise between fidelity and low computational cost.
JOURNAL OF STRUCTURAL ENGINEERING
(2021)
Article
Engineering, Geological
Yu Huang et al.
Summary: This study focused on the prediction of slope dynamic responses in earthquake-induced landslide disasters using recurrent neural network models (including Simple-RNN, LSTM, and GRU). The results indicated that recurrent neural networks outperformed traditional methods in predicting slope dynamic responses, with LSTM and GRU models showing advantages in large data scenarios.
ENGINEERING GEOLOGY
(2021)
Article
Geosciences, Multidisciplinary
David Castro-Cruz et al.
Summary: The prediction of seismic response for critical structures is sensitive to earthquake source and geological setting. Uncertainty issues must be considered in mitigation studies, with physical-based numerical simulations increasing computational burden. A proposed alternative method combines synthetic earthquake simulations with the empirical Green function method to generate realistic seismic response, tested on a real case scenario at the Cadarache nuclear site in France.
Article
Engineering, Civil
Christine A. Goulet et al.
Summary: The NGA-East Project developed a large earthquake ground motion database for the CENA region, including over 27,000 time series and pseudo-spectral acceleration data from 82 earthquakes. This database underwent multiple quality checks and is one of the largest in stable continental regions currently available.
EARTHQUAKE SPECTRA
(2021)
Proceedings Paper
Computer Science, Interdisciplinary Applications
Kirill Gadylshin et al.
Summary: The proposed method utilizes deep learning techniques to reduce numerical errors in seismic modeling, significantly improving performance. By using coarse discretization for finite difference simulation of wavefields and training a deep neural network to remove numerical errors, seismic modeling is accelerated in 2D cases by up to ten times.
COMPUTATIONAL SCIENCE AND ITS APPLICATIONS, ICCSA 2021, PT I
(2021)
Article
Acoustics
Byung Kwan Oh et al.
JOURNAL OF SOUND AND VIBRATION
(2020)
Review
Chemistry, Multidisciplinary
Gyungmin Toh et al.
APPLIED SCIENCES-BASEL
(2020)
Article
Engineering, Civil
Ruiyang Zhang et al.
ENGINEERING STRUCTURES
(2020)
Proceedings Paper
Computer Science, Information Systems
Shreyasi Ghose et al.
PROCEEDINGS OF THE CONFLUENCE 2020: 10TH INTERNATIONAL CONFERENCE ON CLOUD COMPUTING, DATA SCIENCE & ENGINEERING
(2020)
Article
Computer Science, Interdisciplinary Applications
M. Raissi et al.
JOURNAL OF COMPUTATIONAL PHYSICS
(2019)
Article
Computer Science, Interdisciplinary Applications
Yinhao Zhu et al.
JOURNAL OF COMPUTATIONAL PHYSICS
(2019)
Article
Computer Science, Information Systems
Qinglin Qi et al.
Article
Computer Science, Artificial Intelligence
Byung Kwan Oh et al.
APPLIED SOFT COMPUTING
(2017)
Article
Construction & Building Technology
Ziemowit Dworakowski et al.
STRUCTURAL CONTROL & HEALTH MONITORING
(2016)
Review
Engineering, Civil
Swagato Das et al.
JOURNAL OF CIVIL STRUCTURAL HEALTH MONITORING
(2016)
Article
Biochemical Research Methods
Svetlin A. Manavski et al.
BMC BIOINFORMATICS
(2008)