An Ultrasound-Based In Situ Monitoring for Water-Lubricated Thrust Bearing Wear

Water-lubricated bearings are prone to abnormal wear due to mixed lubrication under harsh conditions. Accurate wear measurement is critical for analyzing the wear mechanisms and evaluating the health of bearings. However, the traditional wear monitoring usually needs to disassemble the bearing structure, which is ineffective. The aim of this study is to achieve the in situ monitoring of the water-lubricated thrust bearing based on nondestructive ultrasound technology. The traditional time-domain ultrasound method is limited to accurately identify the wear due to the nonhomogeneity and large attenuation of the waterlubricated bearing, while the water film could influence the echo signal. Therefore, a multiparameter inversion method based on frequency domain features of the ultrasonic signal is performed to simultaneously identify the thickness, density, and attenuation factor of the bearing. Pearson's correlation coefficient of measured and theoretical ultrasonic reflection coefficient amplitude spectrum (URCAS) is taken as the objective function. Finite element simulation and the water-lubricated thrust bearing wear experiment are used to verify the proposed method and its potential for practical scenarios. The results show that the inversion error of the thickness-density-attenuation factor is the smallest in threeparameter inversion. The absolute wear error of the proposed method is 32 and $45~\mu \mathrm{m}$ in the bearing wear test. The water film thickness of less than $30~\mu \mathrm{m}$ has almost no effect on the wear measurement of the bearing pad. The proposed method provides a solution to achieve the in situ monitoring of water-lubricated thrust bearing wear.