An improved adaptive stochastic resonance method for improving the efficiency of bearing faults diagnosis

Abstract

The general scale transformation (GST) method is used in the bistable system to deal with the weak high-frequency signal submerged into the strong noisy background. Then, an adaptive stochastic resonance (ASR) method with the GST is put forward and realized by the quantum particle swarm optimization (QPSO) algorithm. Through the bearing fault simulation signal, the ASR method with the GST is compared with the normalized scale transformation (NST) stochastic resonance (SR). The results show that the efficiency of the GST method is higher than the NST in recognizing bearing fault feature information. In order to simulate the actual engineering environment, both the adaptive GST and the NST methods are implemented to deal with the same experimental signal, respectively. The signal-to-noise ratio (SNR) of the output is obviously improved by the GST method. Specifically, the efficiency is improved greatly to extract the weak high-frequency bearing fault feature information. Moreover, under different noise intensities, although the SNR is decreased versus the increase of the noise intensity, the ASR method with the GST is still better than the traditional NST SR. The proposed GST method and the related results might have referenced value in the problem of weak high-frequency feature extraction in engineering fields.