Feature extraction method for weak mechanical fault signal based on double coupled Duffing oscillator and EMD

Abstract

A new feature extraction method for early weak mechanical fault signal is proposed, which is based on the combination of double coupled Duffing oscillator (DCDO) and empirical mode decomposition (EMD). Firstly, machine fault signal is disintegrated into several intrinsic mode function (IMF) by EMD. After that, each intrinsic mode function which contains fault characteristics is inputted to the DCDO system. The characteristics corresponding to the weak fault signal will be enhanced, and then the fault detection and diagnosis are achieved. This method uses EMD decomposition method to extract the features of non stationary signals, and combines the ability of Duffing oscillator(DO) to restrain the noise and to be sensitive to the weak periodic signal of the same frequency. It can effectively improve the noise ratio of the output signal. Simulation results and experimental data show that the proposed method is effective.