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

中文翻译：

---

基于自适应随机共振和傅里叶分解法的水电机组故障特征提取研究

为了有效地提取噪声干扰下水电机组非平稳振动信号的特征，提出了一种基于遗传算法的自适应随机共振傅里叶分解方法。首先，GA用于优化谐振参数，从而使信号可以达到最佳谐振，并且可以提高信噪比（SNR）。其次，FDM用于处理信号，并选择适当的频带函数进行重构。最后，对重构后的信号进行希尔伯特包络解调分析，从包络谱中获取故障特征。为了证明所提方法的有效性和优越性，通过使用模拟信号和水轮机测得的摆动信号来设计对比实验。结果表明，该方法能有效消除噪声干扰，提高信噪比，提取信号特征频率，在水电机组故障诊断中具有广泛的工程应用价值。