Limits...
Bearing Fault Diagnosis Based on Statistical Locally Linear Embedding.

Wang X, Zheng Y, Zhao Z, Wang J - Sensors (Basel) (2015)

Bottom Line: The measured signal samples usually distribute on nonlinear low-dimensional manifolds embedded in the high-dimensional signal space, so how to implement feature extraction, dimensionality reduction and improve recognition performance is a crucial task.Finally in the feature reduction space pattern classification and fault diagnosis by classifier are carried out easily and rapidly.The results indicate that the proposed approach obviously improves the classification performance of fault pattern recognition and outperforms the other traditional approaches.

View Article: PubMed Central - PubMed

Affiliation: College of Water Conservancy and Hydropower Engineering, Hohai University, Nanjing 210098, China. wangxiang@njit.edu.cn.

ABSTRACT
Fault diagnosis is essentially a kind of pattern recognition. The measured signal samples usually distribute on nonlinear low-dimensional manifolds embedded in the high-dimensional signal space, so how to implement feature extraction, dimensionality reduction and improve recognition performance is a crucial task. In this paper a novel machinery fault diagnosis approach based on a statistical locally linear embedding (S-LLE) algorithm which is an extension of LLE by exploiting the fault class label information is proposed. The fault diagnosis approach first extracts the intrinsic manifold features from the high-dimensional feature vectors which are obtained from vibration signals that feature extraction by time-domain, frequency-domain and empirical mode decomposition (EMD), and then translates the complex mode space into a salient low-dimensional feature space by the manifold learning algorithm S-LLE, which outperforms other feature reduction methods such as PCA, LDA and LLE. Finally in the feature reduction space pattern classification and fault diagnosis by classifier are carried out easily and rapidly. Rolling bearing fault signals are used to validate the proposed fault diagnosis approach. The results indicate that the proposed approach obviously improves the classification performance of fault pattern recognition and outperforms the other traditional approaches.

No MeSH data available.


Feature dimension reduction to rolling bearing multi-domain feature in the dataset: (a) Mapping with PCA; (b) Mapping with LDA; (c) Mapping with LLE; (d) Mapping with S-LLE.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4541876&req=5

sensors-15-16225-f010: Feature dimension reduction to rolling bearing multi-domain feature in the dataset: (a) Mapping with PCA; (b) Mapping with LDA; (c) Mapping with LLE; (d) Mapping with S-LLE.

Mentions: In order to demonstrate the superiority of the presented S-LLE dimensionality reduction method, when S-LLE is carried out in the process of the training sample labeled into clusters, is set to 4 and is set to 4. An experiment was conducted on the dataset to evaluate its dimensionality reduction performance on the sample dataset and make a comparison with PCA, LDA, and LLE as the most representative dimensionality reduction approaches. The experimental results of dimensionality reduction with the four approaches are shown in Figure 10, where it can be seen that PCA, LDA and LLE have poor sample classification performance. PCA and LDA obviously have three classes of overlap and LLE obviously has two classes of overlap. Compared with them, S-LLE can obtain a more clear separation of the clustering on the mapping, so S-LLE can identify each fault accurately for all feature samples. This is due to the fact that S-LLE has a greater ability to discovery local neighbor geometry information in the data manifold by utilizing the class label information. Therefore, we can use the S-LLE algorithm to obtain the original multi-domain feature dataset and select the salient features. This added process can capture intrinsic global geometric structure embedded in the high-dimensional fault features and achieve an efficient classification for fault pattern recognition.


Bearing Fault Diagnosis Based on Statistical Locally Linear Embedding.

Wang X, Zheng Y, Zhao Z, Wang J - Sensors (Basel) (2015)

Feature dimension reduction to rolling bearing multi-domain feature in the dataset: (a) Mapping with PCA; (b) Mapping with LDA; (c) Mapping with LLE; (d) Mapping with S-LLE.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4541876&req=5

sensors-15-16225-f010: Feature dimension reduction to rolling bearing multi-domain feature in the dataset: (a) Mapping with PCA; (b) Mapping with LDA; (c) Mapping with LLE; (d) Mapping with S-LLE.
Mentions: In order to demonstrate the superiority of the presented S-LLE dimensionality reduction method, when S-LLE is carried out in the process of the training sample labeled into clusters, is set to 4 and is set to 4. An experiment was conducted on the dataset to evaluate its dimensionality reduction performance on the sample dataset and make a comparison with PCA, LDA, and LLE as the most representative dimensionality reduction approaches. The experimental results of dimensionality reduction with the four approaches are shown in Figure 10, where it can be seen that PCA, LDA and LLE have poor sample classification performance. PCA and LDA obviously have three classes of overlap and LLE obviously has two classes of overlap. Compared with them, S-LLE can obtain a more clear separation of the clustering on the mapping, so S-LLE can identify each fault accurately for all feature samples. This is due to the fact that S-LLE has a greater ability to discovery local neighbor geometry information in the data manifold by utilizing the class label information. Therefore, we can use the S-LLE algorithm to obtain the original multi-domain feature dataset and select the salient features. This added process can capture intrinsic global geometric structure embedded in the high-dimensional fault features and achieve an efficient classification for fault pattern recognition.

Bottom Line: The measured signal samples usually distribute on nonlinear low-dimensional manifolds embedded in the high-dimensional signal space, so how to implement feature extraction, dimensionality reduction and improve recognition performance is a crucial task.Finally in the feature reduction space pattern classification and fault diagnosis by classifier are carried out easily and rapidly.The results indicate that the proposed approach obviously improves the classification performance of fault pattern recognition and outperforms the other traditional approaches.

View Article: PubMed Central - PubMed

Affiliation: College of Water Conservancy and Hydropower Engineering, Hohai University, Nanjing 210098, China. wangxiang@njit.edu.cn.

ABSTRACT
Fault diagnosis is essentially a kind of pattern recognition. The measured signal samples usually distribute on nonlinear low-dimensional manifolds embedded in the high-dimensional signal space, so how to implement feature extraction, dimensionality reduction and improve recognition performance is a crucial task. In this paper a novel machinery fault diagnosis approach based on a statistical locally linear embedding (S-LLE) algorithm which is an extension of LLE by exploiting the fault class label information is proposed. The fault diagnosis approach first extracts the intrinsic manifold features from the high-dimensional feature vectors which are obtained from vibration signals that feature extraction by time-domain, frequency-domain and empirical mode decomposition (EMD), and then translates the complex mode space into a salient low-dimensional feature space by the manifold learning algorithm S-LLE, which outperforms other feature reduction methods such as PCA, LDA and LLE. Finally in the feature reduction space pattern classification and fault diagnosis by classifier are carried out easily and rapidly. Rolling bearing fault signals are used to validate the proposed fault diagnosis approach. The results indicate that the proposed approach obviously improves the classification performance of fault pattern recognition and outperforms the other traditional approaches.

No MeSH data available.