Research on Fault Feature Extraction of Hydropower Units Based on Adaptive Stochastic Resonance and Fourier Decomposition Method

In order to effectively extract the characteristics of nonstationary vibration signals from hydropower units under noise interference, an adaptive stochastic resonance and Fourier decomposition method (FDM) based on genetic algorithm (GA) are proposed in this paper. Firstly, GA is used optimize parameters so that signal can reach optimal signal-to-noise ratio (SNR) be improved. Secondly, FDM process appropriate frequency band function selected for reconstruction. Finally, Hilbert envelope demodulation analysis was performed reconstructed obtain fault spectrum. prove effectiveness superiority method, comparative experiments designed by using simulated measured swing a unit. The results show remove interference improve SNR characteristic signal, which has extensive engineering application value diagnosis units.