一种旋转机械系统耦合故障诊断的新方法  

New approach for diagnosis on coupled faults of rotating machine system

作　　者：焦卫东[1] 蒋永华[1] 施继忠[1] 王晓燕[1] 

机构地区：[1]浙江师范大学工学院,金华321004

出　　处：《仪器仪表学报》2016年第7期1449-1456,共8页Chinese Journal of Scientific Instrument

基　　金：国家自然科学基金(51575497;51405449);浙江省公益应用技术(2016C31067)项目资助

摘　　要：旋转机械系统中,与转子振动响应相关的裂纹非线性开/合行为以及动-静件瞬态碰摩接触或齿轮啮合冲击、机座松动等缺陷或故障均会激起非线性振动,且不同的故障激励源相互作用,导致复杂的故障耦合形态,增加了故障诊断的难度。为此提出一种基于本征模函数能量矩-灰色关联分析的旋转机械系统耦合故障诊断方法。应用能量矩测度,从振动响应信号的经验模态分解本征模函数中分别抽取特征,然后进一步利用灰色关联分析对耦合故障的关联特性进行定量评价,从而实现复杂耦合故障模式的识别。仿真与实验研究结果验证了方法的有效性。而且,利用本方法可以对故障的严重程度进行排序,可以有力地支持合理的维修决策的制定。Nonlinear vibration can be excited by some drawbacks or faults such as rotor vibration response-dependent nonlinear crack breathing behavior,and transient rotor-stator rubbing,gear meshing impact or loose foundation,etc.,in a rotating machine system.Mutually interaction among different fault excitation sources lead to complex coupled fault morphology. All these make fault diagnosis a difficult task. To solve this problem,a diagnostic approach on coupled faults of rotating machine system was proposed. It was based on energy moment of intrinsic mode functions（ EM-IMFs） and gray relation analysis（ GRA）. Firstly,by using the EM measure,feature vectors were extracted from the IMFs by empirical mode decomposition（ EMD） on vibration responses respectively. Secondly,relevancy characteristics were quantitatively evaluated by using the GRA technique. At last,complex faulty patterns were recognized. Simulated and experimental results show that the proposed approach is effective,and the approach can sort the recognized faults according to their severity,which can be a strong support to make a reasonable maintenance decision.

关 键 词：经验模态分解 本征模函数能量矩 灰色关联分析 非线性振动 故障耦合 

分 类 号：TN911.4[电子电信—通信与信息系统] TN911.7[电子电信—信息与通信工程]