Intelligent Mechanical Fault Diagnosis Using Multisensor Fusion and Convolution Neural Network

Xie, Tingli; Huang, Xufeng; Choi, Seung-Kyum

doi:10.1109/tii.2021.3102017

Cited by 124 publications

(34 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After the model training at different conditions is completed, a five-fold cross-validation method is adopted to obtain the classification accuracy of the trained model. The classification accuracy [ 37 ] is defined as

where, TP , TN , FP , and FN represent the number of true positives, true negatives, false positives, and false negatives, respectively. The results of the confusion matrix and the classification accuracy are shown in Figure 17 and Table 6 , respectively.…”

Section: Fault Diagnosis Of Bearing Abnormal Wearmentioning

confidence: 99%

Fault Diagnosis for Abnormal Wear of Rolling Element Bearing Fusing Oil Debris Monitoring

Zhao

Wang

Han

et al. 2023

Sensors

View full text Add to dashboard Cite

The abnormal wear of a rolling element bearing caused by early failures, such as pitting and spalling, will deteriorate the running state and reduce the life. This paper demonstrates the importance of oil debris monitoring and its effective feature extraction for bearing health assessment. In this paper, a rolling bearing-rotor test rig with forced lubrication is set up and the nonferrous contaminants with higher hardness were introduced artificially to accelerate the occurrence of pitting and spalling. The early failure and abnormal wear of rolling bearings cannot be effectively detected only through the vibration signal; the temperature and oil debris monitoring data are also collected synchronously. Two features regarding the ferrous particle size distribution are extracted and fused with vibration based-features to form a feature set. The sensitive features are extracted from the features set using the Neighborhood Component Analysis method to avoid feature redundancy. Finally, the importance of the oil debris based-features for the diagnosis of abnormal bearing wear is analyzed with different machine learning algorithms. Taking SVM classifier as an example, the experiment results show that the introduction of oil debris based-features increases the diagnostic accuracy by 15.7%.

show abstract

Section: Fault Diagnosis Of Bearing Abnormal Wearmentioning

confidence: 99%

Fault Diagnosis for Abnormal Wear of Rolling Element Bearing Fusing Oil Debris Monitoring

Zhao

Wang

Han

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…For processing sensor data for real-time diagnostics redundancy, numerous parameters have to be analyzed [ 38 ]. To develop predictive models for machines, data collection of the mechanical health condition is needed [ 73 , 74 ]. In AM, sensors are attached to measure and detect thermal, acoustic, optical, and ultrasonic signals, which return valuable information [ 5 ].…”

Section: Additive Manufacturing and Data-driven Preparationmentioning

confidence: 99%

Exploring Machine Learning-Based Fault Monitoring for Polymer-Based Additive Manufacturing: Challenges and Opportunities

Sampedro

Rachmawati

Kim

et al. 2022

Sensors

View full text Add to dashboard Cite

Three-dimensional printing, often known as additive manufacturing (AM), is a groundbreaking technique that enables rapid prototyping. Monitoring AM delivers benefits, as monitoring print quality can prevent waste and excess material costs. Machine learning is often applied to automating fault detection processes, especially in AM. This paper explores recent research on machine learning-based mechanical fault monitoring systems in fused deposition modeling (FDM). Specifically, various machine learning-based algorithms are applied to measurements extracted from different parts of a 3D printer to diagnose and identify faults. The studies often use mechanical-based fault analysis from data gathered from sensors that measure attitude, acoustic emission, acceleration, and vibration signals. This survey examines what has been achieved and opens up new opportunities for further research in underexplored areas such as SLM-based mechanical fault monitoring.

show abstract

“…1) Pre-processing methods: Fig. 5 shows the comparison results across four data pre-processing methods: (a) convert the raw signal into two-dimensional time-frequency domain signal through Short-time Fourier transform (STFT); (b) reshape the frequency domain signal processed by fast Fourier transform into a 32x64 matrix (denoted as RI), and a similar trick can be found [38]; (c) reshape the frequency domain signal into a 64x32 matrix (denoted as RII); and (d) normalize the twodimensional matrix generated by RI (denoted as Norm). It can be seen that STFT is inferior to the other three methods.…”

Section: Sensitivity Analysismentioning

confidence: 99%

Class-Imbalance Privacy-Preserving Federated Learning for Decentralized Fault Diagnosis With Biometric Authentication

Lü

Gao

et al. 2022

IEEE Trans. Ind. Inf.

View full text Add to dashboard Cite

Privacy protection as a major concern of the industrial big data enabling entities, makes the massive safetycritical operation data of wind turbine, unable to exert its great value because of the threat of privacy leakage. How to improve the diagnostic accuracy of decentralized machines without data transfer remains an open issue, especially these machines are almost accompanied by skewed class distribution in the real industries. In this study, a class-imbalance privacy-preserving federated learning framework for fault diagnosis of decentralized wind turbine is proposed. Specifically, a biometric authentication technique is first employed to ensure that only legitimate entities can access private data and defend against malicious attacks. Then, the federated learning with two privacy-enhancing techniques enables high potential privacy and security in low-trust systems. After that, a solely gradient based self-monitor scheme is integrated to acknowledge the global imbalance information for class-imbalanced fault diagnosis. We leverage a real-world industrial wind turbine dataset to verify the effectiveness of the proposed framework. By comparison with five state-of-theart approaches and two non-parametric tests, the superiority of the proposed framework in imbalanced classification is ascertained. An ablation study indicates the proposed framework can maintain high diagnostic performance while enhancing privacy protection.

show abstract

Intelligent Mechanical Fault Diagnosis Using Multisensor Fusion and Convolution Neural Network

Cited by 124 publications

References 33 publications

Fault Diagnosis for Abnormal Wear of Rolling Element Bearing Fusing Oil Debris Monitoring

Fault Diagnosis for Abnormal Wear of Rolling Element Bearing Fusing Oil Debris Monitoring

Exploring Machine Learning-Based Fault Monitoring for Polymer-Based Additive Manufacturing: Challenges and Opportunities

Class-Imbalance Privacy-Preserving Federated Learning for Decentralized Fault Diagnosis With Biometric Authentication

Contact Info

Product

Resources

About