变分框架下多尺度熵相关优化的模态分解在故障诊断中的应用
发布时间：2018-08-04 18:56
【摘要】：针对变分框架下,一种新的模态分解——变分模态分解(Variational Mode Decomposition,VMD)的最优模态分量选择和关键参数辨识问题,借鉴折半查找的思想,提出应用多尺度熵相关系数和频域相关系数来改进VMD的上述关键环节,并通过轴承故障信号仿真研究其频域分解的数据特点,揭示其滤波本质;轴承故障信号仿真及工程应用的结果表明,相对于经验模态分解(Empirical Mode Decomposition,EMD)和总体平均经验模态分解(Ensemble Empirical Mode Decomposition,EEMD),改进后的VMD(IVMD)去噪效果更为明显,是一种有效的自适应频域模态分解方法,可更为准确地提取出微弱特征频率信息,实现轴承故障的正确识别。
[Abstract]:In view of the problem of optimal modal component selection and key parameter identification for a new mode decomposition-variational mode decomposition (Variational Mode DecompositionVMD) framework, the idea of half-searching is used for reference. Multi-scale entropy correlation coefficient and frequency-domain correlation coefficient are proposed to improve the key links of VMD. The characteristics of frequency domain decomposition data are studied by simulation of bearing fault signal, and the essence of filtering is revealed. The simulation results of bearing fault signals and engineering application show that the improved VMD (IVMD) denoising effect is more obvious than that of the empirical mode decomposition (Empirical Mode) and the total average empirical mode decomposition (Ensemble Empirical Mode). It is an effective adaptive frequency domain mode decomposition method, which can extract the weak characteristic frequency information more accurately and realize the correct identification of bearing fault.
【作者单位】： 广东石油化工学院计算机与电子信息学院;广东石油化工学院广东省石化装备故障诊断重点实验室;华南理工大学自动化科学与工程学院;
【基金】：国家自然科学基金项目(61174113,61672174) 广东省自然科学基金项目(2016A030307029)
【分类号】：TE65;TQ050.7
，
本文编号：2164809
[Abstract]:In view of the problem of optimal modal component selection and key parameter identification for a new mode decomposition-variational mode decomposition (Variational Mode DecompositionVMD) framework, the idea of half-searching is used for reference. Multi-scale entropy correlation coefficient and frequency-domain correlation coefficient are proposed to improve the key links of VMD. The characteristics of frequency domain decomposition data are studied by simulation of bearing fault signal, and the essence of filtering is revealed. The simulation results of bearing fault signals and engineering application show that the improved VMD (IVMD) denoising effect is more obvious than that of the empirical mode decomposition (Empirical Mode) and the total average empirical mode decomposition (Ensemble Empirical Mode). It is an effective adaptive frequency domain mode decomposition method, which can extract the weak characteristic frequency information more accurately and realize the correct identification of bearing fault.
【作者单位】： 广东石油化工学院计算机与电子信息学院;广东石油化工学院广东省石化装备故障诊断重点实验室;华南理工大学自动化科学与工程学院;
【基金】：国家自然科学基金项目(61174113,61672174) 广东省自然科学基金项目(2016A030307029)
【分类号】：TE65;TQ050.7
，
本文编号：2164809
本文链接：https://www.wllwen.com/kejilunwen/huaxuehuagong/2164809.html
