Related references
Note: Only part of the references are listed.
Article
Automation & Control Systems
Quan Qian et al.
Summary: In this article, a new ensemble weighting subdomain adaptation network (EWSAN) diagnostic model is proposed to improve the degree of domain confusion. The model utilizes an enhanced joint distribution alignment (EJDA) mechanism with a multiscale top classifier and ensemble voting to obtain reliable pseudolabels. An ensemble weighting maximum mean discrepancy is constructed to enhance fine-grained domain confusion. The effectiveness and superiority of the EWSAN model are validated through multiple experiments.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2023)
Article
Engineering, Mechanical
Jianyu Long et al.
Summary: This article proposes a fault diagnosis method for industrial robots based on an attitude sensor and a multiscale convolutional capsule network (MCCN). By monitoring the attitude of transmission components, fault features are learned from attitude data, and effective fault diagnosis is achieved by fusing multiscale features and spatial-relational features.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Computer Science, Interdisciplinary Applications
Min Huang et al.
Summary: A deep multi-source transfer learning model is proposed in this paper to address the issues of large data discrepancy in source domain and mismatched feature distribution, using MMD to select suitable source domain data and independent domain-specific feature extractors for feature extraction.
SIMULATION MODELLING PRACTICE AND THEORY
(2023)
Article
Automation & Control Systems
Chao Zhao et al.
Summary: This article proposes an adversarial mutual information-guided single domain generalization network for machinery fault diagnosis, which learns domain-invariant representations to address domain shift problems. A domain generation module is designed to generate fake target domains with significant distribution discrepancies, and an iterative min-max game of mutual information is implemented to learn generalized features for resisting unknown domain shift. Extensive diagnosis experiments on two mechanical rigs validated the effectiveness of the proposed method.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Engineering, Mechanical
Quan Qian et al.
Summary: An improved joint distribution adaptation (IJDA) mechanism is proposed to enhance the distribution alignment and match the marginal distributions as well as conditional distributions of two domains. It combines maximum mean discrepancy and CORrelation Alignment (CORAL) to enhance domain confusion and constructs an improved conditional distribution alignment mechanism. In addition, a new I-Softmax loss is introduced to contribute to feature learning and learn more separable features. Experimental results on six cross-machine diagnostic tasks demonstrate that the proposed DDTLN achieves higher performance in transfer fault diagnosis compared to other typical domain adaptation methods.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2023)
Review
Engineering, Mechanical
Yongjian Sun et al.
Summary: This paper introduces the characteristics and challenges of fault diagnosis in building materials machinery equipment, discusses the research status and development directions.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Engineering, Mechanical
Chao Zhao et al.
Summary: This study proposes a semi-supervised domain generalization fault diagnosis (Sem-iDGFD) method, which assigns reliable pseudo labels to unlabeled data with knowledge assistance from labeled data. An entropy-based sample purification mechanism is designed to improve the quality of the pseudo-labeled samples. Experimental results demonstrate that the proposed method achieves higher precision than other common SemiDGFD methods and comparable performance with up-to-date fully-labeled DGFD methods.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Engineering, Industrial
Shaowei Liu et al.
Summary: This paper proposes a deep multi-source adversarial discrepancy matching adaptation network (MADMAN) to enhance the accuracy of cross-domain intelligent diagnosis. The proposed method utilizes the generalization knowledge learned from multiple domains to diagnose unknown tasks and adaptively adjusts the weight factors of multiple source domains using a self-attention mechanism. It also applies discrepancy matching technique to dynamically align the feature distributions of different domains and incorporates an adversarial classifier training method to improve transferability by considering task-specific decision boundaries. Extensive experiments using two bearing datasets demonstrate the superiority of the proposed approach compared to advanced methods.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Automation & Control Systems
Jinchuan Qian et al.
Summary: A valid fault pattern clustering method is proposed in this paper to assist offline fault diagnosis and provide data support for training the fault diagnosis model. The novel sequence discriminative feature extraction network (SDFEN) is developed to extract discriminative features from industrial time series. The proposed method is verified to be feasible and effective through experiments on benchmark processes and flow facilities.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Engineering, Multidisciplinary
DeCai Li et al.
Summary: This study introduced a revised DL model, DC-CN, for diagnosing rotating machinery faults under minor sample conditions, which achieved a good balance between diagnostic effectiveness and data amount. Experimental results demonstrated that DC-CN outperformed other DL models in diagnosing different types of rotating machinery.
Review
Engineering, Multidisciplinary
Zheng Yang et al.
Summary: With the increase in size and complexity of mechanical equipment, traditional intelligent fault diagnosis based on shallow machine learning methods is insufficient for coupling faults. The development of deep learning, particularly the use of Autoencoder-based representation learning, has provided new opportunities for intelligent fault diagnosis. This article introduces the theoretical foundations and training methods of multi-type Autoencoders and reviews the advancements in their applications, aiming to improve representation learning. Two case studies are presented to demonstrate the application of Autoencoder-based methods on ideal and complex engineering systems. The challenges and prospects of Autoencoder-based representation learning are also discussed, offering guidance for future research directions.
