Abstract: In order to detect the liquid atomization status, acoustic emission (AE) signals which originated from the impact of liquid spray on the plate are collected by using AE measurement. The generating mechanism and character of the AE signals, and the characteristic scale (J1) of AE signals which represented the status of liquid atomization are obtained by using power spectrum analysis, wavelet transform and rescaled range (R/S) analysis. Since the energy fraction of AE signals in J1 scale changed regularly with the atomization status, a new method for judging the liquid atomization performance is presented. The liquid was better atomized with the increase of AE energy fraction in J1 scale. When the AE energy fraction in J1 scale increased rapidly, the liquid spray is sufficiently atomized. This method is verified by the photos of liquid spray. Based on the variation of the AE energy fraction in J1 scale, it is found that the atomization performance became better along the axial direction of the spray bundle. It is concluded that multi-scale analysis of acoustic signals is feasible for detecting the liquid atomization status which have the merit of being sensitive, convenient, real-time and safe.

Key words: Acoustic emission, Atomization, Power spectrum, Rescaled range analysis, Wavelet