Correlation warping radius tracking for condition monitoring of rolling bearings under varying operating conditions

Article Engineering, Mechanical

Canonical correlation analysis of dimension reduced degradation feature space for machinery condition monitoring

Xiaomeng Li et al.

Summary: In this paper, a novel health indicator (HI) construction method is proposed, which improves the sensitivity to degradation process and allows quantitative measurement by reducing the dimension of degradation feature space.

MECHANICAL SYSTEMS AND SIGNAL PROCESSING (2023)

添加到收藏夹

Article Acoustics

A kernel canonical correlation analysis approach for removing environmental and operational variations for structural damage identification

Jie-zhong Huang et al.

Summary: Vibration-based damage detection relies on the observation of changes in damage-sensitive dynamic features. However, dynamic features are also sensitive to environmental and operational variations (EOVs), limiting the application of linear methods in removing environmental effects. This study proposes an improved method based on kernel canonical correlation analysis (KCCA) to remove the nonlinear effects of EOVs on dynamic features, demonstrating its success in accurately identifying damage.

JOURNAL OF SOUND AND VIBRATION (2023)

添加到收藏夹

Article Engineering, Mechanical

An integrated framework via key-spectrum entropy and statistical properties for bearing dynamic health monitoring and performance degradation assessment

Renhe Yao et al.

Summary: This paper proposes an integrated framework for mechanical bearing health monitoring (DHM) and performance degradation assessment (PDA) using key-spectrum entropy and statistical properties. The framework includes key spectrum, key-spectrum entropy, joint statistical alarm and fault identification strategy, health phase segmentation strategy, and three-dimensional (3D) key spectrums. Evaluation on eighteen sets of bearing degradation vibration signals demonstrates the validity and practical application prospects of the proposed framework.

MECHANICAL SYSTEMS AND SIGNAL PROCESSING (2023)

添加到收藏夹

Article Engineering, Mechanical

Conditional feature disentanglement learning for anomaly detection in machines operating under time-varying conditions

Haoxuan Zhou et al.

Summary: This paper proposes a novel conditional feature disentanglement learning framework to solve the disturbance in anomaly detection caused by the entanglement between operating conditions (OCs) and health states. By utilizing the variational autoencoder (VAE) and OCs information, the proposed approach performs conditional self-supervised anomaly detection (AD) and develops a feature disentanglement conditional VAE (FDCVAE) network to realize the disentanglement of OCs and health states. Experiments validate the effectiveness of the proposed method and demonstrate the superiority of the constructed anomaly indicator (ANI) in eliminating time-varying operating conditions (TVOCs) interference compared to common fault mechanisms-based and data-driven ANI. The proposed method not only achieves anomaly detection under TVOCs but also provides a new way for representation learning under variable working conditions in machine health management applications.

MECHANICAL SYSTEMS AND SIGNAL PROCESSING (2023)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Intrinsic dimension estimation method based on correlation dimension and kNN method

Haiquan Qiu et al.

Summary: In practical problems, high-dimensional data often exhibits low-dimensional structure, which can be estimated using correlation dimension methods. However, these methods tend to underestimate the true intrinsic dimension of the dataset.

KNOWLEDGE-BASED SYSTEMS (2022)

添加到收藏夹

Article Engineering, Mechanical

An unsupervised feature learning based health indicator construction method for performance assessment of machines

Liang Guo et al.

Summary: This paper proposes a method for constructing health indicators based on unsupervised feature learning, which calculates the health status of machines by building a multiscale convolutional autoencoder network and optimizing hyperparameters through genetic algorithms. Experimental results demonstrate that this method can effectively identify the degradation process of machines and achieve better performance.

MECHANICAL SYSTEMS AND SIGNAL PROCESSING (2022)

添加到收藏夹

Article Engineering, Multidisciplinary

Health indicator construction and status assessment of rotating machinery by spatio-temporal fusion of multi-domain mixed features

Yong Duan et al.

Summary: This paper proposes a multi-domain features-based spatio-temporal fusion method called SALICAE for constructing a novel health indicator for rotating machinery. The method integrates the advantages of self-attention, long short-term memory, and an improved convolutional autoencoder. The effectiveness of the proposed method is verified and its accuracy is approximately 85.3% under different working conditions.

MEASUREMENT (2022)

添加到收藏夹

Article Engineering, Mechanical

Bearing signal models and their effect on bearing diagnostics

P. Borghesani et al.

Summary: This study discusses the signal models of rolling-element bearings and their impact on the signals, develops a generalized model that includes both real and pseudo-cyclostationary characteristics, and assesses the relative importance of first and second-order cyclostationary fault symptoms.

MECHANICAL SYSTEMS AND SIGNAL PROCESSING (2022)

添加到收藏夹

Article Engineering, Multidisciplinary

A health indicator construction method based on self-attention convolutional autoencoder for rotating machine performance assessment

Weipeng Ma et al.

Summary: This paper proposes an unsupervised method for constructing Health Indicators (HIs) by incorporating a basic network unit and a self-attention convolutional autoencoder (SACA) to achieve better feature extraction effects. The experimental results show that the proposed method has better trendability and scale similarity compared to other five methods.

MEASUREMENT (2022)

添加到收藏夹

Article Engineering, Mechanical

Investigation on optimal discriminant directions of linear discriminant analysis for locating informative frequency bands for machine health monitoring

Tongtong Yan et al.

Summary: This article investigates how the LDA algorithm can locate informative frequency bands for machine health monitoring, and proposes a new health indicator to detect fault initiation and track degradation trajectory during a machine life cycle. The effectiveness of the proposed method is validated through simulation and experimental studies. The interpretability of this approach provides a new perspective for developing physics-informed machine learning technology by combining data-driven models with physics-based knowledge.

