参数优化VMD对引风机振动信号处理研究

Research on Vibration Signal Processing of Induced Draft Fan by VMD with Parameter Optimization

针对传统振动信号分析算法经验模态分解(Empirical Mode Decomposition,EMD)在引风机振动信号分析中所呈现的模态混叠和端点效应现象,在含有较强噪声的非线性非平稳的振动信号去噪效果较差,因此,论文提出一种参数优化的变分模态算法(Variational Mode Decomposition,VMD),建立包络熵适应度函数,利用PSO算法确定VMD算法中参数[K,α]最优值,从而提高算法的稳定性和可靠性。

The traditional vibration signal analysis algorithm empirical mode decomposition(EMD)in induced draft fan vibration signal analysis presents the phenomenon of modal aliasing and endpoint effect,and the denoising effect is poor in the nonlinear and non-stationary vibration signal with strong noise,so it is necessary to improve the denoising effect. In this paper,a parameter optimized variational mode decomposition(VMD)algorithm is proposed. The envelope entropy fitness function is established,and the PSO algorithm is used to determine the optimal value of parameters in VMD algorithm,so as to improve the stability and reliability of the algorithm.