变工况条件下滚动轴承故障诊断研究

Research on Fault Diagnosis of Rolling Bearing under Variable Working Condition

针对变工况运行条件下滚动轴承的故障识别率低的问题,提出了自适应白噪声完备集合经验模态分解(CEEMDAN)和排列熵结合的故障诊断方法。首先将轴承原始信号通过CEEMDAN分解成一系列的固有模态函数(IMF)并计算他们的排列熵,将各IMF的排列熵按从大到小排列并选取其中的9个IMF进行重构,最后用粒子群优化-支持向量机(PSO-SVM)模型进行故障分类。该诊断方法在轴承故障数据集上有较好的诊断效果。

Aiming at the problem of low fault identification rate of rolling bearing under variable operating conditions,a fault diagnosis method combining complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN)and permutation entropy was proposed.Firstly,the original signal of bearing was decomposed into a series of intrinsic mode function(IMF)through CEEMDAN and their permutation entropies were calculated.The permutation entropies of each IMF were arranged from large to small and nine of IMFs were selected for reconstruction.Finally,the partical swarm optimization-support vector machine(PSO-SVM)model was used for fault classification.The diagnosis method has a good effect on the bearing fault data set.