基于变分模态分解的机械故障诊断方法研究

Research on Mechanical Fault Diagnosis Method Based on Variational Mode Decomposition

变分模态分解(VMD)是一种新的自适应信号分解方法,该方法的核心思想是假设每个模态的绝大部分都是紧紧围绕在某一中心频率周围的,然后将模态带宽的求解问题转化为约束优化问题,求解出每个模态。将VMD引入到机械故障诊断中,提出一种基于VMD的机械故障诊断方法,并与集合经验模态分解(EEMD)方法进行对比分析。仿真结果表明:VMD方法明显优于EEMD方法,能有效地分解出信号的固有模态;与EEMD方法相比较,该方法模态混叠现象弱,计算效率高,理论充分。将VMD方法成功地应用到转子不同碰摩严重程度的故障数据分析实验中,实验结果进一步验证了该方法的有效性,能够揭示出碰摩故障数据的频率结构,区分碰摩故障的严重程度。

Variational mode decomposition （VMD） is a new adaptive signal decomposition method. The idea of this method is that most of each mode is assumed to be tightly around a center frequency, the sol- ving problem of mode bandwidth is converted into an optimization problem with the constrain conditions, and finally each modal is solved. VMD is introduced into the mechanical fault diagnosis, and a fault di- agnosis method based on VMD is proposed. The proposed method is compared with the ensemble empiri- cal mode decomposition （EEMD）. The simulated result shows VMD method is superior to EEMD meth- od, the intrinsic mode of signal can be effectively decomposed by the VMD method. Compared with the EEMD method, the proposed method has some distinct advantages, such as weak mode mixing phenome- non, high calculation efficiency and sufficient theory. The proposed method has been successfully applied to the rub-impact fault diagnosis of rotor system. The experimental results show that the proposed method is effective, and can effectively reveal the frequency structure in rubbing fault and discern the severity of rub-impact fault.