基于EMD模糊熵与会诊决策融合模型的中介轴承故障诊断技术  

Fault Diagnosis Technology of Inter-shaft Bearing Based on EMD Fuzzy Entropy and Consultative Decision Fusion Model

作　　者：王志[1] 李有儒 田晶[1] 刘丽丽 李吉凯 WANG Zhi;LI You-ru;TIAN Jing;LIU Li-li;LI Ji-kai(Liaoning Key Laboratory of Advanced Test Technology for Aeronautical Propulsion System,Shenyang Aerospace University,Shenyang 110136,China;Institute of Information Science,Beijing Jiaotong University,Beijing 100044,China)

机构地区：[1]沈阳航空航天大学辽宁省航空推进系统先进测试技术重点实验室,沈阳110136 [2]北京交通大学信息科学研究所,北京100044

出　　处：《航空发动机》2019年第5期76-81,共6页Aeroengine

基　　金：国家自然科学基金(11702177);辽宁省自然科学基金(20180550783)资助

摘　　要：针对航空发动机中介轴承故障信号难于识别的问题,提出了1种融合kNN、SVM、CART、RF及GBDT5种算法的会诊决策融合模型。基于中介轴承故障模拟试验台开展某型航空发动机中介轴承故障试验验证。采用EMD算法对采集的振动故障信号进行分解,提取IMF分量的模糊熵作为故障特征。利用建立的会诊模型对中介轴承内环故障、内环-滚动体耦合故障、正常、滚棒剥落、滚棒划伤5种不同状态进行识别。试验研究表明:会诊模型的故障诊断准确率高达92.5%,并表现出良好的泛化能力。In order to solve the problem that it was difficult to identify the fault signal of aeroengine inter-shaft bearing,a consultative decision fusion model combining five algorithms of kNN,SVM,CART,RF and GBDT was proposed.The failure test verification of a type of aeroengine inter-shaft bearing was carried out based on the inter-shaft bearing fault simulation test bench.The EMD algorithm was used to decompose the collected vibration fault signal,and the fuzzy entropy of IMF component was extracted as the fault feature.Fault of the inner ring,fault of inner ring and rolling body coupling,normal,fault of stick-spalled,fault of stick-scratched were identified by established consultative model.The test results show that the fault diagnosis accuracy of the consultative model is as high as 92.5%,and it shows good generalization ability.

关 键 词：经验模态分解 中介轴承 模糊熵 决策融合 会诊模型 故障诊断 航空发动机 

分 类 号：V231.92[航空宇航科学与技术—航空宇航推进理论与工程]