Related references
Note: Only part of the references are listed.
Article
Computer Science, Artificial Intelligence
Chuanjiang Li et al.
Summary: Deep learning-based fault diagnosis methods have achieved remarkable progress, but they are often coarse grained and require large amounts of data, which cannot identify the root causes of mechanical system failures at a finer granularity with limited fault data. In this study, a novel attention-based deep meta-transfer learning (ADMTL) method is proposed to address the challenges of fine-grained fault feature extraction and limited model generalization ability. The proposed method achieves excellent performance in few-shot fine-grained fault diagnosis tasks.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Yifei Ding et al.
Summary: This paper introduces a domain generalization (DG) approach for predicting the remaining useful life (RUL) of bearings under unseen operating conditions. It proposes an adversarial out-domain augmentation (AOA) framework to generate pseudo-domains and increase the diversity of available samples. The framework includes manifold and semantic regularization to ensure consistency. Experimental results validate the effectiveness and superiority of the proposed approach.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Engineering, Industrial
Junchao Guo et al.
Summary: This paper proposes a novel method for rotating machinery fault detection, which achieves multi-sensor data fusion using improved cyclic spectral covariance matrix (ICSCM) and motor current signal analysis. The proposed method adaptively acquires multi-sensor mode components and constructs ICSCM using sample entropy to preserve the interaction relationship between different sensors. The ICSCM is then incorporated into an extreme learning machine classifier for fault type identification. The proposed method has achieved satisfactory results and more reliable diagnosis accuracy than other state-of-the-art algorithms in rotating machinery fault detection.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Engineering, Multidisciplinary
Junchao Guo et al.
Summary: This paper proposes a novel AM-FM demodulation method based on LMSB for extracting fault features from gearbox signals. The method can simultaneously demodulate multi-mesh frequency bands and multi-modulation components. The effectiveness of LMSB is demonstrated through numerical simulations and experimental analysis, showing its superiority over other demodulation techniques. This research provides a new perspective for gearbox fault detection.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2023)
Article
Automation & Control Systems
Junchao Guo et al.
Summary: This article proposes a novel fault detection scheme based on cyclic morphological modulation spectrum (CMMS) and hierarchical Teager permutation entropy (HTPE) for rotating machinery. The scheme uses CMMS to analyze the measured signal and obtain CMMS slices with different frequency bands, and utilizes HTPE for improved feature selection. Experimental results show that the proposed scheme effectively obtains fault features and achieves accurate fault classification and recognition.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Computer Science, Information Systems
Yuxian Duan et al.
Summary: This article introduces an improved affinity propagation algorithm based on optimization of preference (APBOP) for automatic clustering on high-dimensional data. APBOP aims to address the challenges of feature extraction from high-dimensional data and the sensitivity of the clustering performance to preference. The proposed method utilizes dimensionality reduction and preference optimization techniques to improve the effectiveness of affinity propagation.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Jiaqi Xue et al.
Summary: Dimensionality reduction is a crucial technique in data mining, embedding high-dimensional data into a low-dimensional vector space while retaining important information. We propose a novel unsupervised dimensionality reduction model called LLEAN, which utilizes adaptive neighbor strategy and a projection matrix to achieve desirable results. The relationship between pairwise data is investigated, and the augmented Lagrangian multiplier is used for effective optimization. Experimental results demonstrate that LLEAN outperforms state-of-the-art methods on toy data and benchmark datasets.
PATTERN RECOGNITION
(2023)
Article
Automation & Control Systems
Xin Zhang et al.
Summary: This paper proposes a multi-sensor multi-head GAT model for fault diagnosis of large rotating machinery, which can dynamically fuse and mine high-level fault characteristics to improve the effectiveness of diagnosis.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Erfan Babaee Tirkolaee et al.
Summary: This study presents a sustainable-robust aggregate production planning problem that considers workforce productivity, outsourcing option, and supplier resilience. A bi-level decision-making system is designed using MADM and MODM models. The MADM section utilizes a hybrid method based on BWM and WASPAS under T2NN to investigate resilient supplier selection, while the MODM section suggests a multi-objective MILP model treated with the WGP method. The results demonstrate the efficiency of the decision-making system and the sensitivity of total weighted purchase from suppliers to conservatism levels.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Automation & Control Systems
Bin Yang et al.
