In this paper a new method to fault diagnosis for circuit breakers is presented based on the principle of wavelet packet and characteristic entropy. The method introduces the way to abstract the characteristic entropy from vibration signals of HVCB. Firstly, the vibration signal is decomposed by the three-layer wavelet packet, and eight signals of each junction at the third level are reconstructed; Secondly, the characteristic vector is extracted from the envelopes of reconstructed signals; and lastly the Euclidean distance between the vectors of normal state and fault states is seemed as characteristic parameter for fault diagnosis of HVCB. The experimentation without loads indicates this method can easily and accurately diagnose the faults of HVCB, and detect the state change of circuit breakers both in time domain and frequency domain at the same time.

No abstract

No abstract

In this paper, the basic principle of support vector machine is introduced firstly; Then a new method to diagnosis fault for high voltage circuit breakers is presented based on the introduction of wavelet packet and characteristic entropy. The new method decomposes vibration signals with wavelet packet, and extracts entropy parameters from the restructured signals at the third level. Finally, the new method and SVM are applied to the fault recognition of circuit breakers, and the usable process is introduced in detail in the paper. In addition, SVM is compared with the artificial neural network, and the paper concludes that in terms of classification and learning speed, SVM is better than neural network clearly, and SVM is more applicable to fault recognition of circuit breakers.

No abstract