Fault feature signal extraction method for rolling bearings in gas turbine engines based on threshold parameter decision screening

Abstract

Vibrations during the operation of gas turbine engines (GTEs) include shocks from the combustion process as well as vibrations from high-speed rotating components, which may mask or distort the characteristic signals of rolling bearing failure. Compared to those of ordinary rolling bearings, the fault characteristics of rolling bearings in GTEs are more difficult to extract. In response to the difficulty of extracting characteristic signals of rolling bearing faults in gas turbine engines due to the effect of environmental noise, a method for extracting characteristic features of GTE rolling bearings is proposed. First, the particle swarm optimization (PSO) algorithm is used to optimize the parameters in the variational mode decomposition (VMD) algorithm, resulting in K0 modal components. VMD can not only suppress the modal aliasing phenomenon, but also decompose signals at different scales, thereby providing multi-resolution information. Second, a new parameter harmonization equation is introduced, which balances the relationship between the kurtosis and correlation coefficients, and is fused into a single parameter P. Subsequently, high signal-to-noise ratio (SNR) signals are selected ted based on threshold parameter criteria, and these high-SNR signals are used to generate new vibration signals. Finally, the envelope spectrum is used to extract fault characteristics of bearings. According to the results, the fault feature signal extraction (FFSE) method can extract fault characteristics in rolling bearings under both simple and complex transmission paths.