Summary: This article proposes a deep targeted transfer learning (DTTL) method for fault diagnosis, which addresses the issue of data distribution shift and facilitates diagnosis knowledge transfer across related machines. The method relaxes the strict assumption that all target domain data are unlabeled by introducing labeled one-shot target domain samples called anchors. DTTL includes a domain-shared residual network, a target-domain clustering module, and a targeted adaptation module to correct the joint distribution shift. Experimental results on transfer diagnosis tasks across different bearings demonstrate that DTTL achieves higher diagnosis accuracy compared to other methods.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2023)
Article
Engineering, Industrial
Yiming Xiao et al.
Summary: To ensure researchers trust deep diagnostic models, interpretable rotating machinery fault diagnosis (RMFD) research has been developed. However, there is limited work on quantifying uncertainty in results and explaining its sources and composition. This paper proposes a Bayesian variational learning method to introduce uncertainty into the attention weights of Transformer and constructs a probabilistic Bayesian Transformer for trustworthy RMFD. By inferring prior and variational posterior distributions of attention weights, uncertainty is perceived, and an uncertainty quantification and decomposition scheme is developed to achieve confidence characterization of results and separation of epistemic and aleatoric uncertainty. The proposed method is validated in three out-of-distribution generalization scenarios.
JOURNAL OF MANUFACTURING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Zhe Wang et al.
Summary: This study proposes a novel temporal-spatial graph neural network with an attention-aware module (A-TSGNN) for multi-source information fusion, achieving exceptional performance on wind turbine and gearbox datasets.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Engineering, Industrial
Shen Yan et al.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Engineering, Industrial
Shen Yan et al.
Summary: This paper proposes a new machinery anomaly detection method called full graph dynamic autoencoder (FGDAE) for complex operating conditions. It develops a full connected graph (FCG) to obtain global structure information and constructs a graph adaptive autoencoder (GAAE) model to aggregate multi-perspective feature information between channels. The method achieves better performance compared to other popular anomaly detection methods on machinery datasets.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Engineering, Multidisciplinary
Yonghao Miao et al.
Summary: This paper proposes a period-refined maximum second-order cyclostationarity blind deconvolution (PRCYCBD) method using TSA for weak fault detection in rolling element bearings (REBs). The method accurately estimates the iterative period and is suitable for heavy noise environments. Simulation and experimental results show that the proposed method outperforms traditional minimum entropy deconvolution and traditional autocorrelation-improved cyclostationarity blind deconvolution in fault detection.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2023)
Article
Engineering, Electrical & Electronic
Shaoke Wan et al.
Summary: This study proposes a novel multisensor information coupling network (MICN) for bearing fault diagnosis, which handles the signals from the same or different types of sensors, and extracts deeper features by layer by layer fusion. A novel feature-level information coupling model is developed based on the mutual attention mechanism during the multilayer feature fusion process. Several different experiments are designed to validate the efficiency and superiority of the proposed method.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Computer Science, Artificial Intelligence
Xingqiu Li et al.
Summary: This study proposes an integrated deep multiscale feature fusion network (IDMFFN) for predicting the remaining useful life (RUL) of aeroengines using multisensor data. The network utilizes multiscale feature extraction blocks and concatenated blocks to improve prediction accuracy.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Xuegang Li et al.
Summary: In this study, a multi-perspective deep transfer learning (DTL) model called Multi-Perspective DTL (MPDTL) is proposed for enhancing the bearing fault diagnosis under varying working conditions. The MPDTL model integrates multi-perspective information such as space, channel, and sequence to extract discriminative features. It consists of a feature enhancement network (FENet) to improve the quality of characteristics, a bidirectional long-short term memory (BiLSTM) network to extract high-level discriminative features, and an optimization objective module for model updating. Experimental results demonstrate the effectiveness and superiority of the proposed method.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Zhuang Ye et al.
Summary: Bearings are critical components in rotating machinery, and it is important to timely and accurately recognize bearing defects. This paper proposes a novel CNN called Multi-Level Features Fusion Network (MLFNet) for feature learning and fault diagnosis of vibration signals. Experimental results show that MLFNet has good performance in feature extraction and outperforms typical DNNs and state-of-the-art methods in bearing fault diagnosis.
APPLIED SOFT COMPUTING
(2022)
Article
Automation & Control Systems
Tingli Xie et al.
Summary: In this article, a novel intelligent fault diagnosis method based on multisensor fusion and convolutional neural network is explored. The proposed method converts multisignal data into RGB images and uses an improved CNN for classification, resulting in higher accuracy in fault diagnosis.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Article
Engineering, Multidisciplinary
Wei Zhang et al.
