基于ACO-USK优化VMD参数的滚动轴承故障诊断研究

Research on rolling bearing fault diagnosis based on ACO-USK optimized VMD parameters

传统变分模态分解(VMD)技术需要人为主观预设模态分解个数K和二次惩罚因子α,由此可能导致信号的欠分解、过分解、模态混叠或信息丢失等问题,从而影响对滚动轴承早期故障信号的分解效果。本文根据峭度指标对滚动轴承早期故障异常敏感的特点,提出了一种以联合平方峭度(USK)指标为目标函数,结合蚁群优化(ACO)算法的ACO-USK优化方法,对VMD模态分解个数K和二次惩罚因子α进行自适应寻优。研究结果表明:对于滚动轴承早期故障信号,与以包络熵(EE)为目标函数的VMD优化方法对比,本文提出的方法既具有较好的包络谱信噪比(SNRES),又有在计算用时上的优越性,具有一定的工程应用价值。

Traditional variational mode decomposition(VMD)technology requires artificial presets of the number of modal decompositions K and the quadratic penalty factorα,which may lead to under-decomposition,over-decomposition,mode mixing or information loss of the signal.These problems will affect the decomposition effect of early fault signals of rolling bearings.Based on the characteristic that the kurtosis index is extremely sensitive to early fault of rolling bearings,this paper proposes a method using the union squared kurtosis(USK)as the objective function and combined with the ant colony optimization(ACO)algorithm.The ACO-USK optimization method performs adaptive optimization on the number of VMD mode decompositions K and the quadratic penalty factorα.The research results show that for early fault signals of rolling bearings,compared with the VMD optimization method with envelope entropy as the objective function,the method proposed in this paper not only has a good signal noise ratio of envelope spectral(SNRES),but also has advantages in calculation time.Therefore,the method proposed in this paper has good engineering application value.