Title :
Research of an improved PCA method for abnormality diagnosis in synchronous multi-dimensional data stream
Author :
Tongyao Yang ; Bin Wang ; Chuan Li ; Bi He
Author_Institution :
Fac. of Inf. Eng. & Autom., Kunming Univ. of Sci. & Technol., Kunming, China
Abstract :
An improved PCA method is presented which combined the PCA technology with data mining technology. In this method, the problem of the original data stream variation tendency is mapped to the eigenvector space, and the steady-state eigenvector is solved, then the abnormal changes can be diagnosed by analyzing the relationship between the instantaneous eigenvector and the steady-state eigenvector. The method is applied to analyze the synchronous multi-dimensional data stream for tunnel strain monitoring. Result shows this method can reflect the changes of the aperiodic variables timely and realize the anomaly monitoring for multi-dimensional data stream effectively.
Keywords :
data analysis; data mining; eigenvalues and eigenfunctions; principal component analysis; system monitoring; abnormality diagnosis; aperiodic sampling data analysis; aperiodic variables; data mining technology; data stream variation tendency; dissimilarity function; eigenvector space; improved PCA method; instantaneous eigenvector; sliding window technique; steady-state eigenvector; synchronous multidimensional data stream; tunnel strain monitoring; Correlation; Covariance matrices; Data mining; Monitoring; Principal component analysis; Steady-state; Strain; Abnormality diagnosis; Multi-dimensional data stream; PCA; Steady-state eigenvector;
Conference_Titel :
Control Conference (CCC), 2013 32nd Chinese
Conference_Location :
Xi´an
