期刊
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
卷 25, 期 11, 页码 2043-2052出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TNNLS.2014.2303086
关键词
Fault angle; fault recognition and classification; multiwavelet packet entropy; neural network; transmission line
类别
资金
- National Nature Science Foundation of China [U1134205, 51377136]
- Program for New Century Excellent Talents in University in China [NECT-08-0825]
Multiwavelets possess better properties than traditional wavelets. Multiwavelet packet transformation has more high-frequency information. Spectral entropy can be applied as an analysis index to the complexity or uncertainty of a signal. This paper tries to define four multiwavelet packet entropies to extract the features of different transmission line faults, and uses a radial basis function (RBF) neural network to recognize and classify 10 fault types of power transmission lines. First, the preprocessing and postprocessing problems of multiwavelets are presented. Shannon entropy and Tsallis entropy are introduced, and their difference is discussed. Second, multiwavelet packet energy entropy, time entropy, Shannon singular entropy, and Tsallis singular entropy are defined as the feature extraction methods of transmission line fault signals. Third, the plan of transmission line fault recognition using multiwavelet packet entropies and an RBF neural network is proposed. Finally, the experimental results show that the plan with the four multiwavelet packet energy entropies defined in this paper achieves better performance in fault recognition. The performance with SA4 (symmetric antisymmetric) multiwavelet packet Tsallis singular entropy is the best among the combinations of different multiwavelet packets and the four multiwavelet packet entropies.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据