Mechanical Condition Monitoring of Vacuum Circuit Breakers Using Artificial Neural Network

摘要

The vibration signature collected on the framework of the vacuum circuit breaker is decomposed into individual frequency bands using the algorithm of wavelet packets. Main signal energy in these bands is extracted to form the condition eigenvectors. For specific condition classification, the concept of credibility is introduced in this paper. A new recognition algorithm based on the Radial Basis Function (RBF) neural network is proposed and applied to the monitoring of vacuum circuit breakers. Multiple experiments under different conditions verify that the new algorithm can recognize not only the known conditions but also a new condition of the breakers.