基于IBAS-Elman网络的滚动轴承故障诊断研究  

Research on Fault Diagnosis of Rolling Bearing Based on IBAS-Elman Network

作　　者：瞿红春[1] 许旺山 郭龙飞 朱伟华 高鹏宇 QU Hongchun;XU Wangshan;GUO Longfei;ZHU Weihua;GAO Pengyu(College of Aeronautical Engineering,Civil Aviation University of China,Tianjin 300300,China)

机构地区：[1]中国民航大学航空工程学院,天津300300

出　　处：《机床与液压》2020年第16期201-205,共5页Machine Tool & Hydraulics

基　　金：中央高校基本科研业务费资助项目(201935)。

摘　　要：为了提高民航发动机滚动轴承故障诊断正确率,提出基于改进天牛须搜索算法优化Elman神经网络的诊断模型。针对天牛须搜索算法易早熟等缺陷,对天牛质心位置和步长更新操作进行改进,并用改进算法优化Elman网络的学习率、权重和阈值。使用IBAS-Elman模型对滚动轴承故障和正常状态进行诊断,并分析Elman网络延迟向量比例因子对滚动轴承故障诊断率的影响。为了验证IBAS-Elman模型的有效性,将天牛须搜索算法、萤火虫算法和遗传算法作为对比算法。实验结果表明:改进天牛须搜索算法收敛精度优于对比算法。An Elman neural network optimized by improved beetle antennae search algorithm was proposed to improve the fault diagnosis accuracy of rolling bearing of civil aviation engine.Aiming at the defects of beetle antennae search such as precocity,the centroid position and step update operation of beetle were improved.Meanwhile the learning rate,weight and threshold of Elman network were optimized by improved algorithm.The improved beetle antennae search algorithm optimization Elman neural network was used to diagnose rolling bearing fault and normal state,and the influence of Elman network delay vector scale factor on fault diagnosis rate of bearings was analyzed.In order to verify the validity of IBAS-Elman model,the beetle antennae search algorithm,firefly algorithm and genetic algorithm were used as comparison algorithms.The experimental results show that the accuracy of the improved beetle antennae search algorithm is better than that of other comparison algorithms.

关 键 词：滚动轴承 故障诊断 ELMAN神经网络 改进天牛须搜索算法 

分 类 号：TH133.3[机械工程—机械制造及自动化]