Condition Evaluation for Opening Damper of Spring Operated High-Voltage Circuit Breaker Using Vibration Time-Frequency Image

The working condition of opening damper directly affects the opening mechanical characteristics and component lifetime of high-voltage circuit breaker (HVCB). Vibration diagnostic technique is a noninvasive and efficient way for on-line evaluation of the opening damper. Vibration time-frequency image which constructed by using the method of time-frequency analysis (TFA) contains rich feature information of opening damper's working condition. In this paper, a condition evaluation method for opening damper of the spring operated HVCB based on vibration time-frequency image and convolutional neural network (CNN) is proposed and analyzed. The one-dimensional (1D) vibration signal of the HVCB is transformed into three-dimensional (3D) time-frequency image by utilizing Hilbert-Huang transform (HHT). The 3D time-frequency image directly served as a training sample for the CNN, which avoids the tedious and invalid work of manual feature extraction or selection and can achieves an accurate condition evaluation of the HVCB's opening damper. It can be seen that the noninvasive vibration on-line monitoring and the evaluation algorithm for the opening damper of the HVCB is feasible and effective. The test results carried out on several 12-kV vacuum HVCBs substantiate the proposed approach.