Engine Noise Suppression Method Based on Angle Domain Signal

An engine is a complex system with numerous moving parts and multiple types of interfering excitations. Due to strong noise interference, signal processing is often compromised, making it challenging to identify the engine’s characteristic signals. To minimize the impact of engine noise on characteristic information, this study proposes an engine noise suppression method based on angle-domain signals. This method employs equal-angle sampling, synchronous averaging, and wavelet packet decomposition to process the engine’s vibration signal. Additionally, the correlation coefficient method is applied to evaluate the correlation level of the sub-order segment signal and parent-order segment signal to determine the optimal noise suppression output signal. The efficacy of this method is validated using engine simulation signals. The test results show superior noise suppression performance compared to traditional approaches when using angle-domain signal processing. The influencing factors of the noise suppression method are analyzed by the peak ratio method which proposes a new approach for researches on engine noise suppression and feature extraction.