检索结果-内蒙古大学图书馆

A digital twin-driven machine learning framework for structural condition monitoring using multi-datasets

STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL 2025年

作者： Karyofyllas, George Giagopoulos, Dimitrios Jia, Xinyu Papadimitriou, Costas Aristotle Univ Thessaloniki Dept Mech Engn Bldg E14Univ Campus Thessaloniki 54124 Central Macedon Greece Hebei Univ Technol Sch Mech Engn Tianjin Peoples R China Univ Thessaly Dept Mech Engn Volos Greece

Real-time condition monitoring (CM) utilizing vibration measurements offers a proactive approach to detect faults and enable predictive maintenance. The robustness and accuracy of CM applications significantly rely on the quality of training datasets. While numerical model-generated data is commonly used in current solutions, the efficiency of CM frameworks is greatly influenced by dataset quality. This research addresses the challenge by employing data-driven modeling to enhance the efficiency and accuracy of CM frameworks. A hierarchical bayesian Modeling (HBM) framework is introduced to estimate uncertainties in finite element (FE) model parameters at the healthy state. The HBM is particularly adept at addressing uncertainties in model parameters caused by variations in experimental data, material characteristics, assembly processes, and nonlinear mechanisms under diverse loading scenarios. These uncertainties in the parameters of the FE model obtained at the healthy state of the structure are maintained for the model representing the damaged state of the structure. The FE model with the associated uncertainties is used to generate data for training a convolution neural network model for the different health states. Such training based only on observed/inferred uncertainties at the healthy state significantly enhances the robustness and overall accuracy of the CM framework against previous approaches based on arbitrarily imposed uncertainties at the healthy and damaged state of the structure. To validate the effectiveness of the proposed method, multiple experimental structural dynamics tests on a small-scale laboratory truss are conducted at the healthy and damaged states. The results demonstrate the applicability and effectiveness of the developed approach in improving the structural health monitoring process.

关键词： Damage identification finite element modeling machine learning vibration measurements hierarchical bayesian method convolutional neural network condition monitoring uncertainty quantification multiple experimental datasets

来源：评论

学校读者我要写书评

暂无评论

bayesian identification of the optimal soil-water characteristic curve (SWCC) model and reliability analysis of unsaturated loess slope from extremely sparse measurements

引用

ENGINEERING GEOLOGY 2025年 349卷

作者： Zhao, Tengyuan Yang, Yabin Xu, Ling Sun, Pingping Xi An Jiao Tong Univ Sch Human Settlements & Civil Engn Xian 710049 Shaanxi Peoples R China

Soil-water characteristic curve (SWCCs) are crucial in engineering geology and geotechnical engineering for understanding the behavior of unsaturated soils, such as loess, which directly affects permeability, shear strength, and volume change-key factors in slope stability and soil-structure interactions. Conventionally, SWCC estimation relies on multiple (saying approximately ten) measurements fitted to parameterized models. However, in practical applications, especially for medium- or small-scale projects, the availability of SWCC measurements is often extremely limited (e.g., one or two measurements) due to the time-intensive nature of the experiments. This presents significant challenges in accurately identifying suitable SWCC models and performing reliable stability analyses for unsaturated soil slopes. To address these challenges, this study employs a hierarchical bayesian framework that integrates information from similar geotechnical site, enabling robust SWCC estimation and model selection from minimal measurements with the aid of Markov Chain Monte Carlo (MCMC) sampling, thereby quantifying model uncertainties and providing more scientifically informed decision-making for construction in engineering geology. MCMC samples obtained further facilitate both the identification of the most suitable SWCC model and the quantification of associated uncertainties. Then, a reliability-based stability analysis of an unsaturated loess slope is conducted, using the optimal SWCC model and its quantified uncertainty. The proposed methodology is validated through a real-world case study, demonstrating its effectiveness in deriving reliable SWCC models and performing stability analyses under conditions of extremely sparse data. The results highlight the potential of this method as a practical tool for advancing reliability assessments of unsaturated soil slopes in engineering geology.