Summary: Federated learning has shown promising potential in machinery fault diagnostics, but the varying data distributions across clients in real-world industrial scenarios can lead to performance deterioration. To address this, a federated transfer learning method is proposed to bridge the domain gap through prior distributions, extracting client-invariant features for diagnostics while preserving data privacy. Experiments on rotating machinery datasets validate the effectiveness and promise of the proposed method for federated transfer learning in fault diagnostic problems.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2022)
Article
Computer Science, Artificial Intelligence
Simrandeep Singh et al.
Summary: This manuscript proposes a novel approach for the fusion of infrared and visible light images. By optimizing the weight map, the feature level fusion is achieved, resulting in better fusion performance. Experimental results indicate that the proposed method works well for most image datasets and outperforms certain traditional models.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Engineering, Electrical & Electronic
Di Song et al.
Summary: A novel crack detection method is proposed in this study by fusing information from acoustic and vibration signals (AVS) at the data and decision level. The proposed method achieves improved reliability and accuracy by utilizing a one-dimensional convolutional neural network (1D CNN) and modifying preliminary results. Experimental results demonstrate the reliability and superiority of the proposed method compared to other approaches.
IEEE SENSORS JOURNAL
(2022)
Article
Engineering, Electrical & Electronic
Khandaker Noman et al.
Summary: Sample entropy (SE) is a nonlinear measure used to characterize the health status of rolling element bearings by measuring the complexity of vibration signals. However, in continuous monitoring scenarios under noisy conditions, using SE directly can lead to inefficient early fault warning and an inability to differentiate between different fault types. To address this issue, a new measure called oscillatory sample entropy (OSE) is proposed, which separates the principal component of the vibration signal that is sensitive to SE calculation using the tunable Q factor wavelet transform (TQWT). Experimental case studies have shown that OSE not only overcomes the limitations of SE but also outperforms approximate entropy (AE) and fuzzy entropy (FE) for continuous monitoring of bearing health.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Engineering, Multidisciplinary
Vishnu Prasad Muddineni et al.
Summary: This article introduces the objective function optimization in predictive torque control for induction machine using grey relational analysis (GRA). The GRA method is utilized to modify the single-objective function into two independent objective functions for stator flux and torque, identifying suitable control actions. A MATLAB/Simulink model is developed to validate the control algorithm under various operating conditions and compared with experimental results.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2021)
Article
Engineering, Mechanical
Zhuang Ye et al.
Summary: A novel DNN, DMCNet, is proposed for feature learning of gearbox vibration signals. It utilizes two parallel branches for feature mapping and noise reduction, introduces a morphological feature fusion method and recalibrated residual learning, achieving good feature learning of vibration signals.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2021)
Article
Engineering, Electrical & Electronic
Alexandre M. Pessoa et al.
Summary: This paper proposes an extended autocorrelation function model that can generate random variables with negative correlation in R(2) and R(3) for channel modeling in 5G systems. By comparing the proposed Sum-of-Sinusoids (SoS) method with state-of-the-art solutions, it is found that the proposed method achieves lower spatial-mean-square error and mean square error with lower computational complexity.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2021)
Article
Automation & Control Systems
Heng Zhang et al.
Summary: This article proposes a novel framework for RUL prediction and uncertainty management in multisensor systems, which optimizes health indicators obtained from multiple sensors, achieves the optimal weight vector of the fusion model using a multiobjective grasshopper optimization algorithm, introduces an unscented particle filter for RUL prediction, and develops probability distribution of failure threshold and noise parameter adjustment for uncertainty management in prognosis. Experimental results on aircraft turbine engine degradation demonstrate the effectiveness of the proposed framework.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Computer Science, Artificial Intelligence
Xiaoan Yan et al.
Summary: The performance of complex rotor-bearing system deteriorates with time, making it difficult to identify fault categories and severities throughout the entire life-cycle. The proposed DRVAE method utilizes deep learning to effectively improve identification accuracy and feature learning performance.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Chemistry, Physical
Yunlong Shang et al.
JOURNAL OF POWER SOURCES
(2020)
Article
Computer Science, Artificial Intelligence
Hanshu Cai et al.
INFORMATION FUSION
(2020)
Article
Computer Science, Artificial Intelligence
Xiaoan Yan et al.
KNOWLEDGE-BASED SYSTEMS
(2019)
