Research on Intelligent Fault Diagnosis of Rolling Bearing Based on Improved ShufflenetV2-LSTM

Article Engineering, Multidisciplinary

A novel fault diagnostic system for rolling element bearings using deep transfer learning on bispectrum contour maps

Chhaya Grover et al.

Summary: This study demonstrated that transfer learning with convolutional neural networks achieves highly accurate and reliable results in fault diagnosis of rolling element bearings, comparable to state-of-the-art techniques.

ENGINEERING SCIENCE AND TECHNOLOGY-AN INTERNATIONAL JOURNAL-JESTECH (2022)

Add to Collection

Article Automation & Control Systems

Modified Deep Autoencoder Driven by Multisource Parameters for Fault Transfer Prognosis of Aeroengine

Zhiyi He et al.

Summary: This article proposes a modified deep autoencoder method driven by multi-source parameters for fault prognosis of aeroengines. The method utilizes a fused health index to characterize performance degradation and establishes accurate mapping hidden in the health index using adaptive Morlet wavelet. Parameter transfer learning is used to enable the model to have cross-domain fault prognosis capability.

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS (2022)

Add to Collection

Article Engineering, Multidisciplinary

Multiscale inverted residual convolutional neural network for intelligent diagnosis of bearings under variable load condition

Wenlei Zhao et al.

Summary: This article proposes a new method for fault diagnosis of variable load bearings based on a multiscale inverted residual convolutional neural network (MIRCNN). The method first reconstructs the vibration signal using a semi tensor product compressed sensing (CS) method to overcome data storage and transmission speed issues. Then, a convolutional neural network (CNN) is used to process high-dimensional signals, and a multiscale algorithm is applied to the CNN architecture. Experimental results demonstrate the effectiveness and accuracy of the proposed method in improving the ability to identify faults in variable load bearings.

MEASUREMENT (2022)

Add to Collection

Article Engineering, Mechanical

Bearing fault diagnosis method based on adaptive maximum cyclostationarity blind deconvolution

Zhijian Wang et al.

Summary: In this study, a cyclic frequency set estimation method based on autocorrelation function of morphological envelope is proposed to address the issue of determining cyclic frequency in CYCBD. Additionally, a performance efficiency ratio index is introduced to balance the performance and time cost of CYCBD, followed by the use of an equal-step search strategy for adaptively selecting the filter length. The effectiveness of the proposed method is verified through simulation and experiment.

MECHANICAL SYSTEMS AND SIGNAL PROCESSING (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

A novel fault diagnosis method based on CNN and LSTM and its application in fault diagnosis for complex systems

Ting Huang et al.

Summary: This study proposes a novel fault diagnosis method that combines feature extraction and consideration of time delay of fault occurrence, utilizing sliding window processing and a CNN-LSTM model. The method integrates feature and time delay information from obtained samples, significantly improving predictive accuracy and noise sensitivity in fault diagnosis of the Tennessee Eastman chemical process, showcasing superiority over five existing fault diagnosis methods.

ARTIFICIAL INTELLIGENCE REVIEW (2022)

Add to Collection

Article Automation & Control Systems

Subdomain Adaptation Transfer Learning Network for Fault Diagnosis of Roller Bearings

Zhijian Wang et al.

Summary: Due to data distribution discrepancy, fault diagnosis models trained in one scene may fail in classifying data from other scenes. To address this issue, we propose a new model called SATLN, which combines subdomain adaptation and domain adaptation techniques to reduce both marginal and conditional distribution biases, improving classification performance and generalization.

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS (2022)

Add to Collection

Article Automation & Control Systems

A Fusion CWSMM-Based Framework for Rotating Machinery Fault Diagnosis Under Strong Interference and Imbalanced Case

Xin Li et al.

Summary: This paper proposes a fusion framework based on confidence weight support matrix machine (CWSMM) for fault diagnosis of rotating machinery. The framework improves the robustness of fault diagnosis by addressing the class imbalance problem and utilizing the prior knowledge strategy. Experiment results demonstrate promising fault diagnosis performance of the proposed method, specifically under strong interference and imbalanced datasets.

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

Distributed optimization for degenerate loss functions arising from over-parameterization

Chi Zhang et al.

Summary: This paper investigates distributed optimization with degenerate loss functions, proving convergence of distributed gradient descent algorithm in this case, analyzing the trade-off between convergence rate and communication-computation, and validating theoretical findings through experiments in distributed convex optimization and deep learning.

ARTIFICIAL INTELLIGENCE (2021)

Add to Collection

Article Chemistry, Analytical

Intelligent Fault Diagnosis of Hydraulic Piston Pump Based on Wavelet Analysis and Improved AlexNet

Yong Zhu et al.

Summary: An intelligent fault diagnosis method for hydraulic piston pump was proposed, combining wavelet analysis with improved convolutional neural network to extract fault features from vibration signals and achieve intelligent recognition. The method simplifies the network structure compared to classic AlexNet and demonstrates better performance in fault recognition accuracy and stability.

SENSORS (2021)

Add to Collection

Article Engineering, Multidisciplinary

A new fault diagnosis method based on adaptive spectrum mode extraction

Zhijian Wang et al.

Summary: The proposed adaptive spectrum mode extraction method includes spectral segmentation, mode extraction, and feedback adjustment. By utilizing fault feature mapping, spectral aggregation factor, and transboundary criterion, fault characteristic frequencies are effectively extracted, showing advantages for fault feature extraction of rolling bearings.

STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL (2021)

Add to Collection

Article Engineering, Environmental

Data-driven fault diagnosis method based on the conversion of erosion operation signals into images and convolutional neural network

Zhijian Wang et al.

Summary: This study introduces a new fault diagnosis method based on data-driven approach using AlexNet CNN. The method achieves high prediction accuracy for fault classification in different bearing datasets and has been proven feasible in engineering practice.

PROCESS SAFETY AND ENVIRONMENTAL PROTECTION (2021)

Add to Collection

Article Multidisciplinary Sciences

Incipient Fault Diagnosis of Batch Process Based on Deep Time Series Feature Extraction

Min Zhang et al.

Summary: The paper proposes an intelligent incipient fault diagnosis model based on deep time series feature extraction network, which integrates DAE, LD-LSTM, and SAE to obtain efficient features. Experimental results demonstrate that the proposed method can significantly improve accuracy in batch industries.

ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING (2021)

Add to Collection

Article Engineering, Multidisciplinary