为了有效缓解噪声对故障诊断精度的影响和信号分解过程中出现的模态混叠和端点效应问题，本文提出基于WOA-VMD的信号分解与重构方法。首先利用WOA对VMD算法中的分解层数和惩罚因子进行自适应优化，将优化后的参数输入到VMD中对含噪信号进行分解。然后利用累积峭度值占比筛选出相关量较大的IMF分量进行重构，以达到去除噪声的效果。经过实验验证，WOA-VMD能够有效的去掉信号中的大部分噪声，且极大程度的保留了有效信息，同时该方法的在信噪比指标上的表现优于其他方法，为转子-轴承系统故障诊断提供基础。

In order to effectively alleviate the influence of noise on the fault diagnosis accuracy and the problems of modal aliasing and endpoint effect during signal decomposition, this paper proposes a signal decomposition and reconstruction method based on WOA-VMD. Firstly, WOA is used to adaptively optimize the number of decomposition layers and the penalty factor in the VMD algorithm, and the optimized parameters are input into the VMD to decompose the noise-containing signal. Then the IMF components with larger correlation amount are screened out for reconstruction using the cumulative percent kurtosis to achieve the effect of noise removal. After experimental verification, the WOA-VMD can effectively remove most of the noise in the signal and retain the effective information to a great extent, while the method outperforms other methods in terms of signal-to-noise ratio, which provides a basis for the fault diagnosis of rotor-bearing system.

国家自然科学基金项目（面上项目，重点项目，重大项目）