Article
Engineering, Mechanical
Shancheng Cao, Ning Guo, Chao Xu
Summary: Damage localization in plate-type structures using full-field vibration measurements has gained increased attention. A general strategy involves accurately extracting and integrating damage features from different modes for robust localization. However, measurement noise and lack of baseline data may degrade the accuracy of feature extraction, leading to conflicting evidence for damage localization. An enhanced robust principal component analysis and data fusion approach are proposed to address these challenges, resulting in a robust method for detecting multiple damage zones.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Engineering, Multidisciplinary
Sagi Rathna Prasad, A. S. Sekhar
Summary: This article investigates the application of principal component analysis-based statistical pattern analysis for early detection and localization of fatigue-induced transverse cracks in rotor shafts. The study utilizes accelerated fatigue experiments on a customized setup and denoises noise in acquired vibration and strain data using classical principal component analysis. Time- and frequency-domain statistical features extracted from different sensor signals contribute to the development of a new fused health indicator sensitive to rotor shaft cracks.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2021)
Article
Biochemical Research Methods
Kuangnan Fang, Rui Ren, Qingzhao Zhang, Shuangge Ma
Summary: Dimension reduction techniques like PCA, PLS, and CCA are extensively used in the analysis of high-dimensional omics data. Integrative analysis, which outperforms meta-analysis and individual-data analysis, has been developed for multiple datasets with compatible designs. We developed the R package iSFun to facilitate integrative dimension reduction analysis, offering comprehensive analysis options under different models and penalties.
Article
Physics, Applied
Kaixin Liu, Qing Yu, Weiyao Lou, Stefano Sfarra, Yi Liu, Jianguo Yang, Yuan Yao
Summary: Non-destructive ultrasonic testing is important for monitoring the structural health of polymer composites. However, ultrasonic data often appear as noisy signals or images containing artifacts. To reduce human factors, this study proposes an unsupervised method, using nonlinear dimensionality reduction and semantic segmentation, to effectively detect subsurface defects in composite materials.
JOURNAL OF APPLIED PHYSICS
(2022)
Article
Computer Science, Artificial Intelligence
Pei Li, Wenlin Zhang, Chengjun Lu, Rui Zhang, Xuelong Li
Summary: A novel robust kernel principal component analysis method with optimal mean (RKPCA-OM) is proposed to enhance the robustness of KPCA by automatically eliminating the optimal mean. The theoretical proof guarantees the convergence of the algorithm and the obtained optimal subspaces and means. Exhaustive experimental results validate the superiority of the proposed method.
Article
Cell Biology
Ichiro Wakabayashi, Yoko Sotoda, Ryoji Eguchi
Summary: Platelets are a major source of microRNAs in blood. Relationships between circulating platelet-derived miRNAs were investigated in this study to elucidate their significance as biomarkers. Six miRNAs were identified as potentially useful biomarkers reflecting platelet condition and function.
Article
Engineering, Civil
Yixian Li, Limin Sun, Wang Zhu, Wei Zhang
Summary: This study proposes a framework based on the dynamic stiffness theory for input estimation and damage localization in structural health monitoring. By using Fourier transform and principal component analysis, the method can effectively estimate input and reconstruct responses in both frequency and time domains.
FRONTIERS OF STRUCTURAL AND CIVIL ENGINEERING
(2022)
Article
Engineering, Civil
Md Riasat Azim, Mustafa Gul
Summary: This article proposes a damage detection framework for truss railway bridges based on operational strain responses, utilizing principal component analysis to assess the condition of the bridges. By comparing the geometric distance of coordinates in the principal component space between baseline and damaged bridge conditions, useful damage indicators are obtained. The method can assist existing bridge maintenance techniques and contribute to the development of an effective structural health monitoring framework.
STRUCTURE AND INFRASTRUCTURE ENGINEERING
(2021)
Article
Engineering, Chemical
Abdalhamid Rahoma, Syed Imtiaz, Salim Ahmed
Summary: Sparse principal component analysis (SPCA) provides a sparse description of the loading matrix. This article proposes two methods to calculate confidence intervals of the loading values, which outperform traditional PCA and benchmark SPCA methods in fault detection and diagnosis.
CHEMICAL ENGINEERING SCIENCE
(2021)
Article
Engineering, Multidisciplinary
Shancheng Cao, Zhiwen Lu, Dongwei Wang, Chao Xu
Summary: A novel robust multidamage localization method is proposed based on adaptive denoising and data fusion, which significantly enhances the accuracy of multi-damage localization by optimizing the process of damage feature extraction and data fusion.