MECHANICAL SYSTEMS AND SIGNAL PROCESSING (2022)

添加到收藏夹

Article Engineering, Chemical

Fault detection and diagnosis of nonlinear dynamical processes through correlation dimension and fractal analysis based dynamic kernel PCA

Wahiba Bounoua et al.

Summary: This study introduces a novel Dynamic Kernel PCA method for process monitoring in nonlinear dynamical systems, using the powerful theory of the nonlinear Fractal Dimension. The Fractal-based DKPCA integrates the two strategies to overcome the shortcomings of traditional methods and showed superior performance in fault detection and diagnosis compared to contemporary approaches.

CHEMICAL ENGINEERING SCIENCE (2021)

添加到收藏夹

Article Engineering, Multidisciplinary

A nonparametric health index and its statistical threshold for machine condition monitoring

Jingjing Zhong et al.

Summary: Machine condition monitoring uses monitoring data to evaluate machine health conditions and conduct condition-based maintenance. Popular indices for this purpose include kurtosis, entropy, Gini index, and smoothness index, falling into a unified framework. This paper proposes statistical modeling and analysis of normalized square envelope spectrum to construct a nonparametric health index and statistical threshold for machine condition monitoring, addressing the limitations of parametric health indices.

MEASUREMENT (2021)

添加到收藏夹

Article Automation & Control Systems

Key-performance-indicator-related state monitoring based on kernel canonical correlation analysis

Qing Chen et al.

Summary: This paper proposes two improved kernel canonical correlation analysis methods to address key performance indicators (KPI)-related issues, and conducts fault detectability analysis and computational complexity analysis. The efficacy of these methods is verified using the Tennessee Eastman (TE) process.

CONTROL ENGINEERING PRACTICE (2021)

添加到收藏夹

Article Engineering, Multidisciplinary

Research on a rolling bearing health monitoring algorithm oriented to industrial big data

Wenchang Zhu et al.

Summary: The RMT-PCA method utilizes rolling bearing health monitoring data to build a random matrix model, extracting 14 feature indices and constructing a fusion feature index through PCA algorithm to assess bearing degradation. Experimental results demonstrate that the RMT-PCA algorithm is more sensitive to early anomalies and accurately reflects the actual degradation process of the bearing.

MEASUREMENT (2021)

添加到收藏夹

Article Engineering, Multidisciplinary

A chaotic system-based signal identification Technology: Fault-diagnosis of industrial bearing system

Shih-Yu Li et al.

Summary: A smart fault-diagnosis technology is developed using data-mapping strategy and K-Nearest Neighbor (K-NN) algorithm, which achieves high accuracy rates in identifying different fault states of industrial rotating machinery and rolling-bearing systems. Experimental results demonstrate an overall accuracy rate of around 99.16% for the proposed smart detection system.

MEASUREMENT (2021)

添加到收藏夹

Article Engineering, Industrial

A framework to automate fault detection and diagnosis based on moving window principal component analysis and Bayesian network

Arthur Henrique de Andrade Melani et al.

Summary: Through Condition-Based Maintenance strategy, a hybrid framework based on Moving Window Principal Component Analysis (MWPCA) and Bayesian Network (BN) was proposed for automated Fault Detection and Diagnosis (FDD) in machinery. The framework was able to detect and diagnose several simulated failures in a simplified model of a hydrogenerator.

RELIABILITY ENGINEERING & SYSTEM SAFETY (2021)

添加到收藏夹

Article Engineering, Multidisciplinary

A generalized health indicator for performance degradation assessment of rolling element bearings based on graph spectrum reconstruction and spectrum characterization

Xin Wang et al.

Summary: This paper proposes a novel method for performance degradation assessment of rolling element bearings, which establishes health indicators with early warning capability and monotone trend through signal feature enhancement and frequency domain feature extraction, effectively mapping the relationship between degradation and health indicators.

MEASUREMENT (2021)

添加到收藏夹

Article Engineering, Mechanical

Investigations on quasi-arithmetic means for machine condition monitoring

Bingchang Hou et al.

Summary: This paper investigates the use of quasi-arithmetic means (QAMs) and sparsity measures for constructing a health index for machine condition monitoring. Experimental results demonstrate that the proposed framework can detect faults, exhibit degradation tendencies, and be robust to impulsive noises, outperforming existing sparsity measures.

MECHANICAL SYSTEMS AND SIGNAL PROCESSING (2021)

添加到收藏夹

Article Engineering, Multidisciplinary

Deep transient feature learning for weak vibration signal detection

Xiaomeng Li et al.

Summary: In this paper, a novel method based on deep transient feature learning is proposed to extract repetitive vibration transients and suppress in-band noise simultaneously. By constructing a deep model with simulated template signals and mapping noisy signal's time-frequency distribution (TFD) image to pure repetitive transients' TFD image, clean repetitive transients can be reconstructed effectively. Experimental validation results show the proposed method is reliable and effective for repetitive transients extracting.

MEASUREMENT (2021)

添加到收藏夹

Article Engineering, Mechanical

Improved phase space warping method for degradation tracking of rotating machinery under variable working conditions

Peng Luo et al.

Summary: Health Indicators (HI) are crucial for the process of Prognostic and Health Management (PHM), and the IPSW method, by considering comprehensive mutual information and introducing correlation analysis, can effectively extract HI that characterize the damage degradation trend.

MECHANICAL SYSTEMS AND SIGNAL PROCESSING (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic