基于MFOA-KELM的滚动轴承故障诊断

Fault Diagnosis of Rolling Bearings Based on MFOA-KELM

针对滚动轴承故障数据大容量、多样性的特点,提出一种基于飞蛾扑火优化算法(MFOA)优化核极限学习机(KELM)的故障分类识别方法。首先使用主成分分析对原始数据进行降维处理,删除冗余信息,构建训练集和测试集。然后采用飞蛾扑火优化算法对KELM的惩罚系数C与核函数σ进行寻优选择,将具有最优参数的KELM作为故障诊断模型。最后通过美国Case Western Reserve University滚动轴承数据集进行仿真实验,并与一些传统的故障识别方法进行比较,验证了所提方法在准确率和时间效率方面的优越性。

Aiming at the characteristics of large capacity and diversity of rolling bearing fault data,a fault classification and identification method based on the Moth-flame Optimization Algorithm(MFOA)optimized Kernel Extreme Learning Machine(KELM)is proposed.Firstly,principal component analysis is used to reduce the dimension of the original data,delete the redun-dant information,and construct the training set and test set.Then,the moth-flinging optimization algorithm is used to optimize KELM's penalty coefficient and kernel function,and the KELM with the optimal parameters is used as the fault diagnosis model.Fi-nally,the simulation experiment is conducted on rolling bearing data set of Case Western Reserve University in the United States,and compared with some traditional fault recognition methods,the advantages of the proposed method in accuracy and time efficien-cy are verified.