Experimental investigation of cylindrical bearing fault diagnosis with SVM

Parmar, Umang; Pandya, D. H.

doi:10.1016/j.matpr.2020.11.327

Cited by 17 publications

(11 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is one of the common kernel learning methods. More and more scholars are using SVM for fatigue damage detection in bearings [19][20][21][22][23][24]. It is proved by the above scholars that the SVM algorithm can have an exciting effect on the detection of fatigue damage in bearings.…”

Section: Introductionmentioning

confidence: 99%

Data-Driven Fatigue Damage Monitoring and Early Warning Model for Bearings

Deng

2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

Since the manual extraction of features is not sufficient to accurately characterize the health status of rolling bearings, machine learning algorithms are gradually being used for fault diagnosis of bearings, which can adaptively learn the required features from the input data. In this paper, k -nearest neighbor, support vector machines, and convolutional neural networks are successfully applied to the fault diagnosis of bearings, for the benefit of achieving the detection and early warning of bearing fatigue damage. The original samples are segmented into semioverlapping samples. When using k -nearest neighbor and support vector machines as early warning models, we searched their hyperparameters with random search and grid search, and the results showed that support vector machines could achieve 87.1% of bearing detection accuracy and k -nearest neighbor could achieve 100% of detection accuracy. When convolutional neural networks are used as the early warning model, the accuracy can reach 99.75%.

show abstract

Section: Introductionmentioning

confidence: 99%

Data-Driven Fatigue Damage Monitoring and Early Warning Model for Bearings

Deng

2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

show abstract

“…Final accuracy of the top ten ranked features in ANN is 97.6 %, for SVM it is 99.6 % and MLR shows 96% accuracy. Comparison of the previous research (Umang Parmar. 2021) for the same set of data without ranking feature method shows that ANN is 96% accurate in the training but not so good in the validation and testing while SVM shows the 95.6% efficiency which proves that ranking feature method helps in getting the good results.…”

Section: Discussionmentioning

confidence: 85%

Comparison of the supervised machine learning techniques using WPT for the fault diagnosis of cylindrical roller bearing

Parmar¹,

Pandya²

2021

Int. J. Eng. Sci. Tech

Self Cite

View full text Add to dashboard Cite

In this paper, a comparative study of the Artificial Intelligence (AI) techniques for the condition monitoring of the cylindrical roller bearing is presented. For the feature selection, wavelet analysis is applied using the ‘sym2’ as the mother wavelet. Nine features are considered for the training and evaluation of the AI techniques and then effectiveness is compared. Bearing sample data consists of four different conditions as having defective inner race, defective outer race, having defects on roller and a healthy bearing. For the preparation of the sample bearing, laser machine is used for introduction of the micro size defects on the surfaces. Support Vector Machine (SVM), Artificial neural network (ANN), and logistic regression are used with feature ranking method for the data training purpose and their effectiveness of identifying the condition is the major purpose. Feature ranking method is the new way of filtering the right data in right sequence for the data training. In results, Logistic regression found more accurate in comparison with the ANN and SVM for the cylindrical bearing.

show abstract

“…Based on feature extraction and consequent pattern recognition, machine-learning approaches have shown potential for bearing fault diagnosis due to their self-learning optimization. Examples of such approaches are the support vector machine (SVM) [2], Bayesian networks [3] and deep-learning networks [4]. In particular, deep-learning networks, which employ multi-layered deep convolutional networks to extract semantic features, have been proven experimentally to be an effective method for fault diagnosis [5][6][7][8].…”

Section: Introductionmentioning

confidence: 99%

Multi-scale deep residual shrinkage networks with a hybrid attention mechanism for rolling bearing fault diagnosis

Zhang

Wei

Huang

et al. 2023

Preprint

View full text Add to dashboard Cite

The fault diagnosis of rolling bearings based on deep networks is hindered by the unexpected noise involved with accessible vibration signals and global information abatement in deepened networks. To combat the degradation, a multi-scale deep residual shrinkage network with a hybrid-attention-mechanism (MH-DRSN) is proposed in this paper. First, a spatial domain attention mechanism is introduced into the residual shrinkage module to represent the distance dependence of the feature maps. Then, a hybrid attention mechanism considering both the inner-channeled and cross-channeled characteristics is constructed. Through the comprehensive evaluation of the feature map, it provides a soft threshold for the activation function and realizes the feature-map selection adaptively. Second, the dilated convolution with different dilation rates is implemented for multi-scale context information extraction. Through the feature combination of the DRSN and the dilated convolution, the global information of the rolling bearing fault is strengthened and preserved as the fault diagnosis network is deepened. Finally, the performance of the proposed fault-diagnosis model is validated on datasets from Case Western Reserve University (CWRU), Xi’an Jiaotong University and Zhejiang Changxing Sumyoung Technology Co. Ltd (XJTU-SY). Also, the influence of the number of residual shrinkage layers, model optimizers, and different learning rates on the accuracy of the diagnostic model has been discussed. The experimental results show that, compared with common convolution neural networks, the proposed neural diagnosis model provides a higher identification accuracy and better robustness under noise interference.

show abstract

Experimental investigation of cylindrical bearing fault diagnosis with SVM

Cited by 17 publications

References 11 publications

Data-Driven Fatigue Damage Monitoring and Early Warning Model for Bearings

Data-Driven Fatigue Damage Monitoring and Early Warning Model for Bearings

Comparison of the supervised machine learning techniques using WPT for the fault diagnosis of cylindrical roller bearing

Multi-scale deep residual shrinkage networks with a hybrid attention mechanism for rolling bearing fault diagnosis

Contact Info

Product

Resources

About