Use of cyclostationarity based condition indicators for gear fault diagnosis under fluctuating speed condition

Abstract

The vibration based gear health monitoring is one of the condition monitoring techniques, widely used in industry. Under fluctuating speed conditions, vibration based conventional gear fault diagnosis methods like FFT and condition indicators (CI) like rms and kurtosis, fails to differentiate a faulty gear from a healthy one. Under such conditions, cyclic changes are observed in mean and variance of a vibration signal. CI using such statistical parameters for non-stationary gear vibration signal may mislead the entire fault diagnosis approach. In this chapter, CI based on cyclostationarity has been explained and used for gear fault diagnosis for fluctuating speed conditions. This chapter shows an advantage of using cyclostationarity based CI over conventional CI, for example rms and kurtosis, to diagnose fault. Result shows the effectiveness of the cyclostationarity based CI in differentiating gear health. © Springer International Publishing AG 2017.