Limits...
A Spacecraft Electrical Characteristics Multi-Label Classification Method Based on Off-Line FCM Clustering and On-Line WPSVM.

Li K, Liu Y, Wang Q, Wu Y, Song S, Sun Y, Liu T, Wang J, Li Y, Du S - PLoS ONE (2015)

Bottom Line: Firstly, both the fuzzy c-means (FCM) offline clustering and the principal component feature extraction algorithms are applied for the feature selection process.Secondly, the approximate weighted proximal support vector machine (WPSVM) online classification algorithms is used to reduce the feature dimension and further improve the rate of recognition for electrical characteristics spacecraft.The experimental results indicate that the method proposed can obtain better data features of the spacecraft electrical characteristics, improve the accuracy of identification and shorten the computing time effectively.

View Article: PubMed Central - PubMed

Affiliation: Ergonomics and Environment Control Laboratory, Beihang University, Beijing, China.

ABSTRACT
This paper proposes a novel multi-label classification method for resolving the spacecraft electrical characteristics problems which involve many unlabeled test data processing, high-dimensional features, long computing time and identification of slow rate. Firstly, both the fuzzy c-means (FCM) offline clustering and the principal component feature extraction algorithms are applied for the feature selection process. Secondly, the approximate weighted proximal support vector machine (WPSVM) online classification algorithms is used to reduce the feature dimension and further improve the rate of recognition for electrical characteristics spacecraft. Finally, the data capture contribution method by using thresholds is proposed to guarantee the validity and consistency of the data selection. The experimental results indicate that the method proposed can obtain better data features of the spacecraft electrical characteristics, improve the accuracy of identification and shorten the computing time effectively.

No MeSH data available.


Random electrical characteristics data sample.There are 10, 50, 100, and 500 samples randomly extracted from the real data set, combining three categories of typical fault signals, (a)Fault one, (b) Fault two, (c)Fault three.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4636359&req=5

pone.0140395.g010: Random electrical characteristics data sample.There are 10, 50, 100, and 500 samples randomly extracted from the real data set, combining three categories of typical fault signals, (a)Fault one, (b) Fault two, (c)Fault three.

Mentions: The experiment in this paper used 10, 50, 100, and 500 samples randomly extracted from the real data set, combining six elements of the physical process: motor speed, order turning angle, bearing temperature, motor current, top and spinning (Fig 7), and three categories of typical fault signals are shown in Fig 10. We recorded the training time and the accuracy of each method. We changed the parameters of each method, and the highest accuracy was recorded as the result for each method. Finally, the results of each method are compared with each other. The typical fault signals of randomly chosen real data sample and the samples’s dimension reduction by PCA are detailed in Figs 11 and 12, on which the threshold is set as 0.1 and the similarity result is 95%. These data are the input to the next simulation experiments.


A Spacecraft Electrical Characteristics Multi-Label Classification Method Based on Off-Line FCM Clustering and On-Line WPSVM.

Li K, Liu Y, Wang Q, Wu Y, Song S, Sun Y, Liu T, Wang J, Li Y, Du S - PLoS ONE (2015)

Random electrical characteristics data sample.There are 10, 50, 100, and 500 samples randomly extracted from the real data set, combining three categories of typical fault signals, (a)Fault one, (b) Fault two, (c)Fault three.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4636359&req=5

pone.0140395.g010: Random electrical characteristics data sample.There are 10, 50, 100, and 500 samples randomly extracted from the real data set, combining three categories of typical fault signals, (a)Fault one, (b) Fault two, (c)Fault three.
Mentions: The experiment in this paper used 10, 50, 100, and 500 samples randomly extracted from the real data set, combining six elements of the physical process: motor speed, order turning angle, bearing temperature, motor current, top and spinning (Fig 7), and three categories of typical fault signals are shown in Fig 10. We recorded the training time and the accuracy of each method. We changed the parameters of each method, and the highest accuracy was recorded as the result for each method. Finally, the results of each method are compared with each other. The typical fault signals of randomly chosen real data sample and the samples’s dimension reduction by PCA are detailed in Figs 11 and 12, on which the threshold is set as 0.1 and the similarity result is 95%. These data are the input to the next simulation experiments.

Bottom Line: Firstly, both the fuzzy c-means (FCM) offline clustering and the principal component feature extraction algorithms are applied for the feature selection process.Secondly, the approximate weighted proximal support vector machine (WPSVM) online classification algorithms is used to reduce the feature dimension and further improve the rate of recognition for electrical characteristics spacecraft.The experimental results indicate that the method proposed can obtain better data features of the spacecraft electrical characteristics, improve the accuracy of identification and shorten the computing time effectively.

View Article: PubMed Central - PubMed

Affiliation: Ergonomics and Environment Control Laboratory, Beihang University, Beijing, China.

ABSTRACT
This paper proposes a novel multi-label classification method for resolving the spacecraft electrical characteristics problems which involve many unlabeled test data processing, high-dimensional features, long computing time and identification of slow rate. Firstly, both the fuzzy c-means (FCM) offline clustering and the principal component feature extraction algorithms are applied for the feature selection process. Secondly, the approximate weighted proximal support vector machine (WPSVM) online classification algorithms is used to reduce the feature dimension and further improve the rate of recognition for electrical characteristics spacecraft. Finally, the data capture contribution method by using thresholds is proposed to guarantee the validity and consistency of the data selection. The experimental results indicate that the method proposed can obtain better data features of the spacecraft electrical characteristics, improve the accuracy of identification and shorten the computing time effectively.

No MeSH data available.