关键词： Unsaturated soil slopes hierarchical bayesian method Reliability analysis Markov Chain Monte Carlo (MCMC) simulation

来源：评论

学校读者我要写书评

暂无评论

An efficient hierarchical bayesian framework for multiscale material modeling

引用

COMPOSITE STRUCTURES 2025年 351卷

作者： Pyrialakos, Stefanos Kalogeris, Ioannis Papadopoulos, Vissarion Natl Tech Univ Athens MGrp Engn Simulat Lab Inst Struct Anal & Antiseism Res Zografou Campus 9 Iroon Polytech Str Zografos 15780 Greece

This paper introduces a novel approach to infer the material properties of multiscale material systems through a variety of experimental scenarios. We utilize the hierarchical bayesian paradigm which enables us to integrate multiple experimental data at different length scales and/or different material compositions, in a systematic way. Specifically, a probabilistic model is constructed which implements the Transitional Markov Chain Monte Carlo method to extract samples from the posterior distributions of both the multiscale model parameters and the hierarchical hyperparameters. The posterior distribution of the hyperparameters encapsulates the information from all the different experiments and it is utilized to derive an informed set of physical parameters, which can be used for making predictions in future material models. Feed forward neural networks play a crucial role in mitigating the computational effort of implementing the hierarchical bayesian analysis on top of multiscale nonlinear computational homogenization analyses. Their purpose is to learn and accurately emulate the nonlinear constitutive law across multiple length scales. The proposed methodology is demonstrated on a case study of carbon nanotube (CNT) reinforced cementitious material configurations through the investigation of the CNT interfacial mechanical behavior. The hierarchical bayesian framework is applied on a set of measurements gathered from independent literature experiments performed on dissimilar material compositions on the macroscopic structural scale.

关键词： Parameter identification hierarchical bayesian method Multiscale material modeling Neural networks CNT-reinforced concrete

来源：评论

学校读者我要写书评

暂无评论

Uncertainties quantification of fatigue load mixture model using hierarchical bayesian models

引用

INTERNATIONAL JOURNAL OF FATIGUE 2023年第1期174卷

作者： Liu, Xiao-Wei Lu, Da-Gang Shanghai Univ Sch Mech & Engn Sci Shanghai 200444 Peoples R China Harbin Inst Technol Sch Civil Engn Harbin 150090 Peoples R China

There are two primary challenges in current fatigue load probability modeling. Firstly, it is difficult to measure the estimation errors in the mixture model for fatigue loads, particularly in the tails with low-probability and high-stress levels. Secondly, the component number of the mixture model cannot be observed. To address these challenges, this research introduces the hierarchical bayesian mixture model and the Dirichlet process prior. A relative error measure is proposed to reveal the errors of the density tails. The results of an illustrative example show significant relative errors in the tails and a discrete distribution of the mixture number.

关键词： Fatigue loads analysis Mixture model hierarchical bayesian method Dirichlet process Rain-flow method

来源：评论

学校读者我要写书评

暂无评论

Statistical calibration of ultrasonic fatigue testing machine and probabilistic fatigue life estimation

引用

INTERNATIONAL JOURNAL OF FATIGUE 2025年 199卷

作者： Safari, Sina Montalvao, Diogo da Costa, Pedro R. Reis, Luis Freitas, Manuel Univ Bristol Dept Mech Engn Bristol BS8 1TR England Bournemouth Univ Fac Sci & Technol Dept Design & Engn ADDISONIC Res Cluster Poole HouseTalbot Campus Poole BH12 5BB England Atlantica Inst Univ P-2730036 Barcarena Portugal Univ Lisbon Inst Super Tecn IDMEC Av Rovisco Pais 1 P-1050099 Lisbon Portugal

A new statistical technique is proposed to quantify the experimental uncertainty observed during ultrasonic fatigue testing of metals and its propagation into the stress-lifetime predictive curve. hierarchical bayesian method is employed during the calibration and operation steps of ultrasonic fatigue testing for the first time in this paper. This is particularly important due to the significant dispersion observed in stress-life data within the high and very high cycle fatigue regimes. First, the measurement systems, including displacement laser readings and high-speed camera system outputs, are cross-calibrated. Second, a statistical learning approach is applied to establish the stress-deformation relationship, leveraging Digital Image Correlation (DIC) measurements of strain and laser displacement measurements at the ultrasonic machine specimen's tip. Third, an additional hierarchical layer is introduced to infer the uncertainty in stress-life curves by incorporating learned stress distributions and the distribution of fatigue failure cycles. The results identify key sources of uncertainty in UFT and demonstrate that a hierarchical bayesian approach provides a systematic framework for quantifying these uncertainties.

