此页面上的内容需要较新版本的 Adobe Flash Player。

获取 Adobe Flash Player

Study on Complete Analysis of LRE Test Samples Based on PCA

Min WANG(王珉),Niao-qing HU(胡茑庆)

 

(School of Mechatronic Engineering and Automation,National University of Defense Technology,Changsha 410073,China)

 

Abstract-Incomplete data samples have a serious impact on the effectiveness of data mining. Aiming at the LRE historical test samples, based on correlation analysis of condition parameter, this paper introduced principle component analysis (PCA) and proposed a complete analysis method based on PCA for incomplete samples. At first,the covariance matrix of complete data set was calculated; Then, according to corresponding eigenvalues which were in descending, a principle matrix composed of eigen- vectors of covariance matrix was made; Finally, the vacant data was estimated based on the principle matrix and the known data. Compared with traditional method validated the method proposed in this paper has a better effect on complete test samples. An application example shows that the method suggested in this paper can update the value in use of historical test data.

 

Key words-test sample; data mining; correlation analysis; PCA; complete analysis

 

Manuscript Number: 1674-8042(2011)03-0217-05

 

doi: 10.3969/j.issn.1674-8042.2011.03.004

 

References

 

[1] Min Wang, Niao-qing Hu, Si-feng Yang, et al, 2010. Study on application of fault simulation based fault-knowledge base. Journal of Astronautic,  31(4):1253-1258.
[2] Xiao-feng Guo, 1990. Liquid Rocket Engine Testing.  Astronautics Press, Beijing, p.102-103.
[3] Balaji Rajagopalan, Mark W Isken, 2001. Exploiting data preparation to enhance mining and knowledge discovery. IEEE transactions on systems. Man and Cybernetics -Part C: Applications and Review, 31(4): 54-61.
[4] Min Li, Yu Lu, Zhen-zhong Su, 2009. The methods of filling up vacancy in data preprocessing. 5(7):1546-1548.
[5] Soibelman L M, Hyunjoo Kim, 2002. Data preparation process for construction knowledge generation through knowledge discovery in databases. Journal of Computing in Civil Engineering, (1):39 -48.
[6] Somlinski, Walczak, Einax J W, 2002. Exploratory analysis of data sets with missing elements and outliers. Chemosphere, (49):233-245.
[7] Zhi Xiao, Yong Li, Chang-long Li, 2002. Method in data-preprocessing based on correlation analysis. Journal of Chongqing University(Natural Science Edition), 25(6):132-134.
[8] Vlad Popovici, Jean Philippe Thiran, 2002. PCA in autocorrelation space. Proc. 16th International Conference on Pattern Recognition, (2):132-135.
[9] Guo Li, Peng Zhang, Xue-ran Li, et al, 2008. Sensor fault detection based on dynamic principle component analysis. Journal of Data Acquisition & Processing, 23(3):338-341.
[10] Yi Wu, Chao Li, Jian-shu Luo, et al, 2006. Fundamentals of applied mathematics. Higher Education Press, Beijing p.77-78.
[11] Xiao-juan Chen, Wen-bin Guo, 2008. Application of singular value decomposition in g-inverses. Journal of East China Normal University(National Science), (1):25-29.
[12] Ting-feng Xue, Hong-gang Liu, Jian-jun Wu, 2007. Fault detection and diagnosis for sensors of LRE based on PCA. Journal of Astronautics, 28(6):1668-1672.
 

 

[full text view]