Article
Engineering, Multidisciplinary
Ronny Francis Ribeiro Junior et al.
Summary: This study proposes a method for multi-sensor fault detection using a multi-head 1D Convolution Neural Network. The experimental results demonstrate the effectiveness and accuracy of the proposed method in real applications.
Article
Engineering, Mechanical
Shaowei Liu et al.
Summary: A novel data synthesis method called deep feature enhanced generative adversarial network is proposed in this paper to improve the performance of imbalanced fault diagnosis. By integrating a pull-away function, a self-attention module, and an automatic data filter, the quality of synthesized data is improved, the stability of generative adversarial networks is enhanced, and the accuracy and diversity of synthesized samples are timely ensured.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Review
Computer Science, Artificial Intelligence
Dilyara Baymurzina et al.
Summary: This paper reviews the recent works on Neural Architecture Search (NAS) and highlights several crucial concepts and problems in this field.
Article
Green & Sustainable Science & Technology
Reihane Rahimilarki et al.
Summary: Fault detection and classification are crucial techniques in industrial monitoring, and deep learning approaches using convolutional neural networks have shown promise in solving the problem of fault detection in wind turbines. This article presents a novel deep learning method based on time-series analysis and CNN for fault detection and classification in wind turbines, and the simulation results demonstrate the accuracy and effectiveness of the proposed method.
Article
Acoustics
Jiahui Tang et al.
Summary: This paper proposes an intelligent fault diagnosis method based on Bi-directional deep belief network and Quantum genetic algorithm for fault diagnosis of rolling bearings. The method combines forward training and reverse generation to effectively improve the diagnosis accuracy and reduces the similarity between synthesized samples and original samples using a noise time-shift layer.
Article
Engineering, Mechanical
Tianfu Li et al.
Summary: Deep learning methods have advanced the field of Prognostics and Health Management, but handling irregular data in non-Euclidean space remains a challenge. Research has proposed a practical guideline for utilizing graph neural networks for intelligent fault diagnostics and prognostics, and established a framework based on GNN for this purpose.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Engineering, Mechanical
Debasish Jana et al.
Summary: This research introduces a novel deep learning framework for identifying faults in sensor data, locating the faulty sensors, and reconstructing the correct sensor data. The framework performs well in both single and multiple sensor faults and demonstrates high computational efficiency.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Engineering, Mechanical
Yuejian Chen et al.
Summary: In this study, a physics-informed hyperparameters selection strategy is proposed for LSTM modeling and fault detection of gearboxes. The results demonstrate that this strategy can better detect gearbox faults.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Engineering, Mechanical
Weihua Li et al.
Summary: Deep Transfer Learning combines the advantages of Deep Learning in feature representation and Transfer Learning in knowledge transfer, making DL-based fault diagnosis methods more reliable and robust. However, further research is needed to explore the potential of DTL-based approaches in Intelligent Fault Diagnosis.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Engineering, Mechanical
Kai Zhou et al.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Engineering, Mechanical
Chao Zhao et al.
Summary: This paper proposes a novel domain generalization network for fault diagnosis under unknown working conditions, which can exploit domain invariance and retain domain specificity simultaneously. It effectively tackles the problem of target data inaccessibility in real-time cross-domain fault diagnosis.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Computer Science, Interdisciplinary Applications
Xue-yang Zhang et al.
Summary: This paper proposes a transfer sparse auto-encoder (SAE) based on local maximum mean difference (LMMD) and K-means to solve the problems with existing feature-based transfer learning methods. Experimental results demonstrate its superior performance in transfer fault diagnosis compared to other methods.
COMPUTERS & INDUSTRIAL ENGINEERING
(2022)
Article
Engineering, Mechanical
V. Sinitsin et al.
Summary: This paper proposes a novel hybrid CNN-MLP model-based diagnostic method for rolling bearing diagnostics. The method successfully detects and localizes bearing defects using acceleration data from a wireless acceleration sensor, achieving up to 99.6% accuracy.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Engineering, Industrial
Jingkang Liang et al.
Summary: A lightweight network with modified tree-structured parzen estimators (LN-MT) is proposed for intelligent fault diagnosis of rotating machinery in this paper. By introducing a lightweight framework based on global average pooling and group convolution, and utilizing tree-structured parzen estimator for hyperparameter optimization based on Bayesian optimization, models that balance both time and accuracy are found. Comparison experiments show that LN-MT achieves superior fault diagnosis accuracies with few trainable parameters and less calculating time.
IET COLLABORATIVE INTELLIGENT MANUFACTURING
(2022)
Article
Computer Science, Information Systems
Nikita Klyuchnikov et al.
Summary: Neural Architecture Search (NAS) is a promising and rapidly evolving research area. Limited computational resources make training a large number of neural networks difficult, but benchmarks with precomputed results have been introduced to address this issue. However, these benchmarks are only applicable to computer vision, while this work explores NAS in the field of natural language processing.
Article
Engineering, Mechanical
Qi Li et al.
