基于DSmT的航空发动机早期振动故障融合诊断方法  

Diagnosis of aero-engine with early vibration fault symptom using DSmT

作　　者：翟旭升[1] 胡金海[1] 谢寿生[1] 刘佳[2] 李强[2] 

机构地区：[1]空军工程大学工程学院,西安710038 [2]空军工程大学理学院,西安710038

出　　处：《航空动力学报》2012年第2期301-306,共6页Journal of Aerospace Power

摘　　要：提出在航空发动机多个部位安装多个振动传感器组成传感器网络.采用多传感器信息融合技术进行早期振动故障的诊断方法,并引入Dezert-Smarandache理论(DSmT)来处理由早期微弱故障本身所导致的各个传感器信息相互冲突的问题.在构建的早期微弱故障诊断系统框架中,采用基于本征模态函数(IMF)的信息熵特征提取方法提取各路振动数据的特征,采用反向传播(BP)神经网络完成对故障属性的判断并生成各种故障模式的基本置信分配,最后根据DSmT融合规则得到最终的诊断结果.算例表明采用该方法可以有效地解决早期微弱故障条件下的高冲突信息融合问题,故障诊断结果准确可靠.Multisensor network information fusion method was used to diagnose the aeroengine with early vibration fault symtom, and several vibration sensors had been fixed on different positions of the aeroengine to build the muhisensor network. However the information collected from different sensors would highly conflicts when early vibration fault happens, and the information fusion result get by Dempster-Shafer rule would be unreasonable, so Dezert-Smarandache theory （DSmT） was applied to solve the problem. In the early vibration fault diagnosis system proposed in this paper, firstly, a method based on the intrinsic mode functions （IMF） energy entropy was used to extract the signal＇s feature; secondly, basic belief assignment function was constructed based on the output of the back propagation （BP） neural network; lastly, DSmT combination rule was used to combine the different evidences and make the final decision. Two examples suggest the approach is available to solve the problem of high-conflict information fusion when early vibration fault happens, and the diagnosis results are reliable and effective.

关 键 词：早期故障 航空发动机 故障诊断 信息融合 DEZERT-SMARANDACHE理论 

分 类 号：V263.6[航空宇航科学与技术—航空宇航制造工程]