Optimizing the Adaptive Stochastic Resonance and Its Application in Fault Diagnosis

Abstract

This paper presents an adaptive stochastic resonance method based on the improved artificial fish swarm algorithm. By this method, we can enhance the weak characteristic signal which is submerged in a heavy noise. We can also adaptively lead the stochastic resonance to be optimized to the greatest extent. The effectiveness of the proposed method is verified by both numerical simulation and lab experimental vibration signals including normal, a chipped tooth and a missing tooth of planetary gearboxes under the loaded condition. Both theoretical and experimental results show that this method can effectively extract weak characteristics in a heavy noise. In the experiment, each weak fault feature is extracted successfully from the fault planetary gear. When compared with the ensemble empirical mode decomposition (EEMD) method, the method proposed in this paper has been found to give remarkable performance.