摘要
提出了将神经网络与D-S证据理论相结合的故障诊断方法,实现了故障信号的特征级和决策级融合,并应用于轴承的复合故障诊断研究。将BP、RBF、GRNN 3种神经网络的输出结果作为3个证据体,滚动轴承的4种复合故障特征作为系统的识别框架,引入聚类系数作为权值分配,重新计算基本概率赋值,对D-S证据理论进行改进,以提高轴承复合故障诊断的准确性。
A fault diagnosis method based on the combination of neural network and Dempster/Shafer( D-S) evidence theory is proposed. Feature level and decision level fusion of fault signal was realized,which was applied in research of the composite fault diagnosis of bearing. The output results of back propagation( BP),radial-based function( RBF),general regression neural network( GRNN) three kinds of neural networks were used as three body of evidence. Four kinds of compound fault characteristics of rolling bearing were regarded as system identification framework. Clustering coefficient was introduced as the weight distribution,and the basic probability assignment was recalculated. The D-S evidence theory is improved to improve the accuracy of the composite fault diagnosis of bearing.
引文
[1]崔玲丽,吴春光,邬娜.基于EMD与ICA的滚动轴承复合故障诊断[J].北京工业大学学报,2014(10):1459-1464.CUI L L,WU C G,WU N.Composite Fault Diagnosis of Rolling Bearings Based on EMD and ICA Algorithm[J].Journal of Beijing University of Technology,2014(10):1459-1464.
[2]邓丽君,董增寿,宋明远.基于神经网络和证据理论的液压系统故障诊断[J].太原科技大学学报,2016(6):167-171.DENG L J,DONG Z S,SONG M Y.Hydraulic System Fault Diagnosis Based on Neural Network and Evidence Theory[J].Journal of Taiyuan University of Science and Technology,2016(6):167-171.
[3]李伟,梁玉英,朱赛.基于神经网络和证据理论的信息融合在故障诊断中的应用[J].计算机测量与控制,2012,20(11):2888-2893.LI W,LIANG Y Y,ZHU S.Fault Diagnosis Based On Neural Network And Evidence Theory Information Fusion[J].Computer Measurement&Control,2012,20(11):2888-2893.
[4]张锋利,陈文献,贾海英.支持向量机和BP神经网络在水轮发电机轴承故障诊断中的应用[J].电网与清洁能源,2013(4):62-66.ZHANG F L,CHEN W X,JIA H Y.Application of Support Vector Machines and BP Neural Network in the Rolling Bearing of Hydraulic Turbine Generator Fault Diagnosis[J].Power System and Clean Energy,2013(4):62-66.
[5]孙旺旺,任传胜,朱春伟.基于模糊RBF神经网络的滚动轴承故障诊断[J].机械研究与应用,2013(2):13-17.SUN W W,REN C S,ZHU C W.Fault Diagnosis of Rolling Bearing Based on Fuzzy RBF Artificial Neural Network[J].Mechanical Research&Application,2013(2):13-17.
[6]徐富强,郑婷婷,方葆青,基于广义回归神经网络(GRNN)的函数逼近[J].巢湖学院学报,2010(6):11-16.XU F Q,ZHENG T T,FANG B Q.Function Approcimation Based on General Regression Neural Network(GRNN)[J].Journal of Chaohu College,2010(6):11-16.
[7]岑健,胥布工,张清华,等.基于证据理论的免疫检测器在轴承故障诊断中的应用[J].轴承,2009(8):42-46.CEN J,XU B G,ZHANG Q H,et al.Application of Immune Detector in Bearing Fault Diagnosis Based on Dempstershafer Evidential Theory[J].Bearing,2009(8):42-46.
[8]李岸巍,阮豫红.基于MATLAB环境的聚类系数的计算[J].山西师范大学学报(自然科学版).2009(9):32-35.LI A W,RUAN Y H.The Calculation of the Clustering Coefficient Based on MATLAB[J].Journal of Shanxi Normal University(Natural Science Edition),2009(9):32-35.
[9]屈胜男.模糊神经网络和D-S证据理论在齿轮箱故障诊断中的应用[D].哈尔滨:哈尔滨工业大学,2008.