Summary: The study proposes a knowledge mapping-based adversarial domain adaptation (KMADA) method, utilizing a discriminator and feature extractor to learn knowledge from the target domain. KMADA fully utilizes parameters obtained from supervised pre-training to accelerate the adversarial training process, achieving the highest diagnosis accuracy compared to other TL methods.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2021)
Article
Chemistry, Analytical
Xudong Li et al.
Summary: The Frequency-domain Fusing Convolutional Neural Network (FFCNN) improves diagnosis accuracy in fault diagnosis by utilizing convolution operations to filter signals at different frequency bands and combine them as new input signals. FFCNN contains a frequency-domain fusing layer and a feature extractor, providing an effective solution for handling the inconsistent distribution of data in domain adaptation.
Article
Engineering, Multidisciplinary
Shengkang Yang et al.
Summary: The study proposed a multi-source ensemble domain adaptation method to address domain shift issues in rotary machinery fault diagnosis. By constructing anchor adapters for multiple sources and target domains, establishing a multi-source ensemble domain adaptation model, and integrating high-performing classifiers through ensemble of anchor adapters, significant diagnosis performance and robustness were achieved.
Article
Engineering, Multidisciplinary
Jin Si et al.
Summary: This paper proposes an unsupervised deep transfer network with moment matching for fault diagnosis under different working conditions. Two adaptive methods are employed to reduce distribution discrepancy, and the results show the competitiveness of this method in various fault scenarios.
Article
Engineering, Multidisciplinary
Yahui Zhang et al.
Summary: A novel method based on recurrent neural networks is proposed for fault type identification in rotating machinery, utilizing one-dimensional time-series vibration signals converted into two-dimensional images, with the introduction of Gated Recurrent Unit (GRU) and multilayer perceptron (MLP) to achieve the best performance and robustness against noise compared to existing work.
Article
Engineering, Mechanical
Wentao Mao et al.
Summary: In recent years, deep learning techniques have shown promising prospects in bearing fault diagnosis, but introducing discriminant information about different fault types into the model remains a challenge. A new deep auto-encoder method is proposed to address this issue, utilizing a new loss function with structural discriminant information and a gradient descent method for optimization, leading to improved diagnostic accuracy and stability.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2021)
Article
Computer Science, Artificial Intelligence
Xudong Li et al.
Summary: In recent years, deep learning based bearing fault diagnosis has been rapidly developing due to the increasing amount of industrial data. However, the challenges of obtaining labeled data and dealing with the distribution differences in different production environments limit the application of deep learning. Our proposed method utilizes CMD to reduce distribution discrepancies in raw vibration signals from two different domains and successfully diagnose faults in unlabeled data under various working conditions.
Article
Computer Science, Artificial Intelligence
Guangxing Niu et al.
Summary: "The paper proposes a deep learning-based approach for rolling element bearing fault diagnosis to address challenges in existing methods including difficulty in structure decision-making, low accuracy, and learning efficiency. The proposed method integrates principal component analysis, adaptive deep belief network, and particle swarm optimization, achieving high accuracy and convergence rate. Experimental results demonstrate the effectiveness and accuracy of the proposed approach."
Article
Computer Science, Interdisciplinary Applications
Yafei Deng et al.
Summary: Recently, deep transfer learning approaches have been widely developed for mechanical fault diagnosis issue. DA-GAN model shows great superiority in dealing with mechanical partial transfer problem in both TIM and TDM.
COMPUTERS IN INDUSTRY
(2021)
Article
Automation & Control Systems
Jun Zhu et al.
Summary: The fault diagnosis based on data-driven methods is widely researched when supervised samples of the target machine are available, but labeled samples in practical machines are usually scarce. A new transfer learning approach based on multisource domain adaptation is proposed to address this issue, enabling learning from multiple domains for more general diagnosis knowledge that benefits prediction for the target domain.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Engineering, Multidisciplinary
Quan Qian et al.
Summary: The paper introduces a new deep transfer learning network, CAE-DTLN, which incorporates feature extraction, CORAL loss, and domain classification to achieve high diagnostic accuracy and anti-noise performance in mechanical fault diagnosis without labeled data.
Article
Engineering, Mechanical
Dongdong Wei et al.
Summary: This study focuses on intelligent fault diagnosis of machines in the context of changing working conditions. A multiple source domain adaptation method is proposed to learn fault-discriminative but working condition-invariant features to address the data distribution shift issue.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2021)
Article
Engineering, Mechanical
Kaiyu Zhang et al.
Summary: A novel differentiable neural architecture search method is proposed to automate the process of building neural networks and save designing time by gradually reducing candidate operations while retaining trained parameters during pruning; specially designed penalty terms are introduced to search optimal numbers of layers and nodes, reducing complexity of subnetworks and saving calculation time for signal analysis.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2021)
Article
Automation & Control Systems
Huailiang Zheng et al.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2020)
Article
Automation & Control Systems
Xiang Li et al.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2020)
Article
Engineering, Multidisciplinary
Ruixin Wang et al.
Article
Computer Science, Artificial Intelligence
Cheng Cheng et al.
Review
Engineering, Mechanical
Wade A. Smith et al.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2015)
Article
Computer Science, Artificial Intelligence
Sinno Jialin Pan et al.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2010)
