A novel fault diagnosis method of rotating machinery via VMD, CWT and improved CNN

Gu, Jun; Peng, Yuxing; Lu, Hao; Chang, Xiangdong; Chen, Guoan

doi:10.1016/j.measurement.2022.111635

Cited by 78 publications

(35 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The decision made by the fault diagnosis system can be taken based on data in the time domain [ 119 , 120 ] or other domains, for example, the frequency domain [ 121 , 122 , 123 , 124 ], the time–frequency domain [ 125 , 126 , 127 , 128 , 129 , 130 , 131 ] or the time-scale domain [ 132 , 133 , 134 , 135 ].…”

Section: Feature Extraction Methodsmentioning

confidence: 99%

Cloud Based Fault Diagnosis by Convolutional Neural Network as Time–Frequency RGB Image Recognition of Industrial Machine Vibration with Internet of Things Connectivity

Łuczak

Brock

Siembab

2023

Sensors

View full text Add to dashboard Cite

The human-centric and resilient European industry called Industry 5.0 requires a long lifetime of machines to reduce electronic waste. The appropriate way to handle this problem is to apply a diagnostic system capable of remotely detecting, isolating, and identifying faults. The authors present usage of HTTP/1.1 protocol for batch processing as a fault diagnosis server. Data are sent by microcontroller HTTP client in JSON format to the diagnosis server. Moreover, the MQTT protocol was used for stream (micro batch) processing from microcontroller client to two fault diagnosis clients. The first fault diagnosis MQTT client uses only frequency data for evaluation. The authors’ enhancement to standard fast Fourier transform (FFT) was their usage of sliding discrete Fourier transform (rSDFT, mSDFT, gSDFT, and oSDFT) which allows recursively updating the spectrum based on a new sample in the time domain and previous results in the frequency domain. This approach allows to reduce the computational cost. The second approach of the MQTT client for fault diagnosis uses short-time Fourier transform (STFT) to transform IMU 6 DOF sensor data into six spectrograms that are combined into an RGB image. All three-axis accelerometer and three-axis gyroscope data are used to obtain a time-frequency RGB image. The diagnosis of the machine is performed by a trained convolutional neural network suitable for RGB image recognition. Prediction result is returned as a JSON object with predicted state and probability of each state. For HTTP, the fault diagnosis result is sent in response, and for MQTT, it is send to prediction topic. Both protocols and both proposed approaches are suitable for fault diagnosis based on the mechanical vibration of the rotary machine and were tested in demonstration.

show abstract

Section: Feature Extraction Methodsmentioning

confidence: 99%

Cloud Based Fault Diagnosis by Convolutional Neural Network as Time–Frequency RGB Image Recognition of Industrial Machine Vibration with Internet of Things Connectivity

Łuczak

Brock

Siembab

2023

Sensors

View full text Add to dashboard Cite

show abstract

“…In addition, to demonstrate the advantages of the RF classifier in the newly proposed method, 9 other common classifiers, AdaBoost [25], BPNN [26], DA [27], DT [28], GNN [29], KNN [30], NB [19], SVM [32], and CNN [33], are selected for comparison. In all classification models, the parameter optimization method is the same as RF, using Bayesian optimization, trying to minimize the cross-validation error for classification algorithms by varying the parameters.…”

Section: Comparison Of Classifiersmentioning

confidence: 99%

“…e simple implementation, fast training speed, and excellent classification capabilities of RF make it stand out among many machine learning-classification methods. In addition, deep learning models without feature engineering are also developing rapidly [33,34]. However, in the aviation field, according to the guidance for vibration-based diagnostic algorithms, computational efficiency and physical description are emphasized.…”

Section: Introductionmentioning

confidence: 99%

Fractional Dimensionless Indicator with Random Forest for Bearing Fault Diagnosis under Variable Speed Conditions

Huang

Xu²,

Cao³

et al. 2022

Shock and Vibration

View full text Add to dashboard Cite

Fault diagnosis of rolling bearings under variable speed is a common issue in engineering practice, but it lacks an effective diagnosis algorithm, while approaches developed for steady speed cannot be directly applied. Therefore, for effectively identifying bearing faults under variable speed, this paper proposed a multiscale fractional dimensionless indicator (MSFDI) and put forward a fault diagnosis method with random forest (RF). It can overcome the feature space aliasing problem of traditional dimensionless indicators, which will lead to increased diagnosis uncertainty. The multiorder fractional Fourier transform is carried out on bearing signals to get a series of fractional Fourier domain components, which will be used to construct the original MSFDI feature set. Moreover, reliefF selects the sensitive MSFDIs as the input of the RF algorithm to determine the health condition. The effectiveness of the proposed method is verified by experiments and case studies.

show abstract

“…In recent years, various intelligent fault diagnosis methods combining machine learning with traditional signal processing have been proposed [3] . For example, Short Time Fourier Transform [4,5] , wavelet transform [6,7] were used to extract frequency features combined with convolutional neural network (CNN) for fault diagnosis [8] . In [9], the multiscale entropy of the signal and extreme learning machine were applied to achieve the initial looseness detection of pipe clamps.…”

Section: Introductionmentioning

confidence: 99%

Looseness extent recognition of hydraulic pipeline clamps based on multi-scale feature fusion CNN model

Wang

Wei

Yang

2023

Eighth International Symposium on Advances in Electrical, Electronics, and Computer Engineering (ISAEECE 2023)

View full text Add to dashboard Cite

The clamp, as a fixed connection element of hydraulic pipeline system, its condition affects sustainability, reliability and safety of the whole system. Therefore, it is essential to identify level of loosed clamp caused by wear and tear. This paper proposes a clamp looseness extent recognition method based on multi-scale feature fusion convolutional neural network (MFF-CNN). The 2-D input constructed by small sample data of 7 distributed FBG sensors is useful for integration of temporal-spatial feature of clamp looseness. Furthermore, multi-scale fused features in MFF-CNN are more effective to overcome problem of small sample data caused by transient nature of mechanical faults and improves efficiency of CNN model training. The proposed model provides better performance on clamp looseness extent recognition compared to 1DCNN, 2DCNN and MSCNN. The average recognition accuracy reaches up to 96.3%.

show abstract

A novel fault diagnosis method of rotating machinery via VMD, CWT and improved CNN

Cited by 78 publications

References 33 publications

Cloud Based Fault Diagnosis by Convolutional Neural Network as Time–Frequency RGB Image Recognition of Industrial Machine Vibration with Internet of Things Connectivity

Cloud Based Fault Diagnosis by Convolutional Neural Network as Time–Frequency RGB Image Recognition of Industrial Machine Vibration with Internet of Things Connectivity

Fractional Dimensionless Indicator with Random Forest for Bearing Fault Diagnosis under Variable Speed Conditions

Looseness extent recognition of hydraulic pipeline clamps based on multi-scale feature fusion CNN model

Contact Info

Product

Resources

About