Mode Selection in Variational Mode Decomposition and Its Application in Fault Diagnosis of Rolling Element Bearing

Bearings are standard and furthermore enormous parts in any rotating machinery. For the perfect operational rotating system, bearing assumes a key responsibility. Once a calamity happens, it will cause huge budgetary episodes and even security threats. Proper fault diagnosis of the rolling element bearings will put a check on all such issues. In the past, several fault diagnosis strategies have been implemented and found effective due to the involvement of signal decomposition. A proper choice of decomposed modes may extract vital information and evidence on fault type, location, and condition. Inspired by the same fact, the work proposes statistical criteria for mode selection in the vibrational mode decomposition method utilizing entropy and covariance parameters. The criteria have been validated on a test signal sample gathered over bearing with inner race failure. The illustration proves the promising nature of the method which makes the method superior till date.