Application of EMD and SVD in fault identification
Zhihui Zhu, Yunlian Sun · International Power Engineering Conference · 2007
The method based on empirical mode decomposition (EMD) and singular value decomposition (SVD) for power fault identification is presented in this paper. First, fault signal was adaptively decomposed into a series of smooth intrinsic mode functions (IMFs) with different time scales via EMD; second, the matrix is formed by different level IMFs and SVD method decompose the matrix to obtain singular value as eigenvector; finally, support vector machines (SVMs) is used as classifier to identify the fault type. The simulation results show that EMD and SVD can well extract the fault feature and SVMs network can attain high accuracy of fault identification.