关键词： Ultrasonic fatigue testing Calibration hierarchical bayesian method Cyclic lifetime Uncertainty quantification (UQ)

来源：评论

学校读者我要写书评

暂无评论

MEG current source reconstruction using a meta-analysis fMRI prior

引用

NEUROIMAGE 2021年 236卷 118034-118034页

作者： Suzuki, Keita Yamashita, Okito ATR Neural Informat Anal Labs Dept Computat Brain Imaging 2-2-2 Hikaridai Seika Kyoto 6190288 Japan Nara Inst Sci & Technol Grad Sch Sci & Technol Ikoma 6300192 Japan RIKEN Ctr Adv Intelligence Project Computat Brain Dynam Team 2-2-2 Hikaridai Seika Kyoto 6190288 Japan

Magnetoencephalography (MEG) offers a unique way to noninvasively investigate millisecond-order cortical activities by mapping sensor signals (magnetic fields outside the head) to cortical current sources using current source reconstruction methods. Current source reconstruction is defined as an ill-posed inverse problem, since the number of sensors is less than the number of current sources. One powerful approach to solving this problem is to use functional MRI (fMRI) data as a spatial constraint, although it boosts the cost of measurement and the burden on subjects. Here, we show how to use the meta-analysis fMRI data synthesized from thousands of papers instead of the individually recorded fMRI data. To mitigate the differences between the meta-analysis and individual data, the former are imported as prior information of the hierarchical bayesian estimation. Using realistic simulations, we found out the performance of current source reconstruction using meta-analysis fMRI data to be better than that using low-quality individual fMRI data and conventional methods. By applying experimental data of a face recognition task, we qualitatively confirmed that group analysis results using the meta-analysis fMRI data showed a tendency similar to the results using the individual fMRI data. Our results indicate that the use of meta-analysis fMRI data improves current source reconstruction without additional measurement costs. We assume the proposed method would have greater effect for modalities with lower measurement costs, such as optically pumped magnetometers.

关键词： MEG inverse problem fMRI Meta-analysis Source reconstruction hierarchical bayesian method

来源：评论

学校读者我要写书评

暂无评论

Local-scale landslide susceptibility mapping using the B-GeoSVC model

引用

LANDSLIDES 2019年第7期16卷 1301-1312页

作者： Yang, Yang Yang, Jintao Xu, Chengdong Xu, Chong Song, Chao Southwest Petr Univ Sch Geosci & Technol Chengdu 610500 Sichuan Peoples R China Chinese Acad Sci Inst Geog Sci & Nat Resources Res State Key Lab Resources & Environm Informat Syst Beijing 100101 Peoples R China China Earthquake Adm Inst Geol Beijing 100029 Peoples R China

Local-scale landslide susceptibility mapping (LSM) provides detailed information for decision making and planning. Most published landslide susceptibility maps lack reliable information at the local scale due to the spatial heterogeneity being ignored. To enrich the local-scale information of LSM, multiple information fusion methods for the local spatial heterogeneity and regional trends of control factors are needed. However, no information fusion method has been proposed for LSM yet. In this paper, we developed a new integrated statistical method, named B-GeoSVC, under the hierarchical bayesian framework for LSM. Specifically, this model applied the GeoDetector method to fit the regional trends of control factors and employed spatially varying coefficients (SVC) model to fit the local spatial heterogeneity of each control factor. Then, the regional trends and local spatial heterogeneity information were fused within the hierarchical bayesian framework. The B-GeoSVC model was verified using data from the Duwen basin of China, which was in the central region affected by the M-S 8.0 Wenchuan earthquake that occurred on May 12, 2008. Under a cross-validation experiment, the prediction accuracy rate of the B-GeoSVC model was 86.09%, and the area under the curve was 0.93, which suggested that the B-GeoSVC model was able to achieve relatively accurate local-scale LSM and provide richer local information than traditional regional scale LSM. More importantly, not only the B-GeoSVC model could be employed as a general solution to fuse both regional and local-scale information for landslide mapping, but also offer new insights into the broader earth science and spatial statistics.

