A new method for arrival time determination of impact signal based on HHT and AIC

Time-difference method is usually used to locate loose parts in nuclear power plant, the key to which is estimating the arrival time of impact signal caused by the crash of loose parts. However, the dispersion behavior of impact signal and the noise of nuclear power station primary circuit have negative effect on the arrival time determination. In this paper, a method of arrival time determination of impact signal based on Hilbert-Huang Transform (HHT) and Akaike Information Criterion (AIC) is proposed. Firstly, the impact signal is decomposed by Empirical Mode Decomposition (EMD). Then the instantaneous frequency of the first intrinsic mode function (IMF) is calculated, which characterizes the difference between the background noise and the impact signal. The arrival time is determined finally by AIC function. The proposed method is tested through simulation experiment which takes steel balls as the real loose parts. The deviation between the arrival time determined by proposed method and the real arrival time distributes stably under different SNRs and different sensor-to-drop point distances, mostly within the range ±0.5 ms. The proposed method is also compared with another AIC technique and a RMS approach, both of which have more dispersive distribution of deviation, quite a lot out of the range ±1 ms.