Dual-Level Denoising Method for AC Contactor Parameter Data Based on Autocorrelation Function

Parameter data is vital for AC contactor state identification and lifespan prediction. However, conventional data extraction methods often introduce noise when dealing with varying signals, and single-denoising techniques do not yield satisfactory results. This paper proposes a dual-level denoising method based on the autocorrelation function to address this. First, key parameters are acquired from a lifespan testing platform. Next, Empirical Mode Decomposition is applied to raw data to remove steady-state noise. Finally, the autocorrelation function is used to define signal-noise boundaries, and wavelet denoising is performed on noisy components. Experimental results show that this approach improves temporal correlation by 0.1 to 0.2 and surpasses conventional methods in enhancing parameter autocorrelation, confirming its effectiveness in noise reduction.