基于D-S信息融合技术的磁轴承转子故障诊断方法  

Fault Diagnosis Method of Magnetic Rotor based on D-S Information Fusion Technology

作　　者：袁倩[1] 孙冬梅[1] 范文[1] 

机构地区：[1]南京工业大学电气工程与控制科学学院,江苏南京211816

出　　处：《仪表技术与传感器》2016年第11期118-122,共5页Instrument Technique and Sensor

基　　金：国家自然科学基金项目(51277092);江苏省人事厅江苏省博士后资助计划项目(1201012C)

摘　　要：为了提高磁轴承转子故障诊断结果的可靠性,提出了基于D-S信息融合的故障诊断方案:首先选择轴心轨迹作为初步判断,然后以EEMD为基础,分别对多组振动信号的边际谱进行特征频段能量的计算,采用BP神经网络进行故障识别,其结果再经过D-S证据理论做决策融合,最终确定磁轴承转子的故障类型。实验结果表明该方案提高了故障诊断结果的准确性,充分显示了其应用在磁轴承转子故障诊断系统中的可行性。In order to improve the reliability of the magnetic rotor fault diagnosis, fault diagnosis system solutions were proposed based on D-S information fusion technology. First of all, axis orbits were chosen as preliminary judgment of fault diagnosis. Secondly, marginal spectrums of multiple group vibration signals were calculated to the energy of characteristic band respectively based on EEMD.BP neural network was used as the recognition of rotor fault and the D-S theory was used to obtain the final diagnosis result for the multiple diagnostic results of BP neural network. The experimental results show that it effectively improves the veracity of fault diagnosis and fully shows the effectiveness of the system on the fault diagnosis of magnetic rotor.

关 键 词：故障诊断 信息融合 D-S证据理论 磁轴承转子 EEMD 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置]