Using EMD-AR spectrum for concurrent fault diagnosis of engine

Abstract

The problem of concurrent fault diagnosis is one of the difficulties in diesel fault diagnosis because of the complex mechanism of the composite fault. A method combining the empirical mode decomposition (EMD) and autoregressive model (AR) spectrum is proposed and applied to the concurrent fault diagnosis of crankshaft bearing and connecting rod bearing. This method not only can overcome the estimation errors and windowing effects in Hilbert separation algorithm by means of extensive properties of AR spectrum, but also can get the clear spectrum and be easy to extract the fault features. The test results confirm that.