关键词： Landslide susceptibility mapping Spatial heterogeneity Regional and local information fusion GeoDetector SVC hierarchical bayesian method

来源：评论

学校读者我要写书评

暂无评论

bayesian analysis of compound Poisson process with change-point

引用

QUALITY TECHNOLOGY AND QUANTITATIVE MANAGEMENT 2019年第3期16卷 297-317页

作者： Wang, Pingping Tang, Yincai Xu, Ancha East China Normal Univ Sch Stat Shanghai Peoples R China Wenzhou Univ Coll Math & Informat Sci Wenzhou Zhejiang Peoples R China

The compound Poisson process is considered to model the frequency and the magnitude of the earthquake occurrences concurrently. Nevertheless, there are many debates on whether climate change influences the frequency of the natural disasters. In this study, we propose a compound Poisson process with change-point (CPPCP) model to fit the data with two-phase pattern. The hierarchical bayesian method is employed via assigning a common distribution for the unit-specific parameters. For comparison purpose, we also develop the maximum-likelihood method. The simulation study illustrates the applicability of our proposed model and the validity of the hierarchical bayesian method. In the analysis of the earthquake data, CPPCP model outperforms the quadratic linear regression model and the hierarchical bayesian method is superior to the maximum-likelihood method in terms of the model fitting and prediction.

关键词： Compound Poisson process change-point hierarchical bayesian method Gibbs sampler EM algorithm maximum-likelihood method

来源：评论

学校读者我要写书评

暂无评论

Stages of User Engagement on Social Commerce Platforms: Analysis with the Navigational Clickstream Data

引用

INTERNATIONAL JOURNAL OF ELECTRONIC COMMERCE 2019年第2期23卷 179-211页

作者： Kumar, Ashish Salo, Jari Li, Hongxiu Aalto Univ Sch Business Dept Mkt Espoo Finland Univ Helsinki Mkt Helsinki Finland Tampere Univ Technol Dept Ind & Informat Management Tampere Finland

Social commerce platforms have gained prominence in e-commerce, as social media has become an integral part of users' online activities. Therefore, firms have been either developing or utilizing social commerce platforms to increase user engagement by adding social shopping facility onto their electronic commerce platforms. However, managing user engagement and user interaction becomes complex when e-commerce platforms are transformed into social commerce platforms. In this study, we operationalize four distinct stages of the social commerce platform, namely, social identification, social interaction, social shopping, and transaction based on salience theory. Using clickstream data, we empirically measure user engagement in these four states by modeling users' incidence and time spent. Drawing from the PageRank algorithm, we capture the importance of ranking and distance on user engagement. The model also accounts for the effects of situational variables such as weekend;holiday;time of day;and user characteristics, such as gender and social media setting. Our results suggest that ranking and distance have significant effects on users' incidence as well as time spent on social commerce platforms. The insights from this study can be helpful in designing the social commerce platform effectively using only the customers' path navigational clickstream data from the parent social commerce platform.

关键词： Clickstream data Dijkstra's shortest path algorithm hierarchical bayesian method multivariate type-2 Tobit online communities online platforms online shopping PageRank algorithm social commerce platforms

来源：评论

学校读者我要写书评

暂无评论

A hierarchical bayesian approach for multi-site optimization of a satellite-based evapotranspiration model

引用

HYDROLOGICAL PROCESSES 2018年第26期32卷 3907-3923页

作者： Su, Yonghong Feng, Qi Zhu, Gaofeng Gu, Chunjie Wang, Yunquan Shang, Shasha Zhang, Kun Han, Tuo Chen, Huiling Ma, Jinzhu Chinese Acad Sci Northwest Inst Ecoenvironm & Resources Key Lab Ecohydrol Inland River Basin Lanzhou Gansu Peoples R China Lanzhou Univ Minist Educ Lab Western Chinas Environm Syst Lanzhou 730000 Gansu Peoples R China China Univ Geosci Sch Environm Studies Wuhan Hubei Peoples R China

Modelling is an important tool in simulating and partitioning evapotranspiration (ET). To obtain realistic partitioning of ET, a hierarchical bayesian (HB) method was used to fit the Priestly-Taylor Jet Propulsion Laboratory (PT-JPL) model against the multi-tower Flux Network (FLUXNET) datasets. Unique to the HB method is its ability to exchange of information between sites and simultaneously estimate the species-and PFT-level parameters. The results suggested that the sensitive parameters varied at the both species and PFT levels. The parameter beta (water control of soil evaporation) exhibited relatively wide species-and PFT-level posterior distributions, indicating that the original parameterization of soil moisture constraint may be problematic. Generally, the model with parameters determined by the HB approach showed better performance in predicting and partitioning ET than the original model, especially in evergreen needleleaf forests, open shrublands, closed shrublands, and woody savannas. To overcome the problem of parameter uncertainty (equifinality), direct observations of different components of ET are urgently needed in future studies, and assessments the extent to which the parameter uncertainties are reduced by the use of additional data.

关键词： differential evolution Markov chain algorithm evaporation evapotranspiration hierarchical bayesian method process-based ET model transpiration

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：