The vibration signal denoising method for high voltage circuit breaker mechanical condition diagnosis based on dual-tree complex wavelet transform

Abstract

The vibration signal of the operating mechanism during the operation of high-voltage circuit breaker (HVCB) contains a large amount of information, which can be used to reflect the mechanical status of circuit breakers and thus carry out early warning and diagnosis of potential faults. However, the complex operating environment of the circuit breaker contains a large number of noise signals, which makes the vibration signals obscured and thus difficult to extract the feature values accurately and easily. In order to solve this problem, this paper proposes a new denoising method based on Dual-Tree Complex Wavelet Transform (DT-CWT), which is specifically used for the feature extraction of mechanical vibration signals from HVCB. Real vibration signals are first simulated and full-band random noise is injected to replicate the field noise conditions. Various decomposition layers and denoising methods within the DT-CWT framework were tested on these simulated signals to assess their effectiveness. The performance of the proposed methods was further validated using actual acquired HVCB vibration signals. The results show that DT-CWT is a powerful tool for denoising mechanical vibration signals, improving the clarity of the signals and maintaining the integrity of the signals in the presence of noise, providing a significant improvement over conventional wavelet method. This study provides a robust solution for reducing noise in circuit breaker vibration signals and an effective reference method for improving mechanical status diagnosis.