LAUSR

Abstract Active noise control (ANC) provides an efficient way for eliminating low-frequency noise, whereas accurate controlling non-stationary noise of rotating machinery under time-variant secondary path is still a major challenge. To address this, a novel ANC method based on signal decomposition and simultaneous perturbation is developed in this paper. In the proposed method, an indicator of periodic characteristic (IPC) with the degree of cyclostationarity is first presented to evaluate reference noise. Subsequently, a real-time strategy of the IPC combined with fast empirical mode decomposition is designed to adaptively decompose reference noise into a series of intrinsic mode functions, which are stationary and further controlled individually. Besides, the simultaneous perturbation is applied to update the controller weights for having a better tracking ability under the time-variant secondary path. Experiments were conducted to evaluate the capacity of the proposed method for controlling non-stationary reference noise under the time-variant secondary path. The results demonstrate that proposed method is capable of performing better than commonly used algorithms.