DocumentCode :
232987
Title :
Structure damage prediction of the composite material based on EMD
Author :
Cui Jianguo ; Wang Qingtian ; Hua Jiaojiao ; Piao Chunyu ; Zhou Zhiqiang ; Jiang Liying
Author_Institution :
Shenyang Aerosp. Univ., Shenyang, China
fYear :
2014
fDate :
28-30 July 2014
Firstpage :
7784
Lastpage :
7788
Abstract :
Aiming to the characteristics of gradual changes of the structural damage of the composite material, a prediction method on this is proposed based on Experience Modal Decomposition (EMD) and Least Squares Support Vector Machines(LS-SVM)) First, collect the information on what kind of status which the composite material is staying at through by the advanced sensors. Then, extract the status features by using EMD and define the Composite Structural Damage index(SDI) according to the distance and shape similarity. Finally, use LS-SVM to predict structural damage index in order to obtain more information on future damage degree. The experimental results show that the approach is not only workable but also has higher predictive accuracy.
Keywords :
composite materials; least squares approximations; materials science computing; reliability; support vector machines; EMD; LS-SVM; SDI; composite material; composite structural damage index; damage degree; experience modal decomposition; least squares support vector machines; structure damage prediction; Composite materials; Feature extraction; Indexes; Predictive models; Shape; Support vector machines; Vectors; Experience Modal Decomposition (EMD); Least Squares Support Vector Machines (LS-SVM); Structural Damage Index; Structural Damage Prediction;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Control Conference (CCC), 2014 33rd Chinese
Conference_Location :
Nanjing
Type :
conf
DOI :
10.1109/ChiCC.2014.6896299
Filename :
6896299
Link To Document :
بازگشت