APPLICATION OF HIT-OR-MISS WAVELET TRANSFORM AND PRINCIPLE COMPONENT ANALYSIS ON POWER DISTURBANCE IDENTIFICATION

L. Yi ； T. Y. Ji ； M. S. Li ； Q. H. Wu ； T. Wu

《電機工程學刊》 22卷4期 (2015/08) Pp. 127-135

https://doi.org/10.6329/CIEE.2015.4.02

並列摘要

When a power system is disturbed, the voltage waveform would distort and various degrees of oscillation would happen at the onset and end of the disturbance. According to this characteristic, in this paper, a scheme that engages hit-or-miss wavelet transform (HMWT) and principal component analysis (PCA) is proposed for power disturbance detection and classification. The voltage signal to be identified is decomposed by HMWT and a coefficient matrix is obtained, the energy entropy (EE) of which is thereby used for disturbance detection. For the classification, a strategy based on nearest neighbor classifier is used. In the training procedure, a projection matrix and a principle component matrix are obtained for each type of disturbance based on HMWT and PCA. In classification the projection matrix is used to project the HMWT coefficient matrix of an input voltage signal into a principle component matrix. Afterwards, the distance of the principle component matrix from the input signal to the principle component matrix obtained from the training signals is used for classification. Extensive simulation work has been conducted and results have indicated that the scheme proposed in this paper has a high identification rate for common power disturbance detection and classification.

並列關鍵字

Principal component analysis ； power disturbance detection and classification ； hit-or-miss wavelet transform ； energy entropy ； nearest neighbor classifier