Article
Engineering, Civil
Dapeng Wang, Wenda Zhang
Summary: This paper proposes a damage detection method based on principal component analysis (PCA) and deep convolutional neural network (DCNN) using dynamic response measured by FBG sensor arrays. The method accurately predicts damage levels by analyzing the raw dynamic signal with PCA and labeling the training data for DCNN models.
JOURNAL OF CIVIL STRUCTURAL HEALTH MONITORING
(2023)
Article
Nuclear Science & Technology
Wen Zhou, Jie Hou
Summary: This paper combines principal component analysis (PCA) and contribution analysis to achieve real-time monitoring and fault isolation of the system and equipment. Through the application on molten salt reactor (MSR), the advantages and applicability of these two analysis methods are demonstrated. With quantitative analysis and the modified contribution analysis method, the signal causing the anomaly can be diagnosed more accurately.
ANNALS OF NUCLEAR ENERGY
(2022)
Article
Multidisciplinary Sciences
Makoto Mizuno, Hideaki Aoyama, Yoshi Fujiwara
Summary: This study introduces a novel method called complex Hilbert principal component analysis (CHPCA) and constructs a synchronization network using Hodge decomposition, which enables the extraction of significant comovements with a time lead/delay in high-dimensional data and identification of the time-structure of correlations. Applied to Japanese beer market data, this method reveals co-movements across multiple products in the consumer choice process and uncovers remarkable customer heterogeneity.
Article
Biochemical Research Methods
Charles C. David, Chris S. Avery, Donald J. Jacobs
Summary: JEDi software is an upgraded tool that employs multithreading and user-friendly interface for rapid investigation of conformational motions of biopolymers, including multiple chain proteins. It offers options for Cartesian-based coordinates (cPCA) and internal distance pair coordinates (dpPCA) to construct covariance, correlation, and partial correlation matrices.
BMC BIOINFORMATICS
(2021)
Article
Engineering, Multidisciplinary
Shenglan Ma, Shurong Ren, Zhining Chen, Chen Wu, Shaofei Jiang
Summary: This paper proposes a comprehensive evaluation index based on acoustic emission (AE) technology and principal component analysis (PCA) to address the issue of low accuracy of any single AE parameter in identifying damage to wooden beams caused by accumulative damage. The proposed index can divide the damage evolution process of wooden beams into four stages and determine the time of crack initiation. The proposed wooden beam damage detection method based on AE technique provides a reference for further research on the damage evolution mechanism of wooden structures and in situ monitoring methods.
Article
Chemistry, Analytical
Diego Sandoval, Urko Leturiondo, Yolanda Vidal, Francesc Pozo
Summary: In order to reduce maintenance costs of offshore wind turbines, this study proposes enhancing condition monitoring techniques for pitch bearings. Entropy indicators show higher accuracy in detecting damage for low-speed bearings, contributing to a more reliable diagnosis when combined with regular indicators.
Editorial Material
Chemistry, Analytical
Francesc Pozo, Diego A. Tibaduiza, Yolanda Vidal
Article
Energy & Fuels
Cristian Velandia-Cardenas, Yolanda Vidal, Francesc Pozo
Summary: Wind power is a cleaner and cheaper energy source compared to others, but challenges related to operation and maintenance of wind farms contribute to increased costs. A fault detection methodology is proposed in this paper to improve alarm detection for wind turbine gearboxes by applying data analysis and processing techniques to real SCADA data.
Article
Chemistry, Analytical
Jersson X. Leon-Medina, Maribel Anaya, Nuria Pares, Diego A. Tibaduiza, Francesc Pozo
Summary: A data-driven pattern-recognition methodology for structural damage classification was developed and successfully applied to wind-turbine foundations in this study, with performance measures over 99.9% after validation through experimental tests.
Article
Chemistry, Analytical
Diego F. Godoy-Rojas, Jersson X. Leon-Medina, Bernardo Rueda, Whilmar Vargas, Juan Romero, Cesar Pedraza, Francesc Pozo, Diego A. Tibaduiza
Summary: This study performs multivariate time series temperature prediction in an electric arc furnace using a deep learning approach. An attention mechanism is utilized to improve the long-term dependency of temperature predictions. The results suggest that the attention-based mechanism outperforms other recurrent neural network architectures in terms of Average Root Mean Square Error (ARMSE).
