风力发电机轴承复合故障特征提取方法研究

针对较强背景噪声干扰下的风力发电机轴承复合故障特征难以准确提取的问题,提出了基于自适应最大相关峭度解卷积(AMCKD)的风力发电机轴承复合故障特征提取的方法。首先,利用人工鱼群算法(AFSA)以解卷积信号包络谱的相关峭度值为目标函数对最大相关峭度解卷积算法(MCKD)的影响参数自适应寻优;然后,将影响参数优化的MCKD对原始故障信号进行解卷积处理,对解卷积信号进行包络谱分析,通过包络谱中的主导频率成分与轴承各元件的故障特征频率相对比,准确实现轴承复合故障的诊断。仿真和工程应用实例验证了所提方法的有效性与实用性。

Feature extraction method for bearing composite faults of a wind turbine

Aiming at the problem that the complex fault characteristics of wind turbine bearings under strong background noise are difficult to accurately extract, a method based on adaptive maximum correlation kurtosis deconvolution (AMCKD) for the wind turbine bearing composite fault feature extraction was proposed.Using the artificial fish swarm algorithm (AFSA), and taking the correlated kurtosis of the deconvoluted signal envelope spectrum as an objective function, the influential parameters of the maximum correlation kurtosis deconvolution algorithm (MCKD) were adaptively optimized.Then, the optimized MCKD was used to deconvolute the original fault signal, and an envelope spectrum analysis was performed on the deconvoluted signal.Through the comparison of the dominant frequency components in the envelope spectrum with the fault characteristic frequency of each component of the bearing, the diagnosis of the bearing composite faults was accurately realized.Simulation and engineering application examples verify the effectiveness and practicability of the proposed method.