Using Multi-Window Correlation to Improve Sensitivity and Adaptability for Oil Debris Detections

Haokuo Liu,Tongyang Li,Wei Hong,Shaoping Wang

DOI: https://doi.org/10.1016/j.measurement.2021.109236

IF: 5.6

2021-01-01

Measurement

Abstract:Inductive debris detection plays an important role in realizing the real-time monitoring of oil, which can reflect the wear condition of mechanical systems and predict the remaining useful life to prevent serious accidents. However, complex working conditions which usually integrate mechanical vibrations and electrical noises with variable debris velocities are preventing the method from being more accurate. This paper proposed a novel approach based on multiple correlation windows to improve the sensitivity and adaptability for oil debris detections. The errors between the actual and the estimated time delays are analyzed to optimize the design of correlation windows. Then, the theoretical calculation and numerical simulation are conducted to validate the performance. The result indicates that the proposed method can enhance the signal-to-noise ratio (SNR) at least 5 times. Finally, an experiment is carried out to validate that the smaller debris can be detected at variable debris velocities by using the proposed method.