Article
Chemistry, Multidisciplinary
Ramin Ghiasi, Mohammad Noori, Sin-Chi Kuok, Ahmed Silik, Tianyu Wang, Francesc Pozo, Wael A. Altabey
Summary: In this study, a non-probabilistic structural damage identification technique based on an optimization algorithm and interval mathematics is proposed for uncertainty-oriented damage identification. The method takes into account uncertainty quantification and provides support for structural health diagnosis under uncertain conditions. The technique is implemented using the slime mold algorithm (SMA) for model updating and an enhanced variant of SMA (ESMA) is developed. Results show that the proposed method reduces computation time and improves the certainty of damage detection.
APPLIED SCIENCES-BASEL
(2022)
Article
Mathematics
Jersson X. Leon-Medina, Nuria Pares, Maribel Anaya, Diego A. Tibaduiza, Francesc Pozo
Summary: The study introduced a machine learning methodology to improve signal processing and develop classification methodologies for EN applications. The approach enhanced sensor classification accuracy through data preprocessing and machine learning algorithms.
Article
Physics, Multidisciplinary
Juan Carlos Perafan-Lopez, Valeria Lucia Ferrer-Gregory, Cesar Nieto-Londono, Julian Sierra-Perez
Summary: The article presents a clustering hybrid algorithm for automatically grouping datasets into a two-dimensional space using DBSCAN, with parameters MinPts and Eps automated using nearest neighbor and a genetic algorithm. Factor Analysis was used for preprocessing high-dimensional datasets. The performance of the algorithm, FA+GA-DBSCAN, was evaluated using artificial datasets, showing lower error probability in clustering denser datasets.
Article
Chemistry, Analytical
Alejandra Amaya, Julian Sierra-Perez
Summary: This paper presents a data-driven methodology for structural health monitoring (SHM) in reinforced concrete structures using embedded fiber optic sensors and pattern recognition techniques. A prototype structure was built and instrumented with fiber Bragg gratings (FBGs) bonded directly to the reinforcing steel bars embedded in the concrete. Datasets for pristine and damaged states were acquired, and classifiers based on Mahalanobis' distance were developed for supervised and unsupervised pattern recognition, achieving an accuracy of up to 98%.
Editorial Material
Energy & Fuels
Jersson X. X. Leon-Medina, Francesc Pozo
Article
Energy & Fuels
Jorge Mario Tamayo-Avendano, Ivan David Patino-Arcila, Cesar Nieto-Londono, Julian Sierra-Perez
Summary: This research aims to improve the energy output of small wind turbines by investigating the effects of bend-twist coupling on their performance. The study found that using coupling as a passive pitch mechanism can increase energy yield by 3%, while causing only a small increase in blade root flapwise moment and rotor thrust force.
Proceedings Paper
Engineering, Civil
Andres R. Herrera, Esteban Paniagua, Carlos A. Blandon, Carlos A. Riveros-Jerez, Jorge Aristizabal, Julian Sierra-Perez
Summary: This paper presents the development of a low-cost fiber optic-based transducer for force/stress measuring in civil structure pillars. The transducers successfully measured strain levels in reduced scale columns, and a sensing scheme was designed for a 32-story-tall building. It is expected to detect load redistribution during the construction and operational phases.
EUROPEAN WORKSHOP ON STRUCTURAL HEALTH MONITORING (EWSHM 2022), VOL 2
(2023)
Proceedings Paper
Engineering, Civil
Jersson X. Leon-Medina, Nuria Pares, Maribel Anaya, Diego Tibaduiza, Francesc Pozo
Summary: This work presents a novel methodology to improve the structural damage classification of wind turbine foundations. The methodology includes stages such as data acquisition, data normalization, data unfolding, feature extraction, and machine learning classification. Through validation with experimental data, this methodology demonstrates good classification accuracy.
EUROPEAN WORKSHOP ON STRUCTURAL HEALTH MONITORING (EWSHM 2022), VOL 3
(2023)
Article
Energy & Fuels
Juan Pablo Jaramillo-Cardona, Juan Carlos Perafan-Lopez, Jose Luis Torres-Madronero, Cesar Nieto-Londono, Julian Sierra-Perez
Summary: Hydroelectric plants meet Colombian electricity demand, but wind power is emerging as an alternative to improve the energy matrix. This study evaluates the feasibility of Small Wind Turbine projects and concludes that SWTs excel as an off-grid energy system alternative.
CT&F-CIENCIA TECNOLOGIA Y FUTURO
(2022)
Article
Mechanics
Jersson X. Leon-Medina, Maribel Anaya, Diego A. Tibaduiza, Francesc Pozo
Summary: A comparative study of four manifold learning algorithms was conducted in the context of structural health monitoring for damage classification. The results indicated that employing the Isomap algorithm led to the best classification accuracy in the experimental setup.
JOURNAL OF APPLIED AND COMPUTATIONAL MECHANICS
(2021)
