A comparative study of several enhanced energy operators for vibration exciter bearing fault diagnosis.

Rolling bearings of the vibration exciter are prone to failure due to long-term high amplitude alternating impact loads, causing economic losses and threatening production safety. The heavy environmental noise during the operation of the vibration exciter and the high vibration level generated by the eccentric block make the weak bearing fault features submerged and difficult to extract. Teager-Kaiser energy operator is a popular method for extracting bearing fault features. However, it has poor noise-robustness and low accuracy of frequency estimation of the exciter with heavy noise and multiple disturbances. Therefore, three enhanced energy operators-symmetric higher-order analytical energy operator (SHAEO), multi-resolution symmetric difference energy operator (MSDAEO), and symmetric higher-order frequency weighted energy operator (SHFWEO) have been introduced. This paper compares and studies the three energy operators through theoretical analysis and vibration exciter bearing fault diagnosis experiments. The results show that MSDAEO has the most outstanding noise robustness, but has the minimum effect on improving the signal-to-interference ratio (SIR) of the signal among the three energy operators. SHFWEO has the most prominent performance on improving SIR but is sensitive to the signal energy. SHAEO can increase the amplitude of the signal, and its ability to improve signal SIR is higher than MSDAEO but lower than SHFWEO. Its ability to improve signal SNR is the weakest among the three. Finally, the characteristics of preprocessing methods that can be jointly used by the three energy operators in different application scenarios are presented.